1.. .3 a . sag?
II '4‘:- J . . O-
h , ,Ffi “mm an? 3%
......u..afi.._.,.
:3. awn
.. l
; . ‘30:}
“M4311”.
.
4.5%.? fidflfi
.33....

u! .2!

;
sch-Gil;
32.21.11. ”A“... of
t»: i... z?! I

. ‘ £51.! in .... .
$3.1...3... “gm”

195.1.

 

THESiS

k‘xl

2,600

 

_ llillllliliillilllllllilllllilllllill

3 1293 02048 8734
LIBRARY

Michigan State
University

 

 

 

This is to certify that the

dissertation entitled

Mechanisms Underlying Prolactin-lnduced Activation
of Tuberoinfundibular Dopaminergic Neurons

presented by

Kenneth Hentschel

has been accepted towards fulfillment
of the requirements for

Ph.D. degreein Pharmacology and

Toxicology

1/ / 1:4 ajor professy

Date 8/21/00

MSU i: an Affirmative Action/Equal Opportunity Institution 0- 12771

 

4.—

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

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

11/00 C'JCIRC/DaiaDuo.p65-p.14

MECHANISMS UNDERLYING PROLACTIN-INDUCED
ACTIVATION OF TUBEROINFUNDIBULAR
DOPAMINERGIC NEURONS

By

Kenneth Hentschel

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Pharmacology and Toxicology

2000

tubero:

 

 

early It
Studied
dissena
llDA

COmbine
Pharmat
SPCCific
951011316

lH‘Jnedi;

 

a period greater than 3 h but less than 6 h of hyperprolactinemia is required for its
initiation, and that hyperprolactinemia is required to sustain the response.

Secondly, the temporal effects of prolactin on immediate early gene expression in
subpopulations of TIDA neurons was characterized using neural activity mapping with
inducible transcription factors combined with dual immunohistochemistry. Prolactin
differentially regulated the expression of F os-related antigens (FRA), a cellular marker of
genomic activity, in TIDA neurons in different arcuate nucleus subdivisions. Following
the induction of hyperprolactinemia transient increases in FRA expression were first
observed in TIDA neurons of the ventrolateral arcuate nucleus (1.5 h) followed by those
of the dorsomedial arcuate nucleus (3 h). Additionally, prolactin affected FRA expression
in a group of non-catecholaminergic neurons in the arcuate nucleus. These studies
demonstrated that prolactin elicits genomic effects prior to the delayed activation of
TIDA neurons, identified a novel group of prolactin-responsive non-catecholaminergic
neurons and suggested a temporal sequence of activation.

Finally, the role of neurotensin as a mediator of the effects of prolactin on TIDA
neurons was characterized. The first approach utilized neural activity mapping with
inducible transcription factors and dual immunohistochemistry for the detection of FRA
and neurotensin. By 1.5 hours after injection, prolactin increased the percentage of
neurotensin neurons expressing F RA and the increased genomic activity of these neurons
was maintained for 12 h. In addition, prolactin increased total numbers of neurotensin
neurons in the arcuate nucleus after 6-12 h. Taken together, prolactin modulated gene
expression in neurotensin and other neurons, stimulated neurotensin synthesis and

recruited these neurons into an activated pool.

0

SR—li'

TIDA .

TIDA

inept..-

DEMON

no: toni

 

 

 

cimcte
prolacti
maimcn
underly
Prelacti

Unileug

DELHOIE

deiai‘ec

Another approach utilized neurochemistry and the neurotensin receptor antagonist
SR-48692 to demonstrate a role for neurotensin receptors in the inductive activation of
TIDA neurons by prolactin. The neurotensin receptor antagonist had no effect on basal
TIDA neuronal activity or plasma prolactin concentrations. However, the neurotensin
receptor antagonist blocked the delayed stimulatory effects of prolactin on TIDA
neurons. The results demonstrated that neurotensin receptors mediated the delayed, but
not tonic stimulatory effects of prolactin on TIDA neurons.

In conclusion, the inductive activation of TIDA neurons by prolactin was
characterized by a temporal delay of 6-12 h, was dependent upon a period of elevated
prolactin for greater than 3 h for initiation and required hyperprolactinemia for
maintenance of the response. The experiments also lent insight into the mechanisms
underlying prolactin-induced activation of TIDA neurons. Such mechanisms involved
prolactimdependent genomic changes in TIDA and neurotensin neurons in the arcuate
nucleus hours prior to the time of TIDA neuronal activation. The role of neurotensin in
the delayed inductive activation of TIDA neurons by prolactin was also characterized. It
was observed that prolactin caused early and prolonged increases in the genomic activity
of existing neurotensin neurons and of those neurons recruited by prolactin to synthesize
neurotensin. Moreover, results demonstrated that neurotensin receptors mediated the

delayed stimulatory effects of prolactin on TIDA neurons.

To my loving wife, Alicia,
my beautiful daughters, Brynn and Kaylin,

and my wonderful families.

   

L'oolir;~

for she:

 

formed l

“an

Chum;

God-MS:

hBUbba“

dilingi

and Dim

ACKNOWLEDGMENTS

Foremost, I would like to thank my thesis advisor and mentor, Dr. Keith
Lookingland, for fostering my development as a physician-scientist. I am grateful to him
for sharing his diverse knowledge, creativity, expertise in experimental design, technical
skill and proficiency in writing and public speaking. In exercising these strengths, he
outlined an interesting and productive thesis project, and guided me through its
completion.

I am grateful for an insightful, cooperative and supportive guidance committee
formed by the esteemed Drs. Kenneth Moore, John Thomburg, and Cheryl Sisk whose
expert advice kept my thesis project directed and circumscribed.

I appreciate the commentary, efforts and friendship of my collaborators, Drs. Sun
Cheung, Chris McMahon, Annette Fleckenstein and Yvonne Will-Murphy. “Oh my
Godness!” -thanks for the memories!

I would like to thank the technicians, Sarah “Bart” Cooper, Mark Jacob, Thomas
“Bubba” Ingram and Erika Bronz for their contribution to my thesis work through their
diligence and high standards. Best wishes for your bright futures.

I would like to recognize the front office staff, Nelda Carpenter, Mickie Vanderlip
and Diane Hummel for numerous acts of kindness.

Finally, I must thank my family: my wife, Alicia, for her love, support, patience
and for helping me maintain balance; my parents, Christopher and Carol Arm, for their
love and for facilitating this endeavor with kind gifis of computer hardware and software;
and my in-laws, Stan and Claudia Kavan, for their love and morale.

Thanks again everyone, I could not have done it without you!

vi

 

llSl ()

LIST ()

LIST (I;

l (
l‘l
(
l):
l:
I.
G.
H.
l.

2' M;
A,

TABLE OF CONTENTS

LIST OF TABLES ................................................................................ x
LIST OF FIGURES ............................................................................... xi
LIST OF ABBREVIATIONS ................................................................... xvii
1. GENERAL INTRODUCTION ...................................................... 1
A. STATEMENT OF PURPOSE ................................................ 1
B ANATOMY AND FUNCTION OF CENTRAL DA NEURONS ....... 2
C. BIOCHEMISTRY OF CENTRAL DA NEURONS ....................... 4
D NEUROCHEMICAL INDICES OF DA NEURONAL ACTIVITY. 7
E. NEURAL ACTIVITY MAPPING WITH INDUCIBLE
TRANSCRIPTION FACTORS ............................................... 9
F. SUBPOPULATIONS OF TIDA NEURONS ............................... 14
G. REGULATION OF CENTRAL DA NEURONS ........................... 17
1. DA receptor-mediated mechanisms ................................. l7
2. Prolactin feedback ...................................................... 19
3. Afferent regulation of TIDA neurons ............................... 22
H. NEUROTENSIN ................................................................ 24
1. Synthesis ................................................................. 24
2. Central distribution ..................................................... 25
3. Receptors ................................................................ 26
4. Central actions .......................................................... 27
I. THESIS OBJECTIVE .......................................................... 28
2. MATERIALS AND METHODS ..................................................... 29
A. ANIMALS ........................................................................ 29
B. DRUGS ............................................................................ 29

vii

C. SURGICAL MAN IPULATIONS ............................................. 31

1. Intracerebroventricular cannulation and injections ................. 31
2. Intravenous catheterization and injections ........................... 32
D. NEUROCHEMISTRY .......................................................... 33
1. Tissue preparation ...................................................... 33
2. Assays .................................................................... 35
E. PROLACTIN ASSAYS ........................................................ 39
1. Radioimmunoassay for prolactin ..................................... 39
2. Bioassay for prolactin .................................................. 45
F. IMMUNOHISTOCHEMISTRY .............................................. 47
1. Tissue preparation ...................................................... 47
2. FRA and tyrosine hydroxylase detection ............................ 48
3. FRA and neurotensin detection ....................................... 52
4. Neurotensin and tyrosine hydroxylase detection ................... 55
5. Slide preparation and visualization ................................... 57
G. QUANTIFICATION AND STATISTICAL ANALYSES ................. 59
REGULATION OF TIDA NEURONS BY PROLACTIN ..................... 61
A. INTRODUCTION ............................................................... 61
B. RESULTS ........................................................................ 64
C. DISCUSSION .................................................................... 75
EFFECTS OF PROLACTIN ON EXPRESSION OF FRA IN TH-IR
NEURONS IN SUBDIVISIONS OF THE ARCUATE NUCLEUS ........... 80
A. INTRODUCTION ............................................................... 80
B. RESULTS ........................................................................ 83
C. DISCUSSION .................................................................... 106
EFFECTS OF PROLACTIN ON FRA EXPRESSION IN
NON-TH-IR CELLS IN THE ARCUATE NUCLEUS .......................... 113
A. INTRODUCTION ............................................................... 1 13
B. RESULTS ........................................................................ 115
C. DISCUSSION .................................................................... 129

viii

6. ROLE OF NEUROTENSIN IN PROLACTIN-INDUCED

ACTIVATION OF TIDA NEURONS .............................................. 135
A. INTRODUCTION ............................................................... 135
B. RESULTS ........................................................................ 138
C. DISCUSSION .................................................................... 147
7. EFFECTS OF PROLACTIN ON EXPRESSION OF F RA IN
NEUROTENSIN-IR NEURONS IN THE ARCUATE NUCLEUS ........... 151
A. INTRODUCTION ............................................................... 151
B. RESULTS ........................................................................ 154
C. DISCUSSION .................................................................... 170
8. SUMMARY AND CONCLUSIONS ................................................ 175
A. SUMMARY ...................................................................... 175
B. CONCLUDING DISCUSSION ................................................ 179
BIBLIOGRAPHY ................................................................................. 1 85

LIST OF TABLES

Table # Table Title Page

6.1 Effects of SR-48692 on DOPAC and dopamine concentrations in 139
median eminence, striatum and nucleus accumbens of male and
female rats.

Figurt

 

23

2.4

2.6

2.7

2.8

 

Figure #

1.1

1.2

1.3

1.4

1.5

2.1

2.2

2.3

2.4

2.5

2.6

2.7

2.8

3.1

LIST OF FIGURES

Figure Title

Schematic saggital view of the rat brain demonstrating the distribu-
tion of catecholamine-containing perikarya.

Schematic representation of differences in neurochemical and
metabolic pathways in TIDA and NSDA or MLDA nerve terminals.

Schematic representation of extracellular signal-mediated long-term
changes in gene expression in a neuron.

Drawing of a frontal section through the middle ARC depicting
characteristic TH-IR in the ARC subdivisions.

Schematic representation of the feedback regulation of TIDA
neuronal activity and prolactin secretion.

Anatomical locations of nucleus accumbens and striatum samples
collected for neurochemical analysis.

Anatomical locations of median eminence samples collected for
neurochemical analysis.

Effects of duration of chloramine-T oxidative reaction on iodinated
prolactin curve morphology.

Characterization of PD-lO Sephadex G-25M column for fractionation
of rat prolactin for iodination via Lowry protein assay.

Representative rostrocaudal regions of the ARC highlighted in
ventral hypothalamic frontal sections.

Digitized photomicrographs of tissue immunohistochemically
processed for the detection of TH and FRA.

Digitized bright field photomicrograph of tissue immunohisto-
chemically processed for the detection of NT and F RA.

Digitized photomicrographs of tissue immunohistochemically
processed for the detection of TH and NT.

Time course effects of haloperidol on concentrations of prolactin in
plasma, and DOPAC and DA in the median eminence of male rats.

xi

Page

10

15

20

34

36

42

44

49

53

56

58

65

3.4

I”
'1.

4.1

4.2

4.3

4.4

4.6

 

3.2

3.3

3.4

3.5

3.6

3.7

3.8

4.1

4.2

4.3

4.4

4.5

4.6

Time course effects of haloperidol on concentrations of DOPAC and
DA in the striatum of male rats.

Time course effects of haloperidol on concentrations of DOPAC and
DA in the nucleus accumbens of male rats.

Time course effects of PRL-AB on concentrations of plasma
prolactin and DOPAC and DA in the median eminence of
haloperidol-pretreated male rats.

Effects of PRL-AB on concentrations of DOPAC and DA in the
median eminence of haloperidol-treated male rats.

Effects of PRL-AB on concentrations of DOPAC and DA in the
striatum and nucleus accumbens of haloperidol-treated male rats.

Treatment paradigm for comparison of the effects of haloperidol on
median eminence DOPAC concentrations in PRL-AB and PRL-AB
plus prolactin-treated male rats.

Comparison of the effects of haloperidol on median eminence
DOPAC concentrations in PRL-AB and PRL-AB plus prolactin-
treated male rats.

Comparison of the numbers of TH-IR neurons in the DM and VL
subdivisions of rostral, middle and caudal ARC of male rats.

Comparison of %TH-IR neurons expressing FRA in the DM and VL
subdivisions of rostral, middle and caudal ARC of male rats.

Time course effects of haloperidol on plasma prolactin concentra-
tions in male rats.

Time course effects of haloperidol on %TH-IR neurons expressing
FRA in the DM subdivision of rostral, middle and caudal ARC of
male rats.

Time course effects of haloperidol on %TH-IR neurons expressing
FRA in the VL subdivision of rostral, middle and caudal ARC of
male rats.

Time course comparison of the numbers of TH-IR neurons in the DM
and VL subdivisions of rostral, middle and caudal ARC of vehicle-
and haloperidol-treated male rats.

xii

66

67

68

70

71

72

73

84

85

86

87

89

90

4.8

4.9

4.10

 

 

4.13

4.14

4.16

4.17

4.13

4.7

4.8

4.9

4.10

4.11

4.12

4.13

4.14

4.15

4.16

4.17

4.18

Effects of prolactin antiserum on %TH-IR neurons expressing FRA
in the DM subdivision of rostral, middle and caudal ARC of
haloperidol-treated male rats.

Effects of prolactin antiserum on %TH-IR neurons expressing FRA
in the VL subdivision of rostral, middle and caudal ARC of
haloperidol-treated male rats.

Effects of duration of hyperprolactinemia on %TH-IR neurons
expressing FRA in the DM-ARC of haloperidol-treated male rats.

Time course effects of prolactin on %TH-IR neurons expressing F RA
in the DM subdivision of rostral, middle and caudal ARC of male
rats.

Time course effects of prolactin on %TH-IR neurons expressing F RA
in the VL subdivision of rostral, middle and caudal ARC of male rats.

Time course comparison of the numbers of TH-IR neurons in the DM
and VL subdivisions of rostral, middle and caudal ARC of vehicle-
and prolactin-treated male rats.

Comparison of digitized photomicrographs of TH-IR neurons
expressing FRA in the rostral DM-ARC of male rats injected with
vehicle, haloperidol or prolactin.

Comparison of the numbers of TH-IR neurons in the DM and VL
subdivisions of rostral, middle and caudal ARC of diestrous female
rats.

Comparison of numbers of TH-IR neurons in the DM and VL
subdivisions of rostral, middle and caudal ARC of male and diestrous
female rats.

Comparison of %TH-IR neurons expressing FRA in the DM and VL
subdivisions of rostral, middle and caudal ARC of female rats.

Comparison of %TH-IR neurons expressing FRA in the DM and VL
subdivisions of rostral, middle and caudal ARC of male and diestrous
female rats.

Effects of prolactin on %TH-IR neurons expressing FRA in the DM

and VL subdivisions of rostral, middle and caudal ARC of diestrous
female rats.

xiii

91

92

93

95

96

97

98

100

101

102

103

105

 

 

5.7

5.9

 

6.]

5.1

5.2

5.3

5.4

5.5

5.6

5.7

5.8

5.9

5.10

5.11

6.1

Comparison of basal numbers of FRA-IR nuclei in non-TH-IR cells
in subdivisions of the rostral, middle and caudal ARC in male rats.

Digitized photomicrographs showing increased FRA expression in
the middle ARC afier haloperidol injection in male rats.

Time course effects of haloperidol on numbers of FRA-IR nuclei in
non-TH-IR cells in the DM-MC subdivision of rostral, middle and
caudal ARC of male rats.

Time course effects of haloperidol on numbers of FRA-IR nuclei in
non-TH-IR cells in the DM-PC subdivision of rostral, middle and
caudal ARC of male rats.

Time course effects of haloperidol on numbers of FRA-IR nuclei in
non-TH-IR cells in the VL subdivision of rostral, middle and caudal
ARC of male rats.

Effects of prolactin antiserum on numbers of FRA-IR nuclei in non-
TH-IR cells in the DM-MC subdivision of rostral, middle and caudal
ARC of haloperidol-treated male rats.

Effects of prolactin antiserum on numbers of FRA-IR nuclei in non-
TH-IR cells in the DM-PC subdivision of rostral, middle and caudal
ARC of haloperidol-treated male rats.

Effects of prolactin antiserum on numbers of FRA-IR nuclei in non-
TH-IR cells in the VL subdivision of rostral, middle and caudal ARC
of haloperidol-treated male rats.

Time course effects of prolactin on numbers of FRA-IR nuclei in
non-TH-IR cells in the DM-MC subdivision of rostral, middle and
caudal ARC of male rats.

Time course effects of prolactin on numbers of FRA-IR nuclei in
non-TH-IR cells in the DM-PC subdivision of rostral, middle and
caudal ARC of male rats.

Time course effects of prolactin on numbers of FRA-IR nuclei in
non-TH-IR cells in the VL subdivision of rostral, middle and caudal
ARC of male rats.

Effects of SR-48692 on DOPAC and dopamine concentrations in the
median eminence of haloperidol-treated male and female rats.

xiv

116

117

119

120

121

122

123

124

136

127

128

140

6.4

7.1

 

7.2

*4
bu

7.6

7.7

7,3

6.2

6.3

6.4

6.5

7.1

7.2

7.3

7.4

7.5

7.6

7.7

7.8

7.9

Effects of SR-48692 on DOPAC and dopamine concentrations in the
striatum and nucleus accumbens of haloperidol-treated male rats.

Effects of SR-48692 on serum prolactin concentrations in
haloperidol-treated male and female rats.

Effects of SR-48692 on DOPAC and dopamine concentrations in the
median eminence of prolactin-treated male and female rats.

Effects of SR-48692 on DOPAC and dopamine concentrations in the
striatum and nucleus accumbens of prolactin-treated male rats.

Comparison of numbers of NT-IR neurons in the DM and VL
subdivisions of rostral, middle and caudal ARC of male rats.

Comparison of %NT-IR neurons expressing FRA-IR nuclei in the
DM and VL subdivisions of rostral, middle and caudal ARC in male
rats.

Digitized photomicrographs showing increased FRA expression in
NT-IR neurons in the middle ARC after haloperidol injection in male
rats.

Time course effects of haloperidol on %NT-IR neurons expressing
FRA in the DM subdivision of rostral, middle and caudal ARC in
male rats.

Time course effects of haloperidol on %NT-IR neurons expressing
FRA in the VL subdivision of rostral, middle and caudal ARC in
male rats.

Digitized photomicrographs showing increased numbers of NT-IR
neurons in the middle ARC after haloperidol injection in male rats.

Time course effects of haloperidol on numbers of NT-IR neurons in
the DM and VL subdivisions of rostral, middle and caudal ARC in
male rats.

Time course effects of prolactin on %NT-IR neurons expressing FRA
in the DM subdivision of rostral, middle and caudal ARC in male
rats.

Time course effects of prolactin on %NT-IR neurons expressing FRA
in the VL subdivision of rostral, middle and caudal ARC in male rats.

XV

142

143

144

146

155

156

157

159

160

161

162

163

164

 

7.10

7.11

7.12

 

7.13

8.1

7.10

7.11

7.12

7.13

8.1

8.2

Time course effects of prolactin on numbers of NT-IR neurons in the
DM and VL subdivisions of rostral, middle and caudal ARC in male
rats.

Digitized photomicrographs depicting the location of NT-IR neurons
relative to TH-IR neurons in the VL-ARC in male rats.

Time course effects of prolactin on numbers of dual NT- and TH-IR
neurons, and numbers of NT-IR neurons in the DM subdivision of the
middle ARC in male rats.

Time course effects of prolactin on numbers of dual NT- and TH-IR
neurons, and numbers of NT-IR neurons in the VL subdivision of the
middle ARC in male rats.

Schematic summary of the temporal effects of prolactin on IEG
expression in cells, and the recruitment of neurotensin neurons in

ARC subdivisions during the inductive activation of TIDA neurons.

Schematic representation of the theoretical distal circuitry of the
inductive activation of TIDA neurons by prolactin.

xvi

165

167

168

169

177

182

AAA:

ABC

AP- 1

 

ARC
BSA
Ca-Ca'
CAMP
CaRE
CC

CP.\1

 

C RE
CRIB
DA
DAB
DM
DUN
DMSQ
DQPA
DOPAC
EDTA

PM

3V

ABC

AP- 1

ARC

BSA

Ca-CamK

CAMP

CaRE

CC

CPM

CRE

CREB

DA

DAB

DM

DMN

DMSO

DOPA

DOPAC

EDTA

FRA

LIST OF ABBREVIATIONS
third ventricle
aromatic L-amino acid decarboxylase
avidin biotin complex
activator protein-1
arcuate nucleus
bovine serum albumin
calcium calmodulin-dependent kinase
adenosine 3,5-monophosphate
calcium response element
corpus callosum
counts per minute
CAMP response element
CAMP response element binding protein
dopamine
3,3-diaminobenzidine tetrahydrochloride
dorsomedial
dorsomedial nucleus
dimethyl sulfoxide
3,4—dihydroxyphenylalanine
3,4-dihydroxyphenylacetic acid
disodium ethylenediamine tetraacetic acid

fos-related antigens

xvii

GAB

1111’?

in

 

LH
LRG
MAC)
.\1E
min

MLD

GABA

HIPP
icv
IEG

im

ip

IR

iv

LH
LRG
MAO
ME
min
MLDA
NA
ND
NHPP
NIDDK

NMDA

NSDA

NT

gammabutyric acid
hours

hippocampus
intracerebroventricular
immediate early gene
intramuscular
intraperitoneal
immunoreactivity
intraventricular

lateral hypothalamus
late response gene
monoamine oxidase
median eminence
minutes

mesolimbic dopamine
nucleus accumbens
non-detectable
National Hormone and Pituitary Program
National Institute of Diabetes and Digestive and Kidney Diseases
N-methyl D-aspartate
normal rabbit serum
nigrostriatal dopamine

neurotensin

xviii

 

NTR

opt

PBS

PC

PICA

 

PRC
PR1

PRL-A

RIA

SDI
SRE
SRF
SI

TBS

TH

TYR
It

 

NTR
opt
PBS
PC
PKA
PKC
PRL

PRL-AB

RIA

sc
SEM
SRE
SRF
ST
TBS
TH

TI

TYR
VL

VMN

neurotensin receptor
optic tracts

phosphate buffered saline
parvocellular

protein kinase A
protein kinase C
prolactin

prolactin antiserum
every
radioimmunoassay
second

subcutaneous

standard error of the mean
serum response element
serum response factor
striatum

tris-buffered saline
tyrosine hydroxylase
tuberoinfundibular
Triton-X100

tyrosine

ventrolateral

ventromedial nucleus

xix

actix .;
llClITII
neuro.
actix i'
. . i
mIorrr
iranscI

03111011

ICSp0n_

 

PTOIactj
PIOIaCII
Pharma

Charac II

1. GENERAL INTRODUCTION

A. STATEMENT OF PURPOSE

Much of what is known about the mechanisms underlying prolactin-induced
activation of tuberoinfundibular (TI) dopamine (DA) neurons is based upon
neurochemical estimates of neuronal activity. Since these methods rely on changes in
neurochemistry in termini of TIDA neurons, this index provides a composite estimate of
activity of all neurons projecting to this region. While neurochemistry has been useful and
informative, the advent of new techniques (e.g. neural activity mapping using inducible
transcription factors coupled with dual immunohistochemistry) allows the estimation of
neuronal activity at the cellular level, and hence the investigation of differentially-
responsive subpopulations of TIDA neurons.

The overall aim of this proposal is to characterize mechanisms underlying
prolactin-induced activation of TIDA neurons and to identify mediators of the actions of
prolactin. To this end, a combination of anatomical, molecular, neurochemical and
pharmacological approaches are employed to address the following specific aims: (1) To
characterize the temporal effects of prolactin on neurochemical estimates of TIDA
neuronal activity, (2) to characterize the temporal effects of prolactin on immediate early
gene expression in anatomically and neurochemically distinct subpopulations of TIDA
neurons, and (3) to characterize the role of neurotensin as a mediator of the effects of

prolactin on TIDA neurons.

CI35>J
Figu

telent
dopa:
temp.
sensor

imolu

 

0601’an
Iimbic

mgd.
and [he

at. 197

“mm;

Pitui'qx

B. ANATOMY AND FUNCTION OF CENTRAL DOPAMINERGIC
NEURONS

The mammalian brain contains several different dopaminergic neuronal systems
classified alphanumerically by the locations of their perikarya and axon termini (See

Figure 1.1; Dahlstrom and Fuxe, 1964). The largest and most studied is the meso-

telencephalic system, comprised of nigrostriatal dopamine (NSDA) and mesolimbic
dopamine (MLDA) neurons. NSDA neurons have perikarya in the substantia nigra pars
compacta (As-A9) and axons that terminate in the striatum. These neurons regulate
sensorimotor integration and motor movement. Dysfunction of NSDA neurons is
involved in the pathogenesis of Parkinson’s disease (Chase and Oh, 2000). MLDA
neurons have perikarya in the ventral tegmental area (A10) and axons that terminate in
lirnbic structures including the nucleus accumbens, olfactory tubercles, septum,
arnygdala, and various cortical regions. These neurons regulate motivational behavior,
and their dysfunction has been implicated in the pathogenesis of schizophrenia (Snyder et
al., 1974; Matthyssee, 1980).

TIDA neurons have perikarya in the arcuate nucleus (ARC; A12) and axons that
terminate in the external layer of the median eminence. Dopamine released in the median
eminence from TIDA neurons travels in the hypophysial portal blood to the anterior
pituitary where it tonically inhibits prolactin secretion (Ben-Jonathan, 1985). The primary
focus of this dissertation is the regulation of TIDA neurons by prolactin. NSDA and
MLDA systems are incorporated to highlight differences in the regulation of TIDA

neurons compared to “classical” mesotelencephalic neurons.

Cortex

  
 
  
 

:=Ai'6:"'9‘

Cerebellum

A8

“-3?" :‘df‘ qlillllllfiufiilmit ‘

avimui A2

:33. 3...">'.‘IA7IIIII" Pons WM

A villi/Iii”. (“(111111

: Dopamine : —: Norapi. -—--

 

 

Figure 1.1 Schematic saggital view of the rat brain demonstrating the distribution of
catecholamine-containing perikarya (Moore, 1987a). Cell groups Al-A7 are noradrenergic
and groups AVA“, are dopaminergic (Dahlstrom and Fuxe, 1964). Abbreviations: CC,
corpus callosum; HIPP, hippocampus; Norepi, norepinephrine; ST, striatum.

C. BIOCHEMISTRY OF CENTRAL DOPAMINERGIC NEURONS

Current knowledge of the synthesis, storage, release and metabolism of dopamine
has been predominantly derived from studies of NSDA neurons. However, the majority
of neurochemical events occurring in MLDA and TIDA nerve termini are similar to those
described for NSDA neurons. As schematically depicted in figure 1.2, dopamine
synthesis begins when dietary tyrosine is actively transported into the nerve terminal
(Guroff et al., 1961) and converted by the rate limiting enzyme, tyrosine hydroxylase
(TH) to 3,4-dihydroxyphenylalanine (DOPA). DOPA is rapidly decarboxylated by
aromatic L—amino acid decarboxylase (AAAD) to form cytosolic dopamine (Nagatsu,
1973). The majority of newly synthesized cytosolic dopamine is packaged into synaptic
vesicles and stored for release while a fraction remains free in the cytosol where it may
inhibit the synthesis of additional dopamine by end product inhibition of TH, and be
metabolized by monoamine oxidase to 3,4—dihydroxyphenylacetic acid (DOPAC).

Stored dopamine in NSDA and MLDA nerves is released into synaptic clefts of
forebrain structures where it can act at pre-synaptic autoreceptors or post-synaptic
receptors. Activation of pre-synaptic autoreceptors inhibits further synthesis and release
of dopamine in these neurons (Christiansen and Squires, 1974; Roth et al., 1975).
Stimulated TIDA nerves release dopamine into the hypophysial portal vasculature in the
external layer of the median eminence, rather than into a classical synapse like
mesotelencephalic neurons. Dopamine released there acts at post-effector receptors in the
anterior pituitary. TIDA neurons also differ from mesotelencephalic neurons in that they
lack pre-synaptic autoreceptors (Gudelsky and Moore, 1976; Demarest and Moore, 1979;

Lookingland and Moore, 1984).

I -I

T)

NS
NE

 

TY

TYR

TIDA NERVE TERMINAL /‘ ME

Portal Blood
Turf—F
I
I
TYR \
DOPAC
NSDA/MLDA T

 

 

NERVE TERMINALS

Postsynaptic
( MAO ) Receptor

hp
TYR Iii—.001». flu» _t_.DA _.:'_
I \ Rf
TYR

Presynaptic
Autoreceptor

 

 

 

 

Figure 1.2 Schematic representation of differences in neurochemical and metabolic
pathways in TIDA (ABOVE) and NSDA or MLDA (BELOW) nerve terminals (modified
from Moore, 1987a). Abbreviations: AAAD, aromatic L-amino acid decarboxylase; DA,
dopamine; DOPA, 3,4-dihydroxyphenylalanine; DOPAC, 3,4—dihydroxyphenylacet1c
acid; MAO, monoamine oxidase; ME, median eminence; NA, nucleus accumbens; ST,
striatum; TH, tyrosine hydroxylase; TYR, tyrosine.

mka~
VCSICL;

mead

dOpam

howex-

“1111 SI;

IIIIEIy 1

 

 

 

rue C
to DO}
metaboi
may dif
“Haul
emfimet

m? H0

hOmOVa

The synaptic actions of dopamine are terminated in NSDA and MLDA nerves
predominantly via high affinity uptake pumps in the pre-synaptic nerve termini. Once in
the nerve terminal, recaptured dopamine can be repackaged into synaptic vesicles for re-
release or be metabolized by mitochondrial monoamine oxidase to DOPAC. Non-
vesicular cytosolic dopamine transported from the synapse may also feed back to inhibit
the activity of TH, and hence the synthesis of dopamine.

TIDA nerves differ from the mesotelencephalic system in the removal of
dopamine from the synapse. Presynaptic TIDA nerves have a dopamine uptake pump,
however, it is of low affinity and questionable relevance (Demarest and Moore, 1979;
Annunziato et al., 1980). Since TIDA nerves release dopamine in the median eminence
with significant concentration gradient imposed by the fenestrated portal vasculature, it is
likely that the low affinity uptake system in TIDA nerves does not have an important
role. Consequently, in TIDA neurons little to no dopamine is recaptured and metabolized
to DOPAC. Therefore, DOPAC concentrations in TIDA neurons are due to the
metabolism of newly synthesized dopamine (Lookingland et al., 1987a; 1987b). DOPAC
may diffuse out of NSDA, MLDA and TIDA nerves where it may be metabolized by
extraneuronal catechol-O-methyltransferase to form homovanillic acid. A minority of
extraneuronal dopamine that evades uptake may be directly methylated by catechol-O-
methyltransferase to form 3-methoxytyramine, and subsequently oxidized to

homovanillic acid.

rnuhi

Mabfi

1’)

the.
1996'
coupi
coupi
neuro:
neurot
manor

3131.?

mach
101061
1976; I
MAD
afifirfi}
ude

OIHEUIO

 

D. NEUROCHEMICAL INDICES OF DOPAMINERGIC NEURONAL
ACTIVITY

The rate limiting enzyme of dopamine synthesis, TH is intricately regulated on
multiple levels: transcription of mRNA, multiple mRNA splice variants, differential RNA
stability, translation, differential (protein) stability, allosteric modulation, phosphorylation
state, end-product inhibition and inhibitory pre-synaptic autoreceptors (Kumer and Vrana,
1996). The net effect of this diverse and redundant system of regulation, is a tight
coupling of dopamine synthesis, release and metabolism. As a consequence of this
coupling, steady state concentrations of dopamine remain fairly constant within the
neuron despite alterations in rates of release (Moore, 1987a). Accordingly, in vivo
neurochemical estimation of TH activity has been used as an index of dopaminergic
neuronal activity (Roth et al., 1976; Lookingland et al., 1987a; Lindley et al., 1988; Tian
etaL,l99l)

In the past, neurochemical estimates of dopaminergic neuronal activity have been
made following inhibition of dopamine synthesis. For example, a—methyltyrosine inhibits
TH and causes an exponential decline in dopamine concentrations at a rate proportional
to the level of neuronal activity (Brodie et al., 1966; Fuxe et al., 1969; Lofstrom et al.,
1976; Lookingland and Moore, 1984). Similarly, 3-hydroxybenzylhydrazine inhibits
AAAD and causes accumulation of DOPA at a rate proportional to the level of neuronal
activity (Carlsson et al., 1972; Murrin and Roth, 1976; Demarest et al., 1979; Demarest
and Moore, 1980). In most cases these indices provide an accurate and reliable estimate

of neuronal activity. However, these drugs require 30 min or more for product decline, or

are n.
bloc}.
IGULIL‘I
fioni:
neuroy-
conccr
interpr;

Pfiflacz

Concen:
the (1133
Of DO}.-
ACllet}
COHCCUIr
legcukg
meOTIIC
liner do.

 

accumulation, respectively, and are therefore not well suited for the estimation of acute
changes in dopaminergic neuronal activity.

Additionally, neurochemical methods involving inhibition of dopamine synthesis
are not ideally suited for the estimation of TIDA neuronal activity. Since these drugs
block dopamine synthesis and newly synthesized dopamine is preferentially released
(Gudelsky and Porter, 1979), dopamine release is blocked and the secretion of prolactin
from the pituitary is disinhibited. This precludes the concomitant measurement of TIDA
neuronal activity and the functional correlate of TIDA neuronal activity, plasma prolactin
concentrations. Furthermore, the resulting hyperprolactinemia complicates the
interpretation of such experiments because control groups have been exposed to elevated
prolactin for 30 min or more which may influence neuronal activity.

Estimation of the activity of TIDA neurons by measuring changes in
concentrations of the metabolite DOPAC in terminal regions of these neurons overcomes
the disadvantages of neurochemical procedures employing TH inhibition. Briefly, the use
of DOPAC as an index of neuronal activity is based upon the following rationale.
Activity-induced dopamine release is accompanied by an increase in cytoplasmic
concentration of dopamine synthesized de novo. A percentage of this cytoplasmic non-
vesicular dopamine pool is metabolized to DOPAC by monoamine oxidase at a rate
proportional to de novo synthesis (Lookingland et al., 1987a). Hence, procedures which

alter dopaminergic neuronal activity produce proportional changes in DOPAC

concentrations.

 

local
are L.
et a1.
trims
extra;
canto-d

low.

 

 

E. NEURAL ACTIVITY MAPPING WITH INDUCIBLE TRANSCRIPTION
FACTORS

Ten years after initial discovery, neural activity mapping with inducible
transcription factors is now widely utilized and accepted. This technique has led to the
localization and neurochemical identification of neurons within specific brain regions that
are uniquely responsive to a variety of stimuli (Sheng and Greenberg, 1990; Pennypacker
et al., 1995; Chaudhuri, 1997; Harris, 1998). Neural activity mapping with inducible
transcription factors is based upon the following rationale. Neurons respond to
extracellular stimulation by modulating the expression of certain immediate early genes
encoding transcription factors. The basal expression of these genes in quiescent cells is
low. Upon stimulation, genes encoding transcription factors are rapidly and transiently
transcribed (Greenberg et al., 1985; Morgan and Curran, 1986; Bartel et al., 1989). These
transcription factors regulate the expression of late response genes whose expression is
modulated over a time frame of hours (Merlie et al., 1984; Castellucci et al., 1988;
Goldman etal., 1988). The products of the late response genes are thought to serve more
specific effector functions in the neuronal response.

One of the best studied immediate early genes is c-fos. In the last 10 years a great
deal has been learned about the molecular mechanisms underlying the induction of c-fos
and the role of its protein product Fos in expression of late response genes. As shown in
figure 1.3, external stimuli (such as neurotransmitters or membrane depolarization) that
cause increased concentrations of intracellular CAMP or calcium result in
phosphorylation of cAMP response element binding protein (CREB) and its activation as

a transcription factor. This phosphorylation is mediated by protein kinase A (PKA) for

 

Cell .
Rec

 

 

 

Fig" re
In gene
IBPTEVI
Phase .
IEG, -’
I31 Diem
5Cer

 

 

 

  

 

 

Second messen ers I—
Cell Surface 9 (Eg. cAMP. Calciugn)
Receptors .I
Protein kinases T
(15.3.an PKC. Ca~CamK) J SRE IEG
'0 0'00 0
Constitutive TFs —+ (E.g. c-fos)
(5.3. CREB. SRF) I.
IEG mRN A ‘ IEG mRNA
C (5.3. c-fos mRNA) (E-s. c-fos mRNA) N
Y IE pro ein J IEG Dimer U
0 (8.3. Fos) F (E; Fos/Jun) L
S L, V E
0 1‘ 52_;;:;_ Late Response U
Gene
L J S
_. (E-s- TH)
LRG mRNA 4 LRG mRNA
(E.g. TH mRNA) 6':— (5-8- TH mRNA)
LRG protein _
H (E.g. TH protein) [1

 

 

Figure 1.3 Schematic representation of extracellular signal-mediated long-term changes
in gene expression in a neuron (modified from Hughes and Dragunow, 1995).
Abbreviations: AP- 1 , activator protein-1 site; Ca-CamK, calcium-calmodulin-dependent
kinase; CRE, CANIP response element; CREB, CAMP response element binding protein;
IEG, immediate early gene; LRG, late response gene; PKA, protein kinase A; PKC,
protein kinase C; SRE, serum response element; SRF, serum response factor; TFs,
transcription factors.

10

 

 

(PKC
Phos;
pronfi.

detect

 

 

 

express

Dram

in I316 1

r3“? of
{lemme

FOS‘IJUI’]

anfifin.

flan“.

...<

the cAMP pathway, and via unidentified calcium/calmodulin kinase (CAMK) for the
calcium pathway. Phosphorylated CREB interacts with calcium response elements
(CaRE) and CAMP response elements (CRE) within the promoter of immediate early
genes like c-fos to stimulate initiation of transcription. Transcription of c-fos may also be
induced by various signaling pathways that converge upon activation of protein kinase C
(PKC). PKC phosphorylates and thereby activates a serum response factor (SRF).
Phosphorylated dimeric SRF interacts with the serum response element (SRE) in the
promoter of the gene to stimulate its transcription. C-fos mRNA may be transiently
detected 15-30 minutes following stimulation of the neuron (Greenberg et al., 1985;
Morgan and Curran, 1986; Bartel et al., 1989; Hoffman and Murphy, 2000). The
transient expression of c-fos mRNA is due to its inherent instability (10-15 minute half-
life) and end product inhibition. C—fos mRNA is translocated to the cytosol where it is
translated into Fos protein. Generally, these proteins appear 60-90 minutes after
stimulation and persist for 2-5 h (Hoffman et al., 1992; Hoffman and Murphy, 2000).

Fos proteins translocate to the nucleus and heterodimerize with constitutively
expressed Jun-related transcription factors (Morgan and Curran, 1989; Hughes and
Dragunow, 1995). These heterodimers bind AP-l elements (Sheng and Greenberg, 1990)
in late response gene promoters (Kumer and Vrana, 1996) and modulate the transcription
rate of these genes (e.g. TH: Icard-Liepkalns et al., 1992; Guo et al., 1998, and
neurotensin: Kislauskis and Dobner, 1990; Bullock et al., 1994). The effect of the
Fos/Jun heterodimer on transactivation rate of the late response gene depends upon the
combination of Fos- and Jun-related transcription factors composing the dimer. For

example, c-Fos/c-Jun greatly enhances transcription rate, whereas c-Fos/Jun-B represses

ll

 

it. 1'

by a

Shoul
cell 1;
all [m

I99T;I

 

 

 

903161.

it. Ultimately, over a time frame of hours, the late response gene products are synthesized
by a protein synthesis-dependent mechanism.

Neural activity mapping using inducible transcription factors has limitations that
should be recognized to avoid erroneous conclusions. The first consideration is termed
cell type expression specificity and is based on the observation that not all cells produce
all transcription factors. Thus, negative results demand carefirl interpretation (Chaudhuri,
1997; Harris, 1998). Another complexity of neural activity mapping using inducible
transcription factors is due to the convergence/redundancy of multiple signaling pathways
leading to stimulus-coupling uncertainty. Stimulus-coupling uncertainty can make a
causal link between changes in gene expression and a stimulus difficult to establish and
highlights the importance of appropriate controls (Chaudhuri, 1997; Harris, 1998).
Chronically stimulated neuronal systems exemplify stimulus-coupling uncertainty and are
not suited to neural activity mapping using inducible transcription factors. It should also
be recognized that the disappearance of Fos expression does not reliably indicate signal
termination. Fos signal offset is complicated by the time course of Fos per se, stimulus-
coupling uncertainty and staining parameters (Hoffman and Murphy, 2000). The
induction of F 03 does not always indicate that a neuron’s firing rate has increased since
growth factors and other neuromodulators that influence levels of second messengers may
influence Fos expression without concurrently affecting firing rate of the cell (Hoffman
and Murphy, 2000). In an effort to avoid the common pitfalls of neural activity mapping
with inducible transcription factors, this dissertation conservatively considers only

positive data afier acute stimulation and not causal molecular relationships.

On the other hand, neural activity mapping using transcription factor products (i.e.
proteins) like Fos, instead of its mRNA is particularly advantageous. For example, Fos
proteins have greater stability and longer half lives than the mRNA which allows a
greater window for detection. Since Fos synthesis has a latency of at least 30 minutes,
animal handling artifact is minimized (Hoffman and Murphy, 2000). In the TIDA system,
the use of antiserum directed at the conserved region of Fos proteins (i.e. the leucine
zipper) that recognizes Fos and related antigens (F RA) has been shown to increase
sensitivity, and allows the detection of decreases in immediate early gene expression
since some FRA may have prolonged expression, unlike Fos (Hoffman et al., 1992).
Major advantages of this technique are its superior cellular resolution and nuclear
localization of signal. Since FRA proteins are localized within the nucleus dual
immunohistochemistry can be utilized to chemically identify the neuron (via detection of
a key synthetic enzyme in the cytosol) and determine its responsiveness to particular
stimulus. Therefore, the technique may be applied to study subpopulations of neurons that
are ftmctionally unique and intermingled with chemically similar neurons. Furthermore,
the pattern of Fos expression in these neurons can suggest a mechanism for recruitment of
inactive neurons into an activated pool. Accordingly, immuno-histochemical detection of
neurons expressing immediate early gene products represents an early neurochemical
marker of responsiveness and permits the identification of functional heterogeneity in

subpopulations of chemically similar neurons within discrete brain regions.

13

 

sub;

I
dors

 

l)\i~.

113110

 

lli-ll
1181-;
Piece
Inedi
IIaik.
Bruit

Fttte

F. SUBPOPULATIONS OF TIDA NEURONS

There is anatomical and histochemical evidence to divide TIDA neurons into
subpopulations in the ARC. As shown in figure 1.4, the ARC may be divided into
dorsomedial (DM) and ventrolateral (VL) regions by extending a line dorsolaterally at an
angle of 30-50° from the ventrolateral aspect of the third ventricle. The magnocellular
DM-ARC contains TH-immunoreactive (IR) perikarya of TIDA neurons whereas the
parvocellular DM-ARC, located in the ventromedial aspect of the DM-ARC, is devoid of
TH-IR and hence TIDA neurons (Halasz et al., 1985). TIDA neuronal perikarya in the
DM-ARC are smaller and have more intense TH-IR staining than those of the VL-ARC.
Processes from neurons in the DM-ARC are oriented dorsoventrally and terminate in the
medial aspect of the median eminence (Kobayashi and Matsui, 1969; Rethelyi, 1985;
Daikoku et al., 1986) in contrast to those from the VL-ARC which course mediolaterally
and terminate in the lateral aspect of the median eminence (Kobayashi and Matsui, 1969;
Fuxe et al., 1986).

In addition to anatomical differences, there is a discrepancy in TH- and AAAD-IR
between the DM and VL subdivisions of the ARC. TIDA neurons in the DM-ARC
contain both TH-IR and AAAD-IR, whereas those in the VL-ARC contain only TH-IR
(Skagerberg et al., 1988; Misu et al., 1996). Therefore, neurons in the DM-ARC contain
the fiill complement of enzymes necessary to synthesize dopamine, whereas those in the
VL-ARC lack AAAD and produce only DOPA (Meister et al., 1988a; Komori et al.,
1991; Misu et al., 1996; Skagerberg et al., 1988). The different characteristics of

subpopulations of TIDA neurons probably reflects functional heterogeneity.

14

dommmzvnzm Ezzoootenéemvoaofiov
65 E 20.50: 5+5. emu: mo 35ch 05 can: women 3393 HEB—8023 v5 HEB—coocwefi 85 32233 35.3 33 0M4.
Evofioflov BE. 22.2“? BE“ 05 mo noon—me Bone—ES.» 05 82m ocmém me 2mg 5 “a 38828.8 0:: a wfivaofio .3 Somme."
Bogota? was Egofioflov 85 3233.5 83 ONE 2:. .maommmzvnam UM? ”Ems A33 :3 «o 88m 88% vommvoev 8308
35% 05 mo mamfiweomfiouonm .«o “$888803 e Boa Aug—£508 m<m 45535 E 25 ~ ; ”mo—menu x81: 5+3. .«o wasps—Q
”EOE .maommngnm 0% 05 5 ME; 2658355 wawofiou 0% 0628 o5 4385 cowoom Each a mo wEBEQ vé 9:.me

51 0mm

. x In». a. ..
new... ou=oEEH Saves

 

‘ \ “fl“ ambit...“ I o... . \\
«...... .. ... 4
... s .. so..\vl.vnp\.\ Hus \
.....v d
.. . «.5 a e.—
n...... . _ a _ 23> x; .
.. ......»
nib V. .1?“ 1.1.111 \
... ua_==3e>§m \II 1\\
... .... '..... \
.. a a. -.anEeEen \
r..v.\- 1.... WW. \
.. x
. .l’oi; \\
N .. \.~. \
..........3 .3 = ooecwa
. .....n..«.... .. .__.:_._x_.==8..camH
, as...
, + . I .
’.. \
/ . x
x o a x
/ .N- \
1.... \

 

mZO—mgmanm mDmAUDZ HH<DU-<

15

di 13'

6X

0V3!

func

neur

thee

 

 

In support of functionally distinct TIDA neuronal populations, neurons from
different subdivisions demonstrate differential responsiveness to selected stimuli. For
example, activity of TIDA neurons in the DM-ARC, but not VL-ARC, is decreased after
ovariectomy (Cheung et al., 1997), suggesting an estrogen-associated reproductive
function. In the studies described herein, responsiveness of subpopulations of TIDA
neurons is described by ARC subdivision for the detection of functional heterogeneity in

the context of hyperprolactinemia.

{CC'

ICC;

{ECG
dop.

HCUY

meta?
(Guc-V

aUIOr;

 

G. REGULATION OF CENTRAL DOPAMINERGIC NEURONS

1. Dopamine receptor-mediated mechanisms

The activity of central dopaminergic neurons is tightly regulated by dopamine
receptor-mediated mechanisms. However, there are significant differences in the
regulation of mesotelencephalic and TIDA neurons. Traditional non-selective dopamine
receptor agonists (e.g. apomorphine, bromocriptine) inhibit NSDA and MLDA neuronal
activity by acting at presynaptic dopamine autoreceptors and/or postsynaptic dopamine
receptors on neurons that feed back to inhibit these systems. Conversely, non-selective
dopamine receptor antagonists (e.g. haloperidol, raclopride) stimulate NSDA and MLDA
neuronal activity by acting at these same receptors.

In contrast, the activity of TIDA neurons is not acutely affected by non-selective
dopamine receptor agonists and antagonists (Gudelsky and Moore, 1976; Demarest and
Moore, 1979; Moore, 1987a; 1992). For example, for up to 2 h after the administration of
haloperidol, neurochemical estimates of the rates of dopamine synthesis, turnover and
metabolism are increased in termini of NSDA and MLDA, but not TIDA neurons
(Gudelsky and Moore, 1977). These data are consistent with the presence of dopamine
autoreceptors on NSDA and MLDA, but not TIDA neurons. Although TIDA neurons are
not directly responsive to dopamine receptor antagonists, studies show increased
dopamine synthesis (Demarest et al., 1984; Moore, 1987a), turnover (Gudelsky and
Moore, 1976) and metabolite formation 12 h after haloperidol administration (Gudelsky
and Moore, 1977) as a result of their ability to increase circulating concentrations of
prolactin (Moore and Lookingland, 1995). Hence, until recently, it was thought that

mesotelencephalic, but not TIDA neurons are acutely regulated by dopamine receptor-

l7

 

ant

achx
dopa
EaIOI
acfixt
endoé
dom—
1993i
coukj

“Tepu

 

 

mediated mechanisms.

Yet with the advent of selective “second generation” dopamine agonists and
antagonists, studies have revealed that TIDA neurons are also acutely regulated by
dopamine receptor-mediated mechanisms that act independently of prolactin. TIDA
neurons are regulated by inhibitory D. dopamine receptors (Durham et al., 1998).
However, this D dopamine receptor-mediated regulation is activity dependent; i.e.
selective D1 agonists have little effect on basal activity of TIDA neurons, but inhibit
activated TIDA neurons (Berry and Gudelsky, 1990). On the other hand, selective D2
dopamine receptor agonists stimulate TIDA neuronal activity (Berry and Gudelsky, 1991;
Eaton et al., 1993a; 1993b) via an afferent mechanism involving disinhibition of tonically
active dynorphinergic interneurons (Durham et al., 1996). However, there is little
endogenous dopamine tone at these receptors under basal conditions since selective D2
dopamine receptor antagonists have no acute effect on TIDA neurons per se (Eaton et al.,
1993b). The opposing actions of inhibitory D1 and stimulatory D2 dopamine receptors
could account for previous reports of lack of net effect of non-selective dopamine

receptor agonists on TIDA neurons (Durham et al., 1998).

IS

6.”.

at".
Ilt‘.

U)J

I681)
in b<
to 1%

Sl'Ste

expeq
1985 )
depen
expres

I991).

 

2. Prolactin feedback

As depicted in figure 1.5, dopamine released from TIDA neurons in the median
eminence travels in the portal blood to the pituitary gland where it tonically inhibits the
secretion of prolactin from anterior pituitary lactotrophs. Prolactin, in turn, feeds back to
activate TIDA neurons through an unknown mechanism. Feedback activation of TIDA
neurons by prolactin is comprised of interrelated tonic and inductive components
(Demarest et al., 1984; 1986). In female rats, the tonic component determines the basal
rate of prolactin secretion by regulating short-term changes in TIDA neuronal activity in
response to acute changes in circulating prolactin concentrations. This component sets the
basal level of activity of TIDA neurons with a short latency of 2-4 h.

The inductive component involves a change in the capacity of TIDA neurons to
respond to prolonged periods of hyperprolactinemia (Moore et al., 1987b) and is observed
in both males and females. The inductive component stimulates TIDA neuronal activity
to levels 2-3 times basal (Demarest et al., 1985a; Selmanoff, 1985) 12 h following
systemic (Hokfelt and Fuxe, 1972; Selmanoff, 1981; Demarest et al., 1984; 1986) or
central (Annunziato and Moore, 1978) administration of exogenous prolactin, or after
experimental manipulations which result in endogenous hyperprolactinemia (Moore et al.,
1985). This inductive component of prolactin feedback regulation of TIDA neurons is
dependent upon protein synthesis (Johnston et al., 1980) and is associated with increased
expression of TH mRNA in the ARC (Arbogast and Voogt, 1991; Selmanoff et al.,
1991)

Both components are interdependent in that the induction component is expressed

via the tonic component (thus its expression requires elevated prolactin levels), and the

19

 

 
 

 

 

 

 

 

 

 

/ \
/ \
+ ‘7 \
I / 3V \
// Arcuate \\ \
z ’ > l Nucleus \ \
/ +
l C D 1
Median
Eminence/
Dopamine
. Hypophysial
Prolactin Portal System

 

# D2 Dopamine

- T Receptor

Kl’ituitary
Qtotrophs

 

Figure 1.5 Schematic representation of the feedback regulation of TIDA neuronal
activity and prolactin secretion (modified from Moore, 1987a). Abbreviations: 3V, third
ventricle.

20

 

IT.

Cir

prl

 

magnitude of the change in TIDA neuronal activity in response to acute changes in
circulating prolactin is modulated via the inductive component by the preceding history of

prolactin exposure (Demarest et al., 1984; 1986).

21

ph)
acti
cho
Loo
neul
hm
indu
im't‘x'
h} d:

Mal

er 211..

MP;

 

GABA

the GA

not [OK:

 

3. Afferent regulation of TIDA neurons

The basal tonic activity of TIDA neurons is acutely regulated by a wide variety of
afferent neuronal inputs. TIDA neuronal activity is acutely inhibited during selected
physiological states such as stress and lactation. Following immobilization stress the
activity of TIDA neurons in female rats is acutely inhibited in a neuronal circuit involving
cholinergic and 5-hydroxytryptaminergic mediators (Moore, 1987a; Moore and
Lookingland, 1995). Similarly, in the lactating rat suckling rapidly inhibits TIDA
neuronal activity (Demarest et al., 1983) and decreases dopamine content in the
hypophysial portal blood (Ben-Jonathan et al., 1980). While the mechanism of suckling-
induced inhibition of TIDA neurons is largely unknown, there is evidence that this circuit
involves the endogenous mu Opioid enkephalin (Callahan et al., 1996), 5-
hydroxytryptamine (Demarest and Moore, 1981a) and acetylcholine (Moore and
Lookingland, 1995).

Neuropeptides and neurotransmitters also acutely inhibit TIDA neuronal activity.
In male rats, the endogenous kappa opioid dynorphin tonically inhibits TIDA neuronal
activity (Manzanares et al., 1991; 1992), whereas the peptide galanin elicits activity-
dependent inhibition of TIDA neurons, affecting only activated TIDA neurons (Gopalan
et al., 1993). The excitatory amino acid neurotransmitter glutamate, acting through
AMPA receptors, tonically inhibits TIDA neuronal activity via a circuit involving
GABAA receptors (Wagner et al., 1994a; 1994b). Like GABAA receptors, activation of
the GABAB receptor isoforrn also inhibits TIDA neuronal activity, yet this inhibition is
not tonic (Wagner et al., 1994c).

On the other hand, TIDA neuronal activity is acutely stimulated by other

22

DC
31
Pa
Cl .
gle’
l9“

06L

 

neuropeptides and neurotransmitters. For example, the peptides bombesin (Manzanares et
al., 1991; Toney et al., 1992a; Lin and Pan, 1993), neurotensin (Gudelsky et al., 1989;
Pan et al., 1992; Lin and Pan, 1993) and alpha-melanocyte stimulating hormone (Lindley
et al., 1990a) each acutely stimulates TIDA neuronal activity. Additionally in female rats,
glutamate acts via NMDA receptors to tonically stimulate TIDA neurons (Wagner et al.,
1993). However, there exist no data regarding the role of afferent neurotransmitters or

neuropeptides in the delayed inductive activation of TIDA neurons by prolactin.

23

ET.

 

I88

M“,

fror

 

 

H. NEUROTENSIN

1. Synthesis

Twenty five years ago the tridecapeptide neurotensin was isolated from bovine
hypothalamus (Carraway and Leeman, 1973), sequenced (Carraway and Leeman, 1975a)
and synthesized (Carraway and Leeman, 1975b). The biological activity of this peptide
resides predominantly in the carboxyl terminal hexapeptide region (i.e. residues 8-13)
whose amino acid identity is highly conserved across vertebrates (Granier et al., 1982).

Synthesis of neurotensin commences with transcription of neurotensin mRNA
from its respective gene. Since the fourth exon codes for the entire neurotensin precursor,
the elaboration of splice variants is precluded. Regulation of neurotensin gene expression
depends upon interactions of various transcription factors with the promoter region of the
gene. The promoter includes consensus sequences for several cis-acting elements,
including an AP-l site. Mutations of the AP-l site vastly inhibit the expression of
neurotensin mRNA following the application of known inducers (Kislauskis and Dobner,
1990) suggesting that AP-l complexes have a role in the regulation of this gene.

Similar to other neuropeptides, neurotensin is synthesized as part of a larger
precursor protein (170 amino acids). A single copy of neurotensin exists near the carboxyl
terminus (Dobner et al., 1987; Kislauskis et al., 1988). Proneurotensin is proteolytically
cleaved during axonal transport, rather than in the neuronal cell body (N icot et al., 1995).

A consistent product is derived from processing without numerous post-translational

variants.

24

ARt
CI a.‘
Ct 211‘
V1,
(Ere;

‘31 al.

which

“ten:

and};

 

2. Central distribution

Neurotensin exhibits heterogeneous regional distribution (Jennes et al., 1982; Uhl
et al., 1976, 1977). Neurotensin-IR is abundant in the central nervous system (Carraway
and Leeman, 1976; Goedert et al., 1985). Neurotensin—IR is found in particularly high
levels in the hypothalamus, and the majority of neurosecretory cells capable of
neurotensin synthesis are located in the ARC (Merchenthaler and Lennard, 1991). The
neurotensin-IR fibers from these neurons densely ramify over the neurohemal zone of the
median eminence (Jennes et al., 1982; Ibata et al., 1984; Kiss et al., 1987).

Immunohistochemical colocalization studies suggest the existence of two
functionally distinct groups of neurotensin neurons located in different sudivisions of the
ARC. In the DM-ARC, neurotensin is colocalized with TH (Hokfelt et al., 1984; Everitt
et al., 1986; Meister and Hokfelt, 1988b), galanin and GABA (Ibata et al., 1983; Hokfelt
et al., 1984; Everitt et al., 1986; Meister and Hokfelt, 1988b; Ciofi et al., 1993). In the
VL-ARC, neurotensin may be colocalized with grth hormone-releasing hormone
(Everitt et al., 1986; Meister and Hokfelt, 1988b; Niimi et al., 1991) and/or TH (Hokfelt
et al., 1984; Everitt et al., 1986; Meister and Hokfelt, 1988b) but not AAAD (Meister et
al., 1988b; Okamura et al., 1988a). There are conflicting reports concerning the extent to
which TH and neurotensin coexist in the DM-ARC. Some studies report coexistence is
extensive (Hokfelt et al., 1984; Everitt et al., 1986) while others report it is rare (Meister

and Hokfelt, 1988b).

25

at:

lex

at q
sup

pat}

in In

 

“1*..th

3. Receptors

Radioligand binding experiments have revealed two classes of neurotensin
binding sites, low and high affinity (Mazella et al., 1983; Kitabgi et al., 1985). The low
affinity sites selectively bind levocabastine, whereas the high affinity sites are relatively
levocabastine-insensitive (Kitabgi et al., 1987; Schotte et al., 1988).

The high affinity neurotensin binding sites have been classified as NTR1 (Mazella
et al., 1985). Cloning and expression studies demonstrate that this receptor belongs to the
superfamily of G-protein—coupled receptors and is linked to multiple signal transduction
pathways. Known NTRl-associated second messenger systems include PKC-dependent
intracellular calcium mobilization (Berry and Gudelsky, 1992), formation of cAMP
(Yamada et al., 1993), phosphatidyl inositol hydrolysis (Watson et al., 1992) and the
formation of nitric oxide (Marsault and Frelin, 1992). The low affinity levacobastine-
sensitive neurotensin binding sites, also belonging to the superfamily of G-protein-
coupled receptors, have been classified as NTR2 (Chalon et al., 1996).

Neurotensin receptors are distributed throughout the hypothalamus in moderate
density (Rostene and Alexander, 1997). In the hypothalamus, the NTR2 isoforrn is
predominantly expressed by glial cells, whereas the NTR1 isoforrn is predominantly
expressed by neurons (Schotte et al., 1988). NTR; is distributed throughout the
mediobasal hypothalamus on both nerve termini and perikarya (Dana etal., 1989; Boudin
et al., 1996). Within the mediobasal hypothalamus, NTR] mRNA has been localized
within neurons of the DM-ARC (Goedert et al., 1984; Meister et al., 1989), including

TIDA neurons per se (Alexander, 1997).

26

4. Central actions

Neurotensin stimulates central dopaminergic systems (e.g. NSDA and MLDA
systems), both as a neuromodulator and neurotransmitter. Neurotensin modulates the
effects of other transmitters on neuronal firing rate even at concentrations too low to
influence basal activity (Baldino et al., 1985a). For example, neurotensin promotes
repetitive firing of cholinergic neurons of the basal forebrain (Alonso et al., 1994) by
abrogating the prolonged calcium-dependent, potassium-mediated hyperpolarization
following stimulation (Farkas et al., 1994). Neurotensin also facilitates basal and
potassium-induced dopamine release in central dopaminergic systems (Okuma and
Osumi, 1982).

Within the ARC of hypothalamic slices, neurotensin produces excitatory
electrophysiological effects that persist in calcium-free medium in vitro (Herbison et al.,
1986; Lin and Pan, 1993). Furthermore, in vivo, administration of neurotensin produces
dose-dependent activation of TIDA (Widerlov et al., 1982; Gudelsky et al., 1989; Pan et
al., 1992), NSDA (Pinnock, 1985) and MLDA (Widerlov et al., 1982) neurons.

Given the distribution of neurotensin and its receptors within the mediobasal
hypothalamus and the stimulatory effects of this peptide on TIDA neurons, it is possible

that neurotensin has a role in the delayed inductive effects of prolactin on TIDA neurons.

27

 

err

dc}:

 

tem

oft]

I. THESIS OBJECTIVE

The studies described herein aim to test the hypothesis that neurotensin mediates
the delayed inductive activation of TIDA neurons by prolactin. If neurotensin mediates
the delayed inductive activation of TIDA neurons by prolactin, then: prolactin will elicit
temporally-dependent neurochemical effects and changes in immediate early gene
expression in TIDA neurons; the neurotensin receptor antagonist SR-48692 will block the
delayed-inductive activation of TIDA neurons by prolactin; and prolactin will elicit
temporally-dependent changes in immediate early gene expression in neurotensin neurons

of the mediobasal hypothalamus.

28

 

Tel.

CSII

 

mg.
day
Of 5'

days

2. MATERIALS AND METHODS

A. ANIMALS

All experiments were performed using gonadally-intact male and female Long-
Evans rats weighing 200-225 g at time of purchase from Harlan Laboratories
(Indianapolis, IN). Animals of the same gender were housed 1-3 animals per cage,
maintained in a temperature- (22 i 1°C), humidity- (30 - 60%) and light-controlled
(illuminated between 06:00 - 18:00 h) enviromnent, and provided with food (Harlan
Teklad; Bartonville, IL) and tap water ad libitum. In experiments involving female rats,
estrous cycles were monitored daily by microscopic examination of epithelial cytology of
vaginal lavage, and only those animals displaying a vaginal smear consistent with the first
day of diestrus afier two consecutive cycles were used. In order to minimize the effects
of stress, animals were allowed to accommodate to their housing for an average of 3-7
days before use, were handled daily in a manner similar to that required of the day of

experimentation and were decapitated in a room separate from yet adjacent to their

housing.

B. DRUGS

The dose of haloperidol (Sigma Chemical Co., St. Louis, MO) was calculated as
its free base and dissolved in 0.3% tartaric acid (Mallinckrodt Chemical Works, St. Louis,
MO). All other doses of drugs were calculated as the appropriate salt. 2- (1- (7-Chloro-4-
quinolinyl) -5- (2,6-dimethoxy-phenyl) pyrazol-3-yl) carbonylaminotricyclo (3.3.1.1.3'7)
decan-2-carboxylic acid (SR-48692) was kindly supplied by Dr. Danielle Gully (Sanofi

Recherche; Toulouse, France) and solubilized in 0.9% saline containing 10% dirnethyl

29

m7.

 

in 1
ad; ‘

mu

9‘

'f1

sulfoxide (DMSO; Sigma). Distilled water was used to dissolve biological rat prolactin
(NIDDK-rPRL-B-7; Lot # AFP-64SZB) generously supplied by Dr. A.F. Parlow (National
Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), National Hormone
and Pituitary Program (N HPP).

The anesthetic Equithesin was utilized in two instances: (1) prior to intracerebro-
ventricular cannulation surgery, and (2) minutes prior to transcardial perfusion and
fixation for immunohistochemistry experiments. Equithesin was prepared by stirring
42.51 g chloral hydrate (Sigma), 9.72 g sodium pentobarbital (Sigma) and 21.26 g
magnesium sulfate in 443 ml of warmed 1,2-propanediol (propylene glycol). Afier these
compounds were completely dissolved, 120 ml of 95% ethanol was added and the total
volume was brought to 1000 ml with double distilled water. The final solution was
measured into 100 ml aliquots and stored in amber drug bottles at room temperature.

Antiserum to rat prolactin (rPRL-B-6; NIDDK NHPP) was generated previously
in rabbits. Prolactin was dissolved in normal saline, emulsified in complete Freund’s
adjuvant (Sigma) in a ratio of 1:2 (v/v), and 1 mg equivalents were injected s.c. at
multiple sites in a total volume of 1 ml. Booster injections of 200 pg prolactin emulsified
in saline and incomplete Freund’s adjuvant were repeated monthly until a high antibody
titer was obtained.

Prolactin antiserum specificity and titer were determined by radioirnmunoassay.
Specificity was determined by incubating prolactin antiserum (126,000) and radiolabeled
rat prolactin with various amounts (0-1 pg) of unlabeled rat prolactin, growth hormone,
thyroid-stimulating hormone and luteinizing hormone (NIDDK NHPP). Unlabeled rat

prolactin competed on an equimolar basis with radiolabeled prolactin for binding to the

30

pit“

an:

8112';

F"

 

 

prolactin antiserum with a lower limit of sensitivity of 50 pg/tube. In contrast, prolactin
antiserum did not bind with up to 100 ng/tube of growth hormone or 1 ug/tube of thyroid-
stimulating hormone or luteinizing hormone (Toney et al., 1992b).

Doses and routes of administration of drugs, biological rat prolactin and prolactin

antiserum are indicated in the legends of the appropriate figures.

C. SURGICAL MANIPULATIONS
I. Intracerebroventricular cannulation and injections

Intracerebroventricular (i.e.v.) injections were administered to freely moving rats
via cannula guides which were implanted 3-14 days prior to the experiment. Rats were
anesthetized with Equithesin (3 ml/kg, i.p.) and positioned in a stereotaxic frame (David
Kopf Instruments, Tujunga, CA) with the incisor bar set at 2.4 mm below the horizontal
plane (Paxinos and Watson, 1986). A 23-gauge stainless steel guide cannula, 10 mm in
length, was implanted 1.4 mm right lateral to bregma and 3.2 mm below the dura mater
(Paxinos and Watson, 1986), and anchored to the skull with stainless steel screws (Small
Parts Inc., Miami Lakes, FL) and dental cement (Dentsply, York, PA). Following surgery
rats were injected (0.2 ml/rat, i.m.) with Combiotic (200,000 U/ml procaine penicillin G
and 250 ug/ml dihydrostreptomycin sulfate; Pfizer Inc., New York, NY) per AUCAUC
guidelines and a stainless steel stylet (10 mm in length) was inserted in the guide cannula
to maintain its patency. Male rats were allowed 3-5 days for recovery prior to the
experiment. Female rats were utilized after two regular estrous cycles following the
surgery (i.e. 8-14 days recovery). At the time of the experiment, prolactin or vehicle was

injected into freely moving rats over a minute interval in a volume of 3 u]. The volume

31

st
Uh
ha

of

“e
mo
Che
Cep
PE.
(Sch
hag

met

finer
hen
The:
thq

03ij

 

was delivered with a 10 pl Hamilton microsyringe (Reno, NV) connected by a 30 cm
length of PE-lO tubing (Clay Adams/Beckton-Dickinson, Sparks, MD) to a 30-gauge
stainless steel injector which protruded 1 mm beyond the tip of the guide cannula and into
the lateral ventricle. Cannula placement was verified post-mortem in appropriate frontal
brain sections with the aid of a dissecting microscope and only those animals with the tip

of the cannula tract in the right lateral cerebral ventricle were used in the study.

2. Intravenous (external jugular) catheterization and injections

Intravenous injections of prolactin antiserum or its normal rabbit serum vehicle
were administered to conscious, freely moving rats via catheters implanted on the
morning of the experiment. Rats were anesthetized with diethyl ether (J .T. Baker
Chemical Co., Philipsburg, NJ), the right external jugular vein was exposed and its
cephalic aspect ligated. A segment of polyethylene tubing (I.D. 0.58 mm, CD. 0.96 mm;
PE-SO; Clay Adams/Becton-Dickinson, Sparks, MD) full of 12.5% (v/v) heparinized
(Schein Pharmaceutical Inc., F lorham Park, NJ) saline was inserted through a nick in the
distal right external jugular vein. The tubing was gently threaded caudally to the level of
the right atrium where placement and patency of the catheter was verified by aspiration of
dark venous blood. The catheter was then anchored at two points along its course
through the external jugular vein, and the free end of the catheter was tunneled s.c. over
the right shoulder and out a small opening over the dorsal aspect of the cervical spine.
The incision was closed with two surgical staples. The catheter was purged with 200 pl
of heparinized saline, trimmed to 2-4 cm external length and capped with the tip of a

common pin. The entire procedure was completed in 7-14 minutes. After surgery,

32

 

Ct

ar.

ICI‘.‘
CUl

COT.

a g1.
Pall.
flelc‘

Ola

 

individually caged rats were allowed to recover for 3-4 hours prior to use. At the time of
intravenous injections, the patency of the catheter was verified via aspiration of blood.
Conscious, freely moving rats with patent catheters were slowly injected with prolactin
antiserum (200 til/rat) or its normal rabbit serum vehicle (200 ul/rat) followed by 200 pl

of 12.5% (v/v) heparinized saline.

D. NEUROCHEMISTRY
1. Tissue preparation

Following appropriate treatments rats were decapitated and brains were quickly
removed. Brainstems were transected at mid-cerebellar level and placed upright on their
cut surfaces on aluminum foil directly over dry ice. Each brain was sliced into
consecutive 600 um frontal sections beginning at approximately interaural 11.20 mm and
bregrna 2.20 mm (Paxinos and Watson, 1986) and extending caudad in a -9 oC cryostat
(Harris model CTD, International Equipment Co., Needham, MA) and thaw-mounted on
a glass slide. Using a modification (Lookingland and Moore, 1984) of the method of
Palkovits (Palkovits, 1973; Palkovits, 1974; Palkovits and Brownstein, 1983) the terminal
fields of TIDA, NSDA and MLDA neurons were microdissected as follows. With the aid
of a stereoscope and a round 21-gauge punch tool (fashioned from a 21-gauge stainless
steel hypodermic needle; Beckton-Dickinson Co., Rutherford, NJ), the nucleus
accumbens was bilaterally dissected from sections 1 and 2 (as illustrated in figure 2.1)
and placed in 100 pl of tissue buffer (50 mM sodium phosphate; 30 mM citrate buffer;
pH 2.5). Similarly, the striatum was bilaterally dissected from section 2 (figure 2.1) and

placed in 100 u] of tissue buffer. The median eminence was dissected from sections 8 and

33

 

Fig
hat
nUc

Dun
Usir
Bra;

 

 

 

SECTION #2

Figure 2.1 Anatomical locations of nucleus accumbens (stippled) and striatum (cross-
hatched) samples collected for neurochemical analysis. With the aid of a stereoscope, the
nucleus accumbens was bilaterally microdissected from the 1" (bregma 2. 20 mm) and 2'“1
(bregma 1.60 mm) serially collected 600 um frontal sections using a round Zl-gauge
punch tool. Similarly, the striatum was bilaterally microdissected fi'om the 2"" section
using a round l8-gauge punch tool. Figures modified from Paxinos and Watson, The Rat
Brain in Stereotaxic Coordinates, Academic Press, New York, 1997.

34

115

2-2)

 

 

 

lnsl
1111c
em

861

811
he
few.

90

9 using an oval l8-gauge punch tool (as illustrated in figure 2.2) and placed in 50 pl of

tissue buffer containing 15% methanol. All tissues were stored at —20 °C until assayed.

2. Assays

On the day of the assay, samples were thawed, sonicated for 3 5 (Heat System-
Ultrasonics, Plainview, NY USA) and centrifuged (microfuge model #152, Beckman
Instruments, Inc., Palo Alto, CA) for either 30 5 (samples from median eminence and
nucleus accumbens) or 45 3 (samples from striatum). Supematants from median
eminence and nucleus accumbens were transferred via Hamilton microsyringe to a second
set of microtubes over ice; their volumes were brought to 65 pl/tube with tissue buffer
(50 mM sodium phosphate; 30 mM citrate buffer; pH 2.5) containing 15% methanol.
Supematants from striatum were similarly transferred to a second set of microtubes,
however the final volume of these samples was brought to 100 pl/tube. Volumes of eluent
for the various brain regions were previously established based upon relative
concentrations of catecholamines and the optimal range of assay detection.

The remaining tissue pellets were dissolved with 100 pl (median eminence and
nucleus accumbens) or 200 p1 (striatum) of 1.0 N sodium hydroxide for at least 24 h at
room temperature and spectrophotometrically assayed for protein (Lowry et al., 1951) as
follows. For median eminence samples, 100 pl of undiluted protein solution was added
by Hamilton microsyringe to labeled glass cuvettes (12 x 75 mm) which served as the
spectrophotometric reaction vessels. Fifiy p1 of each striatum protein sample was
combined with 50 pl of 1.0 N sodium hydroxide in appropriate glass cuvettes to attain a
four-fold dilution of the total protein content of the original microtube. Similarly, 50 pl of

35

 

 

SECTION #8

 

SECTION #9

e, the median eminence was
(bregma -3.30 mm) serially

°f
-2.56 mm) and 9

(bregma

dissected fi'om the 8‘“
collected 600 pm frontal sections using an oval 18-gauge punch tool. Figures modified

from Paxinos and Watson, The Rat Brain in Stereotaxic Coordinates, Academic Press,

New York, 1997.

Figure 2.2 Anatomical locations of median eminence (cross-hatched) samples collected
nucro

for neurochemical analysis. With the aid of a stereosc

36

 

 

 

5U

each nucleus accmnbens protein sample was combined with 50 pl of 1.0 N sodium
hydroxide in apprOpriate glass cuvettes to attain a 2-fold dilution of the total protein
content of the original microtube. Standards containing 12.5, 25 and 50 pg of bovine
serum albumin (Sigma) dissolved in 100 pl 1.0 N sodium hydroxide were also transferred
into appropriate glass cuvettes for the protein assay. For colorimetric reaction, 1 ml
reagent A (100:1:1, 0.19 M sodium carbonate : 40 mM cupric sulfate : 70 mM potassium
sodium tartrate) was added to each cuvette, vortexed and allowed 10 min incubation.
Subsequently, each cuvette received 100 pl of reagent B ( 1:1, 2 N Folin & Ciocalteu’s
phenol reagent (Sigma): double distilled water), was vortexed and allowed 30 min
incubation. Protein content of a sample was estimated by its absorbance on a Gilford
spectrophotometer (model Stasar III, Gilford Instrument Laboratories, Inc., Oberlin, OH)
tuned to 700 nm wavelength relative to standards run in the same reaction.

DOPAC and dopamine contents of standards and tissue samples were determined
concurrently using ESA Coulochem Liquid Chromatographs (model 5100A; Bedford,
MA) with electrochemical detection and Waters solvent delivery systems (model 510;
Milford, MA) set at a flow rate of 0.9 ml/min as described previously (Chapin et al.,
1986; Lindley et al., 1990b). Fifly pl of supernatant was injected onto a C18 reverse
phase analytical column (5 pm spheres; 250 x 4.6 mm; Biophase ODS, Bioanalytical
Systems, Inc., West Lafayete, IN) which was guarded by a precolumn cartridge filter (5
pm spheres; 30 x 4.6 mm). High-performance liquid chromatography mobile phase
consisted of 50 mM sodium phosphate buffer (pH 2.65 - 2.70), 30 mM citric acid, 0.1
mM disodium ethylenediamine tetraacetic acid (EDTA; Sigma), 0.30% sodium octyl

sulfate and 20-25% methanol. Depending on the characteristics of the column,

37

ingredients of the mobile phase were slightly adjusted to maintain separation of
compounds of interest and minimize total retention times for efficiency (Chapin et al.,
1986).

Amine content in the striatum was determined using an electrochemical detector
(LC4A, Bioanalytical Systems, Inc.) equipped with a TL-5 glassy carbon electrode set at a
potential of +0.75 V relative to a silver/silver chloride reference electrode. Since the
median eminence and nucleus accumbens samples contain lesser amounts of DOPAC and
dopamine than the striatum, supematants from these regions were analyzed on more
sensitive electrochemical detection systems which consisted of a single coulometric
electrode (Coulochem model #5100A; ESA, Bedford, MA) in series with dual electrode
analytical cells (models #5021, BSA). The potential of the conditioning cell electrode for
these systems was set at +0.40 V; analytical cell electrodes were set at +0.12 V and -0.40
V, relative to an internal silver reference electrode. The current signal from the second
electrode of the analytical cell was recorded by a Hewlett-Packard integrator (model
#3395; Hewlett-Packard, Avondale, PA). Use of coulometric electrodes increases the
sensitivity of the measurement of compounds of interest, and with the analytical
electrodes set up in an oxidation-reduction mode, only compounds oxidized at the first
electrode and reduced at the second electrode are detected. For example, the solvent
front, composed predominantly of ascorbic acid, is minimized in this way as ascorbic acid
does not back-reduce and is therefore not detected. The net effect of the oxidation-

reduction mode of detection is increased selectivity and reduced noise signal (Lindley et

al., 1990b).

38

The contents of DOPAC and dopamine in each sample were determined by
comparing peak heights reported by the integrator with those of the standards run on the
same day. The lower limit of sensitivity of this assay for DOPAC and dopamine was

approximately 2 - 5 pg per sample.

E. PROLACTIN ASSAYS

Following appropriate treatments rats were decapitated and trunk blood was
collected in heparinized glass vacutainer tubes (143 USP units/tube; Becton-Dickinson
Co., Franklin Lakes, NJ) maintained on ice. Whole blood was centrifuged (model DPR-
6000, International Equipment Co., Needham, MA) for 20 min at 2000 rpm and plasma
collected and stored at -20°C in glass vials containing 100 pl saturated sodium citrate

solution until assayed for prolactin.

I. Radioimmunoassay for prolactin

Serum prolactin concentrations were determined by double-antibody
radioimmunoassay using the procedure developed by Drs. Parlow and Raiti of the
National Hormone and Pituitary Program (NHPP), National Institute of Diabetes &
Digestive & Kidney Diseases (NIDDK; Rockville, MD). This procedure was optimized
and performed as follows.

Prior to assay, a range (0.03 - 2.00 ng/SOpl ) of standards were made using highly
purified rat prolactin (NIDDK rat PRL-RP-3; potency, 30 IU/mg as determined in pigeon
local crop assay by Nicoll) diluted in 0.1 M phosphate buffered saline (PBS) containing

2% bovine serum albumin (BSA; Sigma, RIA grade, fiaction V), aliquotted 50 pl per

39

microtube and stored at -20°C for no more than 3 months prior to use. Anti-rat prolactin
S-9 antisera (NIDDK) was reconstituted, diluted to 135,000 using 0.1 M PBS containing
0.05 M EDTA and 1:400 normal rat serum (Sigma), aliquotted 100 ml/bottle and stored
at -20°C until use. The solution of 0.1 M PBS/0.05 M EDTA/1:400 normal rat serum
was stored at -20°C in 1 ml aliquots. One hundred pg of rat prolactin for iodination (rat
PRL-L6; NIDDK) was dissolved in 125 pl of 0.01 M NaHCO3 (pH 8.6), aliquotted 5
pl/sterile siliconized 1.5 ml Eppendorf tube, and stored at -80°C for no more than 1 week
before use.

On assay day l, a set of rat prolactin standards, a tube of 0.1 M PBS/0.05 M
EDTA/1:400 NRS, a tube of rat serum pool, rat prolactin primary antiserum and samples
were thawed. All solutions were gently vortexed prior to use. Buffer (0.1 M PBS, 2%
BSA, pH 7.4) was added to each disposable glass (12 x 75 mm) cuvette such that after
subsequent addition of standard, sample or serum pool volume, the final volume was 500
pl. Cuvettes for the measurement of non-specific background and primary antibody
controls received 500 pl of buffer. Appropriate cuvettes were hand-pipetted 50 pl of
standards, three different volumes of unknown samples and 10, 20 and 50 pl of pooled
serum. The three dilutions for samples were established in previous experiments from
the 20 - 80% (linear) binding range for the expected concentration of prolactin. For
example, if prolactinemia in the range of 2 — 250 11ng was expected, 10, 20 and 50 pl
volumes of sample were utilized. If prolactinemia in the range of 0.1 — 70 ng/ml was
expected, 50, 100 and 200 pl volumes of sample were utilized. Repeating pipettor
systems (e.g. Digiflex automatic pipette) were not used in these assays because personal

observations revealed that: (1) rapid aspiration of solutions produced air bubbles in the

40

syringe, and (2) clots within plasma samples often plugged the syringe, contributing to
error and variability. All cuvettes were run in duplicate. Two hundred pl of 1:35,000
anti-rat prolactin S-9 antiserum was added to antibody control, standard, serum control
and sample cuvettes; alternatively, 200 pl of 0.1 M PBS/0.05 M EDTA/1:400 NRS was
added to non-specific background control cuvettes. Cuvettes were gently vortexed,
covered and incubated at 4°C for 18 - 24 h.

On assay day 2, prolactin was iodinated by the chloramine-T method (N iswender
et al., 1968) and prolactin tracer added to cuvettes as follows. The siliconized tube of rat
prolactin for iodination (4 pg rat PRL-I-6 per 5 pl aliquot) was thawed and utilized as the
reaction vessel. Accordingly, 20 pl of 0.5 M PBS (pH 7.4) was added and gently
vortexed. Ten pl of Na'zsl (l mCi; Amersham Corporation, Arlington Heights, IL) was
added to the reaction vessel, gently vortexed and allowed to incubate for 15 5. To
commence the oxidation of 125 I to prolactin, 10 pl chloramine-T (Sigma) solution (1.5 mg
chloramine-T/ml 0.05 M PBS, pH 7.4) was added to the reaction vessel, gently vortexed
and incubated for no more than 30 3. Dr. Parlow warned that only 4 - 5 pg of chloramine-
T should be utilized for every pg prolactin to be iodinated as oxidative damage to
prolactin could interfere with assay performance. A preliminary experiment testing
duration of chloramine-T reaction (at 4 pg chloramine-T/pg prolactin) showed evidence
of prolactin damage (shoulder on curve probably represents prolactin fragments) after 45
s incubation, and established 30 s incubation as optimal (figure 2.3). To quench the
oxidative reaction, 50111 of sodium metabisulfite (Sigma) solution (2.5 mg sodium meta-

bisulfite / ml 0.05 M PBS, pH 7.4) was added to the reaction vessel, gently vortexed and

41

 

Figl
‘3us
“’35
V1235,
11.5
30 0
mg:
for 1
10 L3

 

 

2.3e+6 —

2.0e+6 —

l.8e+6 —

1.5e+6 —

l.3e+6 —

l.0e+6 —

CPM/IOul

7.5e+5 ~

 

5.0e+5 —

2.56+5 _

 

 

 

FRACTION

Figure 2.3 Effects of duration of chloramine-T oxidative reaction on iodinated prolactin
curve morphology. In a siliconized tube, 4 pg rat prolactin for iodination in 5 pl NaHCO3
was combined with 20 p1 of PBS. Ten pl of Naml (1 mCi) was added to the reaction
vessel, gently vortexed and allowed to incubate for 15 5. Ten pl chloramine-T solution
(1.5 mg/ml PBS) was added to the reaction vessel, gently vortexed and incubated for 15,
30 or 45 5. To quench the oxidative reaction, 50 pl of sodium metabisulfite solution (2.5
mg/ml PBS) was added to the reaction vessel, gently vortexed and allowed to incubate
for 15 s. The reaction mixture was fractionated on a PD-lO Sephadex G-25M column and
10 pl of each fraction was quantified in an automated gamma counter.

42

allowed to incubate for 15 s.

The reaction mixture was fractionated on a PD-lO Sephadex G-25M column
(Pharmacia Biotech, Uppsala, Sweden) previously rinsed with 5 ml of 0.05 M PBS (pH
7.4). Phosphate buffered saline was added dropwise to keep the column filter moist and
facilitate the elution of the reaction mixture. A preliminary experiment utilized the Lowry
protein assay to demonstrate that rat prolactin (rat PRL-I-6) elutes from the PD-10
Sephadex G—25M column in the 6th or 7th serially collected fraction (10 drops or 500
pl/fraction) (figure 2.4). To be conservative, the first 12 fractions were collected into test
tubes containing 1 ml of 0.1 M PBS and 2% BSA (pH 7.4).

The collected fractions were vortexed and 10 pl from each test tube was counted
for 1 min in an Micromedic plus series automatic garmna counter (model 4/600, ICN
Biomedicals, Inc., Costa Mesa, CA) coupled to an IBM PC running bundled AGC
operating software (version 2.07). The fraction containing maximal cpm (usually the 7th
fraction) was diluted with 0.1 M PBS/2% BSA (pH 7.4) to 15,000 cme 100 pl solution.
After appropriate radioactivity of the iodinated prolactin tracer solution was established,
each cuvette received 100 pl of tracer, was vortexed, covered and allowed to incubate for
18 - 24hat 4°C.

On day 3, goat anti-rabbit gamma globulin antiserum (Accurate Chemical &
Scientific Corporation, Westbury, NY) was reconstituted and diluted 1:200 with 0.1 M
PBS (pH 7.4). All cuvettes except total count controls received 200 pl of secondary
antiserum, were gently vortexed, covered and incubated for 18 - 24 h at 4°C.

On day 4, antibody-bound radiolabeled prolactin was precipitated and quantified.

All cuvettes except total count controls received 2 ml of cold 0.1 M PBS (pH 7.4)

43

10 e

PROTEIN (ug/lOOul)
:

 

1 1 1 1 1 1 1 1 1 1 1
0 2 4 6 810121416182022

FRACTION

Figure 2.4 Characterization of PD-10 Sephadex G-25M column for fractionalization of
rat prolactin for iodination via Lowry protein assay. A column was rinsed with 5 m1 PBS
prior to use. The solution containing 100 pg rat prolactin dissolved in 10 ml PBS was
added to the column and separated into 500 pl fractions. The collected fractions were
vortexed and 100 pl of each fraction was analde for protein content via Lowry protein
assay relative to standards run on the same day.

44

containing 5% polyethylene glycol (Sigma; 8,000 mw) and were centrifuged (3,000 rpm;
Damon/IEC model PR-6000; Needham Heights, MA) for 20 min at 4°C. After
centrifugation, supematants were decanted and cuvettes were air-dried. Dry cuvettes
were loaded into racks and quantified in an Micromedic plus series automatic gamma
counter coupled to an IBM PC running bundled AGC operating software.

The procedure as described routinely yielded 40 - 50% binding and an R value of
0.99. Using a 100 pl aliquot of plasma, the lower limit of sensitivity was 0.1 ng/cuvette,
and the intra— and interassay coefficients of variance were 4.6% and 6.7%, respectively.
The primary antiserum for rat prolactin showed no significant cross-reactivity with rat
growth hormone, thyroid-stimulating hormone, follicle-stimulating hormone, or luteiniz-

ing hormone as reported by the supplier.

2. Bioassay for prolactin

Antiserum to prolactin interferes with plasma prolactin measurement by double
antibody radioimmunoassay (Hattori et al., 1994). Consequently, plasma prolactin was
measured indirectly by rat Nb2 node lymphoma cell bioassay in experiments in which
prolactin antiserum was administered. The rat Nb2 lymphoma cell bioassay for prolactin
is based on the observation that rat Nb2 node cell lines replicate (approximately 16-fold
over 72 h) in a dose-dependent manner in response to prolactin (Tanaka et al., 1980; Gout
et al., 1980; Noble et al., 1980). The assay is sensitive (lower limit 1.22 ng/ml), specific
(clonal expansion is blocked by prolactin antiserum, and unaffected by rat growth
hormone; Tanaka et al., 1980) and reproducible, yielding results comparable to prolactin

radioimmunoassay (Lawson et al., 1982).

45

Frozen plasma samples were packed in a cooler of dry ice and shipped to Dr.
David M. Lawson (Wayne State University School of Medicine, Detroit, MI) who
performed conventional rat Nb2 node lymphoma cell bioassays for prolactin as follows.
The Nb2 node cell line was originally derived from Drs. Noble, Beer and Gout
(University of British Columbia, Vancouver, BC, Canada) and stored in liquid nitrogen.
Prior to the assay, viable cells were recovered by thawing in a 37°C bath of filter-
sterilized Fischer’s medium containing 10% fetal calf serum, 10% horse serum, 1 x 10'4
M 2-mercaptoethanol, 50 IU/ml penicillin and 50 pg/ml streptomycin (stock medium).
The cells were then centrifuged (300 x g, 10 min), the pellet gently resuspended in stock
medium and dispensed into tissue culture flasks which were supplied with a 95% air - 5%
CO2 gas mixture. The medium was changed after 24 h and the flasks were incubated at
37°C for 72 h. Cell numbers were monitored and when they reached 1x 106 cells/ml, they
were diluted lO-fold with stock medium. Cells were used for assay when the doubling
time was in the range of 18 - 20 h. Twenty four h before assay the cells were repeatedly
centrifuged and resuspended in stock media containing 10%, 1% and 0% fetal calf serum.
After washing the cells of fetal calf serum, the cells were diluted to 1 x 105 cells/ml
aliquot in Fischer’s medium and one aliquot was added to each well of a 24-well tissue
culture plate.

Standard rat prolactin (NIDDK-RP-3; 11 IU/mg) was diluted with Fischer’s
medium to give a final concentration of 0.01 - 40.00 ng/50 pl. Each sample was diluted
with Fischer’s medium to three concentrations and brought to a final volume of 50 pl.
Fifiy pl of standards and samples were added to duplicate wells of the tissue culture

plates. The plates were incubated in a humidified incubator at 37°C for 72 h with

46

constant exposure to 95% air - 5% CO2 gas mixture. The entire well content was
removed, diluted lO-fold with particle-free electrolyte solution (Hematall; Fisher
Scientific, Itasca, IL) and counted in a model B Coulter Counter (Fisher Scientific) using
a 70 pm aperture and a 100 pl mercury manometer. Cell counts from the Coulter Counter
were entered into a Micromedic MAAC microprocessor and analyzed using a slight

modification of a point to point linear RIA program (Lawson et al., 1982).

F. IMMUNOHISTOCHEMISTRY
1. Tissue preparation

After appropriate treatments, rats were anesthetized with Equithesin (4 ml/kg, i.p.)
and transcardially perfused with cold 0.9% saline (~20 ml/rat) followed by 0.1 M PBS
(pH 7.4) containing 4% paraformaldehyde (Sigma; 150 - 180 ml/rat) delivered via
peristaltic pump (rabbit model; Rainin Instrument Co., Wobum, MA) at a flow rate of 10
ml/min. Brains were removed and post-fixed in 20 ml of 0.1 M PBS containing 4%
paraformaldehyde (pH 7.4) for 18 - 24 h at 4°C. Brains were cryoprotected in 20 ml
aliquots of 0.1 M PBS (pH 7.4) containing 20% sucrose changed daily until brains
equilibrate and sink. Brains were sectioned at -15°C using an IEC Minotome cryostat
(International Equipment Co., Needham Heights, MA) into 30 pm frontal sections
through the entire rostrocaudal extent of the arcuate nucleus, beginning approximately 2.0
mm posterior to bregma (Paxinos and Watson, 1986). Sections were collected as 8 sets
of serial sections in a multiwell tray of cryoprotectant (30% (w/v) sucrose, 1% (w/v)
polyvinylpyrrolidone, 30% (v/v) ethylene glycol in 0.1 M PBS, pH 7.2; Watson et al.,

1986) and stored at -20°C until required for immunohistochemistry.

47

Tissue collection and storage in this fashion facilitates tissue handling and
minimizes variation in experiments. Considering that each well contains sections no
closer than 240 pm, correction factors for error due to duplicate quantification of the
same cells in adjacent sections are unnecessary. This method also provides two replicates

of rostral, middle and caudal regions of the arcuate nucleus (figure 2.5) per well of tissue.

2. F RA and tyrosine hydroxylase detection

Two non-adjacent wells of tissue per brain (i.e. 4 replicate sections per region)
were thawed and washed of cryoprotectant in 4 rinses (15 min/rinse) of 0.05 M Tris
(GibcoBRL, Grand Island, NY)-buffered saline, pH 7.6 (TBS). Unless otherwise stated,
tissue was continually gyrated (100 rpm) during latent periods between solution changes
on a gyrotory shaker table (model G2; New Brunswick Scientific Co., Edison, NJ) at
room temperature. Tissues were treated with 3% hydrogen peroxide (Sigma) in TBS for
10 min, and rinsed (4 x 10 min, TBS) to minimize non-specific background due to
endogenous peroxidases. Tissues were treated with 1% sodium borohydride (w/v;
Sigma) in TBS for 20 min, and rinsed (6 x 10 min, TBS) to neutralize non-specific
binding due to reactive aldehydes fiom fixative solutions. Tissues were treated with 3%
normal donkey serum (Sigma) in TBS containing 0.3% (v/v) of the detergent Triton-
X100 (TX; Research Products International Corp., Elk Grove Village, IL) for 30 min to
block non-specific binding with secondary antiserum. Tissue was treated with 1% (v/v)

avidin solution (Avidin-Biotin Blocking kit; Vector Laboratories, Burlingame, CA) in

48

ROSTRAL ARC

A

 

Figure 2.5 Representative rostrocaudal regions of the ARC highlighted in ventral
hypothalamic frontal sections. Distances posterior to bregma for these regions were 2.3 -
2.8 mm for rostral ARC, 2.8 - 3.3 for middle ARC and 3.3 — 3.8 mm for caudal ARC
(Paxinos and Watson, 1997). 3V, third ventricle; ARC, arcuate nucleus; DMN,
dorsomedial hypothalamic nucleus; f, fornix; LH, lateral hypothalamus; opt, optic tract;
VMN, ventromedial hypothalamic nucleus. '

49

TBS-TX for 20 min, and rinsed (3 x 10 min, TBS-TX) to prevent non-specific binding to
endogenous avidin. Tissues were treated with 1% (v/v) biotin solution (Avidin-Biotin
Blocking kit; Vector Laboratories) in TBS-TX for 20 min, and rinsed (3 x 10 min, TBS-
TX) to decrease non-specific binding due to endogenous biotin.

Anti-FRA antiserum (OA-11-824, Genosys Biotechnologies, The Woodlands,
TX) was generated in sheep against a synthetic peptide derived from a conserved region
of the c-fos gene common to all F RA genes (amino acid sequence: M-F-S-G-F-N-A-D-Y-
E-A-S-S-S-R-C), including F OS, FRA], FRA2, and FOSB (Hesketh, 1995). FRA
antiserum specificity was determined by the manufacturer and another research group
(Curran et al., 1985). Preliminary dual immunohistochemical titration studies determined
that the optimal dilution for F RA antiserum was 1:5,000. Accordingly, anti-FRA
antiserum was diluted in TBS-TX containing 1.5% normal donkey serum, added to tissue
and incubated for 48 h on a gyrotory shaker table at 4°C.

After 4 rinses of TBS-TX (15 min/rinse), tissue—bound FRA antiserum was
detected with biotinylated donkey anti-sheep IgG antiserum (Accurate Chemical &
Scientific Corp.) diluted 1:500 in TBS-TX and incubated for 1.5 h. Excess secondary
antiserum was removed with 6 x 10 min rinses in TBS-TX and remaining signal fi'om the
antibody complex was amplified by the avidin-biotin complex system (Vector
Laboratories; Hsu et al., 1981). Vector ABC Elite kit reagents A (avidin) and B
(biotinylated horse radish peroxidase) were diluted 1:200 (v/v) in TBS-TX and allowed to
stand for 30min before incubation with tissue for 1.5 h. Unbound avidin-biotin complex
(was washed from tissue via 3 TBS-TX rinses (10 min/rinse), followed by 3 rinses of

similar duration in sodium acetate (0.175 M CH3COONa, pH 8.2; JT Baker).

50

The chromagen utilized for F RA localization was nickel-intensified 3,3’-
diarninobenzidine tetrahydrochloride (DAB; Sigma) freshly made using the Hsu recipe
(Hsu et al., 1982). Tissue was incubated in 0.175 M sodium acetate containing 0.05%
(w/v) DAB, 0.01% hydrogen peroxide, and 2.5% nickel sulfate (w/v; Sigma) for 3 - 5 min
as determined by rate of color development. After 3 rinses of sodium acetate and 3 rinses
of TBS-TX (10 min/rinse), tissues were treated with 3% normal horse serum (Sigma) in
TBS-TX for 30 min to block non-specific binding with secondary antiserum.

Monoclonal (IgG. subclass) anti-tyrosine hydroxylase antiserum (#22941, Incstar
Corp., Stillwater, MN) was generated in mouse against a fragment of purified tyrosine
hydroxylase isolated from rat PC12 cells. The antiserum is believed to have wide species
cross-reactivity because it recognizes an epitope derived from the midportion of the
tyrosine hydroxylase molecule where extensive species homology exists. The
manufacturer reports that this tyrosine hydroxylase antiserum did not cross react with
dihydropterdine reductase, dopamine-B-hydroxylase, phenylethanolamine-N-methyl-
transferase, phenylalanine hydroxylase or tryptophan hydroxylase as determined by
western analysis. Preliminary dual immunohistochemical titration studies determined that
the optimal dilution for tyrosine hydroxylase antisertun was 1:1,000. Accordingly, anti-
tyrosine hydroxylase antiserum was diluted 1:1,000 in TBS-TX containing 1.5% normal
horse serum, added to tissue and incubated for 48 h on a gyrotory shaker table at 4°C.

After 4 rinses of TBS-TX (15 min/rinse), tissue-bound tyrosine hydroxylase
antiserum was detected with biotinylated horse anti-mouse IgG antiserum (BA-2000;
Vector Laboratories) diluted 1:200 in TBS-TX and incubated for l h. Excess secondary

antiserum was removed with 6 x 10 min rinses in TBS-TX. The chromagen utilized for

51

tyrosine hydroxylase localization was rhodamine-tagged avidin D (A-2002; Vector
Laboratories). Tissues were incubated rhodamine-tagged avidin diluted 1:500 in TBS-TX
for l h in darkness and rinsed in TBS (6 x 10 min/rinse). This double
immunohistochemistry procedure routinely allowed visualization of tyrosine hydroxylase-
immunoreactive (red-orange fluorescent) neurons which contained FRA-immunoreactive

(black) nuclei (figure 2.6).

3. F RA and neurotensin detection

Tissue was thawed, rinsed of cryoprotectant, and non-specific background was
minimized via treatments with 3% hydrogen peroxide, 1% sodium borohydride, 3%
normal donkey serum, 1% (v/v) avidin solution, and 1% (v/v) biotin solution, as
described for FRA and tyrosine hydroxylase detection. Anti-FRA antiserum (OA-11-824,
Genosys Biotechnologies) was diluted in TBS-TX containing 1.5% normal donkey
serum, added to tissue and incubated for 48 h on a gyrotory shaker table at 4°C. Tissue-
bound FRA antiserum was detected with biotinylated donkey anti-sheep IgG antiserum
(Accurate Chemical & Scientific Corp.) diluted 1:500 in TBS-TX and incubated for 1.5 h,
and remaining signal from the antibody complex was amplified by Elite avidin-biotin
complex kit (Vector Laboratories) diluted 1:200 (v/v) in TBS-TX and incubated with
tissue for 1.5 h. The chromagen utilized for FRA localization was nickel-intensified
DAB freshly made using the Hsu recipe (Hsu et al., 1982) and incubated with tissue 3 - 5
min as determined by rate of color development. After 3 rinses of sodium acetate and 3

rinses of TBS-TX, tissues were treated with 3% normal goat serum (Sigma) in TBS-TX

52

5()um

 

Figure 2.6 Digitized photomicrographs of tissue immunohistochemically processed for
the detection of TH and FRA. Fluorescent TH-IR neurons under dark field optics (left
panel) and the same image under bright field optics (right panel) showing blackened
FRA-IR nuclei colocalized to some, but not all, of these neurons. Arrowheads represent
TH-IR neurons with FRA-IR nuclei; i.e. double-labeled neurons.

53

for 30 min to block non-specific binding with secondary antiserum.

Polyclonal anti-neurotensin antiserum (#IHC 7351, Lot #971456, Peninsula
Laboratories, Belmont, CA) was generated in rabbit against synthetic neurotensin (amino
acid sequence: E-L-Y-E-N-K-P-R-R-P-Y-I-L; Carraway and Leeman, 1975). The
manufacturer reports that this neurotensin antiserum exhibited 100% cross reactivity with
synthetic neurotensin, 58% cross reactivity with guinea pig neurotensin (amino acid
sequence: E-L-Y-E-N-K-S—R-R-P-Y-I-L; Rostene and Alexander, 1997), yet did not cross
react with neurotensin 8-13, porcine neuromedin N, kinetensin, substance P, bombesin or
eledoisin by western analysis. One hundred percent cross reactivity is expected for rat as
this species has identical amino acid identity with the synthetic antigen. Preliminary dual
immunohistochemical titration studies determined that the optimal dilution for
neurotensin antiserum was 1:8,000. Antiserum was diluted accordingly in TBS-TX
containing 1.5% normal goat serum, added to tissue and incubated for 48 h on a gyrotory
shaker table at 4°C.

After 6 rinses of TBS-TX, tissue-bound neurotensin antiserum was detected with
biotin-SP-conjugated AffiniPure goat anti-rabbit IgG, F(ab’)2 fragment specific antiserum
(#111-065-006; Lot #38262; Jackson ImmunoResearch Laboratories, West Grove, PA)
diluted 1:500 in TBS-TX and incubated for 2 h. Excess secondary antiserum was
removed with 6 x10 min rinses in TBS-TX, and remaining signal from the antibody
complex was amplified by the avidin-biotin complex system. Vector ABC Elite kit
reagents were diluted 1:200 (v/v) in TBS-TX and incubated with tissue for 2 h. Unbound
avidin-biotin complex was rinsed from tissue via 3 TBS-TX rinses (10 min/rinse),

followed by 3 rinses of similar duration in TBS-TX (pH 7.2).

54

The chromagen utilized for neurotensin localization was DAB without nickel
freshly made according to the Hsu recipe (Hsu et al., 1982). Tissue was incubated in
TBS-TX (pH 7.2) containing 0.05% (w/v) DAB, and 0.01% hydrogen peroxide for 7 - 10
min as determined by rate of color development. Tissues were rinsed (10 min/rinse) 3
times with TBS-TX (pH 7.2) and 3 times with TBS-TX prior to storage or mounting.
This double immunohistochemistry procedure routinely allowed visualization of
neurotensin-immunoreactive (brown) neurons which contained FRA-immunoreactive

(black) nuclei (figure 2.7).

4. Neurotensin and tyrosine hydroxylase detection

Tissue was thawed, rinsed of cryoprotectant, and non-specific background was
minimized via treatments with 3% hydrogen peroxide, 1% sodium borohydride, 3%
normal goat serum, 1% (v/v) avidin solution, and 1% (v/v) biotin solution, as described
above. Anti-neurotensin antiserum was diluted 1:8,000 in TBS-TX containing 1.5%
normal goat serum, added to tissue and incubated for 48 h on a gyrotory shaker table at
4°C. Tissue-bound neurotensin antiserum was detected with biotin-SP-conjugated
AffiniPure goat anti-rabbit IgG, F(ab’)2 fragment specific antisertun (Jackson
ImmunoResearch Laboratories) diluted 1:500 in TBS-TX and incubated for 2 h, and
remaining signal from the antibody complex was amplified by Elite avidin-biotin
complex kit (Vector Laboratories) diluted 1:200 (v/v) in TBS-TX and incubated with
tissue for 2 h. The chromagen utilized for neurotensin localization was nickel-intensified
DAB freshly made and incubated with tissue 7 - 10 min as determined by rate of color

development. After rinsing, tissues were treated with 3% normal horse serum in TBS-TX

55

 

 

rig
PT: ,
10;,
FR.

 

Figure 2.7 Digitized bright field photomicrograph of tissue immunohistochemically
processed for the detection of NT and FRA. Clear or blackened FRA-IR nuclei were
localized within brown NT-IR perikarya. Arrowheads represent NT-IR neurons with
F RA-IR nuclei; i.e. double-labeled neurons.

for 30 min to block non-specific binding with secondary antiserum.

Anti-tyrosine hydroxylase antiserum (Incstar Corp.) was diluted 121,000 in TBS-
TX containing 1.5% normal horse sertun, added to tissue and incubated for 48 h on a
gyrotory shaker table at 4°C. Tissue-bound tyrosine hydroxylase antiserum was detected
with biotinylated horse anti-mouse IgG antiserum (Vector Laboratories) diluted 1:200 in
TBS-TX and incubated for 1 h. The chromagen utilized for tyrosine hydroxylase
localization was rhodamine-tagged avidin D (Vector Laboratories) diluted 1:500 in TBS-
TX and incubated with tissues for 1 h in darkness. This double immunohistochemistry
procedure allowed visualization of tyrosine hydroxylase- (red-orange fluorescent) and
neurotensin- (black) immunoreactive perikarya in the same section by alternating between

dark and bright field optics, respectively (figure 2.8).

5. Slide preparation and visualization

Following immunohistochemistry, sections were stored in TBS at 4°C in darkness
until mounted. Glass slides were washed, rinsed 5 times with distilled water and air-
dried. Clean slides were twice dipped in warm (60°C) subbing solution containing
0.188% gelatin (300 bloom swine; Sigma) 0.0188% chromium potassium sulfate (Sigma)
in double-distilled water. Sections were floated in TBS onto subbed slides and air-dried.
Dry slides were loaded into racks and dehydrated via graded series of filtered ethanol
baths (70%, 90% and 100%; 10 min/rinse). Tissues were delipidated in a series of xylene
baths (2 x 15 min/rinse) and slides coverslipped using DPX mountant (Fluka, St. Louis,

MO).

57

 

Figure 2.8 Digitized photomicrographs of tissue immunohistochemically processed for
the detection of NT and TH. Blackened NT-IR perikarya visualized under bright field
optics (left panel) and the same image under dark field optics (right panel) showing
localization of fluorescent TH-IR.

58

Tissues processed by immunohistochemistry using DAB or DAB-nickel as the
chromagen were viewed by standard bright field microscopy on a Leitz Laborlux
microscope (model S, Wetzlar, Germany). When rhodamine was the chromagen, dark
field fluoromicroscopy was employed on the same microscope (equipped with mercury
lamp and model 103 housing, Leitz). Bright field photomicrographs were captured on
Motophoto (color) 400 film using an interfaced photosystem (model Wild MPSZ8/32,
Leitz) with automated exposure. Since dark field images were routinely over-exposed
using the automated photosystem, these photomicrographs were manually exposed for 1-
15 8. Images were digitized using a scanner (model 5100C, Hewlett Packard) and
PrecisionScan (version 2.0, Hewlett Packard) software/driver, and labeled using
PhotoShop software (version 5.0, Adobe, San Jose, CA). Magnifications are indicated in

the legends of the appropriate figures.

G. QUANTIFICATION & STATISTICAL ANALYSES

Group averages from prolactin radioimmunoassays were based upon individual
animal averages derived from 2 replicates at each of 3 dilutions, and only values which
were extrapolated from the linear portion of the standard curve (20 — 80% binding of
radiolabeled prolactin) were included.

In all immunohistochemistry experiments, 2 - 6 sections per region (rostral,
middle or caudal ARC figure 2.5) per animal were bilaterally quantified at 400x
magnification for total numbers of single- and double-labeled cells in each ARC
subdivision: dorsomedial-magnocellular, dorsomedial-parvocellular and ventrolateral

(figure 1.4). Consequently, the contribution of one animal was based on no fewer than 6

59

sections, and as many as 18 sections through the arcuate nucleus. Individual animal
means were generated and compiled to produce a group average. Since raw counts were
not normally distributed, they were systematically transformed prior to statistical analyses
using a square root function.

Statistical analyses of the data were conducted using SigmaStat software (version
2.0, Jandel Scientific, San Rafael, CA). Statistical analyses of two groups with normal
distribution and equal variance were conducted using Student’s t-test. If normality or
equal variance failed, the Mann-Whitney rank sum test was utilized. Statistical analyses
of greater than two groups with normal distributions and equal variances were conducted
using one way analysis of variance followed by Dunnett’s test for multiple comparisons
versus a control group. If normality or equal variance tests failed, statistical analyses
were conducted using Kruskal-Wallis one way analysis of ranks followed by Dunn’s test
for multiple comparisons versus a control group (Steel and Torrie, 1960). Differences
were considered significant if the probability of error was less than 5% and the desired
power of tests was set as 0.80. Group averages i one standard error of the mean were

utilized for representation in all graphs and tables.

60

3. REGULATION OF TIDA NEURONS BY PROLACTIN

A. INTRODUCTION

The mammalian brain contains several different dopaminergic neuronal systems.
The largest and most studied is the mesotelencephalic system, comprised of NSDA and
MLDA neurons. NSDA neurons have perikarya in the substantia nigra pars compacta and
axons that terminate in the striatum. These neurons regulate sensorimotor integration and
motor movement. MLDA neurons have perikarya in the ventral tegmental area and axons
that terminate in limbic structures including the nucleus accumbens, olfactory tubercles,
septum, amygdala, and various cortical regions. These neurons regulate motivational
behavior. TIDA neurons have perikarya in the arcuate nucleus and axons that terminate in
the median eminence. Dopamine released in the median eminence from TIDA neurons
travels in the hypophysial portal blood to the anterior pituitary where it tonically inhibits
prolactin secretion. Much less is known about the TIDA neuronal system in comparison
to the “classical” mesotelencephalic system, hence the primary focus of this chapter will
be upon the regulation of TIDA neurons by prolactin, incorporating the NSDA and
MLDA neuronal systems for comparison.

TIDA neurons lack dopamine autoreceptors characteristic of NSDA and MLDA
systems (Demarest and Moore, 1979), and instead, are regulated in part by the feedback
effects of prolactin (for review see Moore and Lookingland, 1995). Feedback activation
of TIDA neurons by prolactin is comprised of interrelated tonic and inductive
components (Demarest et al., 1984; 1986). The tonic component determines the basal
rate of prolactin secretion by regulating short-term changes in TIDA neuronal activity in
response to acute changes in circulating prolactin concentrations. This component is

61

unique to females and sets the basal level of activity of TIDA neurons with a latency of 2-
4 h. The inductive component involves a change in the capacity of TIDA neurons to
respond to more prolonged periods of hyperprolactinemia (Moore et al., 1987a) and is
observed in both males and females. The inductive component stimulates TIDA neuronal
activity to levels 2-3 times basal (Demarest et al., 1985b; Selmanoff, 1985) 12 h
following systemic (Hokfelt and Fuxe, 1972; Selmanoff, 1981; Demarest et al., 1984;
1986) or central (Annunziato and Moore, 1978) administration of exogenous prolactin, or
afier experimental manipulations which result in endogenous hyperprolactinemia (Moore
et al., 1985).

Much of what is known regarding prolactin feedback regulation of TIDA neurons
is based upon pharmacological studies employing non-selective dopamine receptor
agonists (e.g. bromocriptine; Demarest et al., 1985b) and antagonists (e.g. haloperidol;
Demarest and Moore, 1979) to alter endogenous prolactin secretion. While it is generally
accepted that these ligands bind to D2 receptors on lactotrophs and thereby act indirectly
via changes in circulating prolactin to regulate the activity of TIDA neurons (Moore and
Lookingland, 1995), these neurons are also acutely regulated by a D2 receptor-mediated
mechanism which acts independently of prolactin (Eaton et al., 1993b; Durham et al.,
1997; 1998). Moreover, many early neurochemical estimates of TIDA neuronal activity
employed drugs (e.g. a-methyltyrosine, NSD 1015) that by design inhibit dopamine
biosynthetic enzymes (Moore, 1987b). Since newly synthesized dopamine is
preferentially released from TIDA neurons (Moore and Lookingland, 1995), these
inhibitors reduce dopamine release, and thereby remove dopamine inhibition of prolactin

secretion (Reyrnond and Porter, 1982). Because these drugs were administered to all

62

experimental groups (including controls), a period of hyperprolactinemia preceding and
during the time of sampling may have confounded the results, especially those concerning
the early time course effects of prolactin on TIDA neurons.

The overall aims of the experiments described in this study were to characterize
the temporal aspects of the delayed inductive activation of TIDA neurons by endogenous
hyperprolactinemia and confirm that prolactin mediates the stimulatory effects of
haloperidol on these neurons. The underlying hypothesis to be tested is that delayed
activation of TIDA neurons by prolactin is dependent upon a period of sustained
hyperprolactinemia for both the initiation and maintenance of this response. If the
delayed inductive activation of TIDA neurons by prolactin is dependent upon a period of
sustained hyperprolactinemia for both initiation and maintenance of the response, then
exposure to 3 or 6 h periods of hyperprolactinemia followed by 9 or 6 h periods,
respectively, of hypoprolactinemia will prevent the delayed inductive activation of TIDA
neurons. Dopaminergic neuronal activity will be estimated in these studies by measuring
the concentrations of the dopamine metabolite DOPAC in terminal regions of these
neurons which requires no pharmacologic intervention. Specific prolactin antiserum
(PRL-AB) will be utilized to immunoneutralize circulating prolactin thereby avoiding the

use of dopaminergic agonists.

63

B. RESULTS

Haloperidol is a D2 dopamine receptor antagonist that stimulates prolactin
secretion via blockade of the tonic inhibitory action of dopamine on pituitary lactotrophs.
As shown in figure 3.1, systemic administration of haloperidol (1 mg/kg; sc) increased
plasma prolactin concentrations in male rats to peak levels by l h, and maintained a state
of hyperprolactinemia for at least 12 h. Haloperidol also increased median eminence
DOPAC concentrations by 6 and 12 h post-injection, but had no effect on dopamine
concentrations in this region. Systemic administration of haloperidol induced rapid (by 1
h) and prolonged (through 12 h) increases in DOPAC concentrations in both the striatum
(figure 3.2) and nucleus accumbens (figure 3.3) and acute decreases in dopamine
concentrations in both the striatum (for 1-3 h; figure 3.2) and nucleus accumbens (for 1-6
h; figure 3.3).

The time course effects of PRL-AB on concentrations of plasma prolactin in male
rats pretreated with haloperidol were investigated to determine the capacity of PRL-AB to
immunoneutralize endogenous hyperprolactinemia (figure 3.4, upper panel). DOPAC and
dopamine concentrations in the median eminence of these rats were measured
concurrently (figure 3.4, lower panel). As expected, plasma prolactin concentrations in
normal rabbit serum-treated rats were elevated 1 h afier coadministration of haloperidol.
In haloperidol-treated rats, a single injection of undiluted PRL-AB (200 til/rat; iv) caused
a prompt (by 1 h) decrease in plasma prolactin concentrations to non-detectable levels
which persisted until at least 3 h afier administration. By 6 h however, male rats
coadministered haloperidol and PRL-AB had elevated plasma prolactin concentrations

compared with zero time vehicle-treated controls. PRL-AB had no effect on median

64

 

 

 

 

40 -
E
H) 30 -
5
a:
E 20' * a:
U
E
o 10-
Dd
D4
0 # .
0 1 3 6 12
*
20 * 2|:
’5“ 15
E)
E/ 120
g 10 ' g...
g :1 60
Q 5 r G
30
0 l 3 6 12
0 r 4 . l
0 1 3 6 12
HOURS AFTER HALOPERIDOL

Figure 3.1 Time course effects of haloperidol on concentrations of prolactin in plasma
(UPPER PANEL), and DOPAC and DA in the median eminence (LOWER PANEL) of
male rats. Rats were injected with haloperidol (1.0 mg/kg; sc; solid symbols and
columns) and decapitated either 1, 3, 6 or 12 h later. Zero time controls were injected
with 0.3% tartaric acid (1.0 ml/kg; sc; open symbols and column) and decapitated 12 h
later. Symbols represent means and vertical lines 1 SEM of 7-8 determinations of
concentrations of prolactin in plasma (nyml) and DOPAC (ngmg protein) in the median
eminence. Columns represent the means and vertical lines 1 SEM of 7—8 determinations
of DA concentrations (rig/mg protein) in the median eminence. *, Values that are
significantly different (p<0.05) from zero time controls.

65

 

 

 

 

 

 

 

* *
a 80 »
E :1:
E) 60 —
L)
< 40 -
a.
O
Q 20 -
0 n 1 I l
0 1 3 6 12
160 -
_L
A 120 -
on
E
g 8. -
a
40 -
0
0 1 3 6 12
HOURS AFTER HALOPERIDOL

Figure 3.2 Time course effects of haloperidol on concentrations of DOPAC (UPPER
PANEL), and DA (LOWER PANEL) in the striatum of male rats. Rats were injected
with haloperidol (1.0 mg/kg; so; solid symbols and columns) and decapitated either 1, 3,
6 or 12 h later. Zero time controls were injected with 0.3% tartaric acid (1.0 ml/kg; sc;
open symbol and column) and decapitated 12 h later. Symbols represent means and
vertical lines 1 SEM of 7-8 determinations of concentrations (rig/mg protein) of DOPAC
and DA in the striatum. *, Values that are significantly difi‘erent (p<0.05) from zero time

controls.

66

100 -

 

 

 

 

 

 

* *
A 80 -
00 :1:
E.
5 6O - *
L)
< 40 -
A.
O
Q 20 _
O 1 1 I 1 1
0 1 3 6 12
160 1
Q;_
120 — * * *
373
E
E3 80-
<
D
40 ~
0
0 1 3 6 12
HOURS AFTER HALOPERIDOL

Figure 3.3 Time course effects of haloperidol on concentrations of DOPAC (UPPER

PANEL), and DA (LOWER PANEL) in the nucleus accumbens of male rats. Rats were
injected with haloperidol (1.0 mg/kg; sc; solid symbols and columns) and decapitated
either 1, 3, 6 or 12 h later. Zero time controls were injected with 0.3% tartaric acid (1.0
kag; sc; open symbol and column) and decapitated 12 h later. Symbols represent means
and vertical lines 1 SEM of 7-8 determinations of concentrations (ng/mg protein) of
DOPAC and DA in the nucleus accumbens. *, values that are significantly different

(p<0.05) from zero time controls.

35 r
30 . 0 Vehicle

0 Haloperidol
25 -

20r

10*

PROLACTIN (ng/ml)

 

 

120

DOPAMINE
‘5‘ 8 8

 

 

 

0

 

12'

DOPAC (rig/mg)

10*

 

 

 

0 l 3 6

HOURS AFTER PRL-AB

Figure 3.4 Time course effects of PRL-AB on concentrations of plasma prolactin
(UPPER PANEL), and DOPAC and DA in the median eminence (LOWER PANEL) of
male rats. Rats were injected with haloperidol (1.0 mg/kg; sc; solid symbols and
columns) and PRL-AB (200ul/rat; iv) and decapitated either 1, 3, or 6 h later. Zero time
controls were obtained from rats injected with haloperidol and normal rabbit serum
(NRS; 200ul/rat; iv) 1 h prior to decapitation. Vehicle controls (open symbol and column)
were injected with 0.3% tartaric acid (1 ml/kg; sc) and NRS 1 h prior to decapitation.
Symbols and columns represent means and vertical lines 1 SEM of 7—8 determinations of
concentrations of prolactin in plasma (ng/ml; UPPER PANEL), DOPAC (ng/mg protein;
LOWER PANEL) and DA (ng/mg protein; LOWER PANEL INSET) in the median
eminence. *, Values that are significantly different (p<0.05) from zero time controls.

68

eminence DOPAC concentrations in male rats through 3 h, but was decreased at 6 h.
PRL-AB had no effect on DA concentrations in this region at any time after its
administration. These studies established that a single injection of PRL-AB (200 til/rat;
iv) immunoneutralized haloperidol-induced hyperprolactinemia for at least 3 h.

Recall from figures 3.1-3.3, haloperidol increased concentrations of DOPAC 12 h
afier injection, but had no effect on dopamine concentrations in the median eminence,
striatum and nucleus accumbens. PRL-AB blocked the delayed (12 h) stimulatory effect
of haloperidol on DOPAC concentrations in the median eminence and had no effect on
dopamine concentrations in this region (figure 3.5). In contrast, PRL-AB failed to block
the stimulatory effects of haloperidol on DOPAC concentrations in the striatum and
nucleus accumbens, and dopamine concentrations in these regions were no different from
controls (figure 3.6).

The results presented in figure 3.5 are consistent with the view that prolactin
mediates the stimulatory effects of haloperidol on TIDA neurons (for review see Moore
and Lookingland, 1995), and suggest that delayed activation of these neurons by prolactin
is dependent upon a period of sustained hyperprolactinemia for both the initiation and
maintenance of this response. To determine if this is the case, haloperidol-treated rats
received injections of either PRL-AB or NRS every 3 h such that animals were exposed
to periods of hyperprolactinemia of 3, 6 or 12 h duration, followed by 9, 6, and 0 h,
respectively, of immunoneutralized circulating prolactin (figure 3.7, upper panel).
Responses in these animals were examined 12 h after haloperidol administration. As
shown in figure 3.8 (left panel), 12 h of sustained ha10peridol-induced

hyperprolactinemia increased DOPAC concentrations in the median eminence, whereas

69

205:8 0885-202? 80c Amodvn:
Ecuommv bamomiwfi 0:: 35 30:3 .... 005580 560:: on: E <Q 0:“ U<mOQ m0 @805 wEEE 223.5588 «0 323::5800
04. m0 2mm L 8:: 30E? 0:“ 8on 28200: 38200 defiance: Bowen 3 o mmz 5.3 Ea dogafisoov 20.30 s 2 068200
:30 Um meSE 0. C Eon 0:853 o\om.o 55, 088:: 0:03 A2828 :33 205:8 20£0> doufiamoov 283 a o E m a .2 ant—100$
mmz £5 0:: .:0_§E800 2800 a S 32:32:: 5S, 0800? 203 205:00 :385800 Bowen : o E m 0 m3 vat—100$ marl—ME
5_; 0:: dozfifimooc 0:83 a E A3528 0:00. mom Mme—Ea: 0.: 62.5022 ES» 0800? 203 Sam .38 20.: 08854001303:
L0 85580 560E 05 E 32m Emma <9 0:: 32m Hmmd U<mOQ m0 2055:0800 :0 mars—Mm .«0 mcoobm mam 03$..—

 

 

 

 

 

 

 

 

 

 

mmz mmz
o o
.m G
.omq w
W -SW
.Su

SF -2\u1
w a fi
.ooqtox w
.ommw

.02 ... .mm

 

Being—mm I
2023/ H.

70

STRIATUM

DOPAC (ng/mg)
DA (ng/mg)

   

NRS NRS PRL-AB

1:] Vehicle
— Haloperidol

NU CLEUS ACCUMBENS

DOPAC (ng/mg)
DA (Hg/mg)

   

NRS NRS PRL-AB

Figure 3.6 Effects of PRL-AB on concentrations of DOPAC (LEFT PANELS) and DA
(RIGHT PANELS) in the striatum (UPPER PANELS) and nucleus accumbens (LOWER
PANELS) of haloperidol-treated male rats. Rats were injected with haloperidol (1.0
mg/kg; sc; solid columns) 12 h before decapitation, and with PRL-AB (200ul/rat; iv; q 3
h) 6 h before decapitation. Controls were injected with haloperidol 12 h before
decapitation, and with NRS (ZOOuI/rat; iv; q 3 h) 6 h before decapitation. Vehicle controls
(open columns) were injected with 0.3% tartaric acid (1.0 ml/kg; sc; open columns) 12 h
before decapitation, and with NRS 6 h before decapitation. Columns represent means and
vertical lines 1 SEM of 7-9 determinations of concentrations (rig/mg protein) of DOPAC
and DA in the striatum and nucleus accumbens. *, values that are significantly different

(p<0.05) from vehicle-treated controls.

71

TIME (11)

nun—c
_
-LN

 

A

G
NRS ‘

 

NRS-1 -- DJ

 

n NRS! -0«
I NRS! -\0

 

OJ
NRS ‘
PRL
AB
PRL
AB
PRL
AB

 

 

1

 

PRL
AB

PRL
AB

 

 

 

NRS: NRS
NRsn NRS‘
NRS!

NRsu

12

 

 

 

1

VEHICLE or DECAPITATION
HALOPERIDOL

ON
\0
p—r
p—n
no
N

 

NRS‘ -'- u

 

1 NRS‘ ‘—

 

 

DJ
NRS
PRL
AB
PRL
AB
PRL
AB

 

L

 

 

PRL
AB
PRL
AB

 

 

 

‘NRsn NRS:
NRS‘
NRS!

 

 

 

m .
a:
z _ ,
f V
PROLACTIN W
VEHICLE or
HALOPERIDOL DECAPITATION

Figure 3.7 Treatment paradigm for comparison of the effects of haloperidol on median
eminence DOPAC concentrations in PRL-AB- (UPPER PANEL) and PRL-AB + prolactin-
treated (LOWER PANEL) male rats. Horizontal shaded bars represent periods of exposure to
hyperprolactinemia before decapitation at 12 h. UPPER PANEL: Controls were injected with
0.3% tartaric acid (1 ml/kg; sc; 0 h) and NRS (200ul/rat; iv; 0 h; q 3 h; arrow heads) before
decapitation at 12 h; there was no hyperprolactinemia. Haloperidol (1 mg/kg; sc; 0 h)-treated rats
were injected with either NRS or PRL-AB (200ul/rat; iv; 0 h; q 3 h; arrow heads) such that
groups of rats were exposed to initial periods of hyperprolactinemia lasting 3, 6 or 12 h in
duration. LOWER PANEL: Controls were injected with 0.3% tartaric acid (1 ml/kg; sc; 0 h),
NRS (200ul/rat; iv; 0 h, q 3 h; arrow heads) and prolactin (lOug/rat; icv; 11 h) before
decapitation at 12 h such that the initial duration of exposure to hyperprolactinemia was zero,
despite hyperprolactinemia at the time of measurement. Haloperidol (1 mg/kg; sc; 0 h)-treated
rats were injected with either NRS or PRL-AB (200ul/rat; iv; 0 h; q 3 h; arrow heads) and
prolactin (lOug/rat; icv; 11 h) before decapitation at 12 h such that groups of rats were exposed

to initial periods of hyperprolactinemia lasting 3, 6 or 12 h in duration and hyperprolactinemia at
the time of measurement.

72

0:05:00 005005-202?
ea: sodve 2200: 3:80:00 20 as 38 3858000203 5: as: .. 385:5 502: 2:: a $40.72: 502: M538.: 0:05
205055050 <9 0:0 3000:: MERE 055058050 05909 90 205E000 a-” :0 29m 9 8:: 085:0.» 0:0 0:005: 05 5:000:00: 3:509
.: N: :0 0885:0000 0:03 0:0 .:05050E5:00 0:00 05050: :050 n 2 5:00:05 “550555800 0:00 050.53 $2: .50 0E5 0:: 50 $5ng : m b0>0
9:2 53 00:00.0: 0:03 0:05:00 :30— : _ 0805:0000 0:03 0:0 550505—500 305250: :050 a 2 $0: 50.51 03 5:00:05: .50 5500.3: 03:; 0
3:08: 8:0 :20: :2 2:: a: 0358:. a E 08:38: 0.0.0:: 8 E 005189 002 850350 as Baseéfiaoam 502$: 55::
.: N: :0 55850000 5:: 5505:5600 0:00 058:0: $90 .50 0E5 0:: :0 wfigmon : m b0>0 9:2 :53 00:00:: 0:03 0.05:00 .:050:=0 E : N:
:0 0 .m 9500— 00:0580:0:&0§: :0 0050: 005:: 0: 000098 0:03 0:0: .50 3:05 :05 5000 550533800 80530—0: .50 0:5 0:: :0 95:53: : m
b0>0 E 3053008 maxim: :0 A3 m50.51508 9:2 55:0 :03 00:02:: 0:03 0:0: 00:005._0050:0_0: ”997?..— .E99 :80: : N: 0885:0000 0:0
35:38 :0:0 :00 905:: o: 0_0E0> 0:00 058:0: find :0 $55.00 0:0: :00 Mme—RE 0.: 305030: :050 53 00500:: 0:03 0:0: =< .80: 20:.
0885:5028: + mafia: 0:0 9:74am 5 305055050 U<mO9 00:05::0 5.00:: :0 605320: (:0 0:00.50 0:: :0 550980 ad 05“....

:0 SEmzfiofiomnamé: 20555:

N: o m o
o o
1m
.0—
-0—
1|l 1m:
-m—
2 o m c .8
i_ ,0
on T2
u.n ism .3
. . . :00

L5“: -3 305000009: I :E .on
0_0E0> H

AME + 9085: 941.999

 

 

V1

(gm/3U) ovaoo

 

 

73

in rats exposed to only 3 or 6 h of haloperidol-induced hyperprolactinemia median
eminence DOPAC concentrations were not significantly different from vehicle-treated
controls. No duration of hyperprolactinemia effected concentrations of dopamine in the
median eminence. This experiment was repeated in another set of animals except that rats
exposed initially to 3 or 6 h of hyperprolactinemia (followed by 9 and 6 h of
hypoprolactinemia, respectively) received an icv injection of prolactin l h prior to
decapitation (figure 3.7, lower panel). As shown in figure 3.8 (right panel), median
eminence DOPAC concentrations were increased in response to prolactin in rats exposed
to 6 or 12 h of haloperidol-induced hyperprolactinemia, but prolactin-treated rats exposed
to 3 h of hyperprolactinemia were not significantly different from vehicle-treated
controls. No duration of hyperprolactinemia effected concentrations of dopamine in the

median eminence.

74

C. DISCUSSION

Passive immunoneutralization of circulating hormones represents a useful
alternative to disruptive pharmacological and surgical procedures to study neuroendocrine
feedback regulation of the central nervous system. In lieu of altering hormone secretion,
injection of antiserum directly into the blood causes rapid and selective binding of the
targeted hormone thereby rendering it inaccessible to its site of action. In the case of
prolactin, passive immunoneutralization has been used successfully to ascribe its role in a
variety of neurologic (Roky et al., 1995; Crisostoma et al., 1997), autonomic (Mills and
Ward, 1985) and neuroendocrine functions (Milenkovic et al., 1990; Parker et al., 1991;
Pape and Tramu, 1996), including feedback regulation of TIDA neurons (Gudelsky and
Porter, 1980; Porter et al., 1990; Toney et al., 1992b; Wagner etal., 1993).

Interpretation of the results of immunoneutralization studies is dependent upon
both selectivity of antiserum for its targeted hormone and capacity of antiserum to bind
sufficient free hormone to eliminate its biological actions. PRL-AB used in these studies
was generated in rabbits using NIDDK rat prolactin as the directed antigen. Since this
preparation was obtained from anterior pituitary extracts the possibility existed that PRL-
AB may also contain antibody directed against other (albeit minor) contaminant
hormones. Radioimmunoasssay analyses of binding specificity for this antiserum
revealed, however, that PRL-AB used in these experiments was selective for prolactin,
having no affinity for luteinizing hormone or growth hormone (Hentschel et al., 2000a).

The capacity of PRL-AB to eliminate free bioactive prolactin from the circulation
was evaluated by determining the time course effects of PRL-AB on plasma prolactin

concentrations quantified using the Nb2 lymphoma bioassay (Canon et al., 1991).

75

Intravenous administration of PRL-AB decreased plasma prolactin to non-detectable
levels for up to 3 h in haloperidol-pretreated male rats. The capacity of PRL-AB to bind
prolactin in the face of activated secretion may be exceeded after 3 h, and hence repeated
injections of PRL-AB were utilized to maintain more prolonged periods of
hypoprolactinemia.

The primary function of TIDA neurons is to suppress the secretion of prolactin
from the anterior pituitary. Indeed, experimental procedures which either; 1) disrupt
synthesis and release of dopamine from TIDA neurons, 2) prevent access of dopamine to
the anterior pituitary lactotrophs (i.e. pituitary stalk section; pituitary transplantation), or
3) block pituitary D2 receptors, all increase prolactin secretion (Ben-Jonathan, 1985). In
females, the inhibitory effect of dopamine on the synthesis and secretion of prolactin is
opposed by estrogen, consequently circulating prolactin levels in females are higher than
males. Higher basal prolactin in females, in turn, tonically stimulates TIDA neurons such
that under normal conditions the activity of these neurons is 2-3 times that of males
(Moore and Lookingland, 1995). In males, testosterone has no effect on prolactin
secretion (Toney et al., 1991), and circulating prolactin levels are insufficient to tonically
activate TIDA neurons (Demarest and Moore, 1981b).

The results of the present study confirm that TIDA neurons in male rats are
unresponsive to acute PRL-AB-induced hypoprolactinemia. Median eminence DOPAC
concentrations remain unchanged up to 3 h, yet are decreased at 6 h. These results are
consistent with a previous study in which median eminence DOPAC concentrations were
decreased 12 h, but not 4 h after PRL-AB injection in male rats (Hentschel et al., 2000a).

These results differ from another study (Demarest and Moore, 1981b) in which

76

bromocriptine was used to depress endogenous prolactin secretion. In that study DOPA
accumulation was not decreased in the median eminence of male rats until 48 h after the
first injection of bromocriptine. The findings of Demarest and Moore may be reconciled
with our results in the light of recent reports (Durham et al., 1997;1998). It is now known
that in male rats TIDA neuronal activity is tonically inhibited by dynorphinergic neurons,
and that these neurons are inhibited by D2 receptor agonists such as bromocriptine.
Accordingly, while bromocriptine suppressed endogenous prolactin secretion and
removed this stimulus from TIDA neurons, this drug simultaneously disinhibited TIDA
neuronal activity. Hence the effect of hypoprolactinemia could have been masked. The
data suggest that in male rats prolonged periods of hypoprolactinemia are required to
reduce basal activity of TIDA neurons. Furthermore, since median eminence DOPAC
concentrations remained decreased at 6 h at which time hyperprolactinemia had overcome
the capacity of PRL-AB to immunoneutralize circulating prolactin, the data also suggest
that once TIDA neuronal activity has been depressed there is a latent period following
acute changes in circulating prolactin concentrations that precedes activation of these
neurons in male rats.

Prolactin induces changes in TIDA neurons which result in delayed activation of
these neurons to levels 2-3 times basal (Demarest et al., 1985a; Selmanoff, 1985), and
maintenance of this response during chronic stimulation (Moore et al., 1985). This
inductive component of prolactin feedback regulation of TIDA neurons is dependent
upon protein synthesis (Johnston et al., 1980) and is associated with increased expression
of tyrosine hydroxylase mRNA in neuronal perikarya in the arcuate nucleus (Arbogast

and Voogt, 1991; Selmanoff et al., 1991). These features suggest that prolactin induces

77

long-term changes in gene expression which result in an increased capacity of TIDA
neurons to synthesize and release dopamine.

Temporal aspects of prolactin feedback related to the initiation and maintenance
of activation of TIDA neurons were determined in the present study using the
haloperidol-induced hyperprolactinemia experimental model in combination with
systemic administration of PRL-AB. This approach permitted the examination of the
effects of finite initial periods of hyperprolactinemia on TIDA, NSDA and MLDA
neurons, and of the dependence on circulating prolactin in this process. The results
confirm that haloperidol-induced hyperprolactinemia is associated with delayed (by 6-12
h) activation of TIDA neurons (Moore and Lookingland, 1995), rapid, prolonged
activation of NSDA and MLDA neurons (Gudelsky and Moore, 1977), and that the
stimulatory effects of haloperidol on these neurons are mediated by prolactin in TIDA
neurons (Gudelsky and Porter, 1980), but not in NSDA or MLDA neurons.

The results are consistent with regulation of NSDA and MLDA, but not TIDA
neurons by autoreceptors (Nowycky and Roth, 1978; Demarest and Moore, 1979).
Accordingly, the acute decrease in dopamine concentrations in terminal regions of NSDA
and MLDA, but not TIDA neurons may be due to autoreceptor-mediated disinhibition of
these systems and the transient inability of synthesis to keep pace with release.

Moreover, these results reveal that more than 3 h of exposure to elevated prolactin
is necessary to prime TIDA neurons to the stimulatory actions of prolactin, and that
hyperprolactinemia need not continue beyond 6 h for priming to occur. Indeed, median
eminence DOPAC concentrations in rats exposed to hyperprolactinemia for 6 h (followed

by 6 h of PRL-AB-induced hypoprolactinemia) were significantly elevated in response to

78

central administration of prolactin as compared with vehicle-treated controls, whereas
values in rats exposed to hyperprolactinemia for only 3 h (followed by 9 h of PRL-AB-
induced hypoprolactinemia) were not. The observation that median eminence DOPAC
concentrations in rats exposed to 6 h of hyperprolactinemia followed by 6 h of
hypoprolactinemia were not significantly different from zero time controls suggests that
priming is not sufficient to induce delayed activation of these neurons in the absence of
circulating prolactin.

In summary, the results of the present study confirm: (1) that the activity of NSDA
and MLDA neurons are not regulated by prolactin, (2) the temporal characteristics of the
delayed (6-12 h) inductive activation of TIDA neurons by prolactin, (3) that prolactin
mediates the activation of TIDA neurons by haloperidol, and (4) that prolonged periods of
hypoprolactinemia are required to depress the basal activity of TIDA neurons in male
rats. The results of the present study also reveal that delayed induced activation of TIDA
neurons by prolactin is dependent upon a priming period of hyperprolactinemia of greater
than 3 h duration for initiation and sustained hyperprolactinemia for maintenance of this

response.

79

4. EFFECTS OF PROLACTIN ON EXPRESSION OF FRA IN TH-IR
NEURONS IN SUBDIVISIONS OF THE ARCUATE NUCLEUS

A. INTRODUCTION

TIDA neurons located in the ARC project to the median eminence where
dopamine released from these neurons is transported in the hypophysial portal blood to
the anterior pituitary where it tonically inhibits prolactin secretion from lactotrophs.
TIDA neurons lack dopamine autoreceptors characteristic of other central dopaminergic
systems (Demarest and Moore, 1979) and are regulated by prolactin feedback. Increases
or decreases in circulating prolactin concentrations produce corresponding changes in the
synthesis and metabolism of dopamine in terminals of these neurons in the median
eminence (for review see Moore and Lookingland, 1995).

In addition to the "tonic" stimulatory action of prolactin that sets the higher basal
TIDA neuronal activity in females as compared with males (Moore and Lookingland,
1995), prolactin induces changes in TIDA neurons that culminate in both delayed
activation of these neurons to levels 2-3 times basal (Demarest et al., 1985a; Selmanoff,
1985) and maintenance of this response during chronic stimulation (Moore et al., 1985).
This "inductive" component of prolactin feedback regulation of TIDA neurons is
dependent upon protein synthesis (Johnston et al., 1980) and associated with increased
expression of TH mRNA in the ARC (Arbogast and Voogt, 1991; Selmanoff et al.,
1991). These features suggest that prolactin induces long-term changes in gene expression
that result in an increased capacity of TIDA neurons to synthesize and release dopamine;

such changes are characteristic of late-response gene expression.

80

In TIDA neurons, various stimuli and neuroendocrine manipulations induce the
rapid transient expression of immediate early genes including FRA (Hoffman et al.,
1993a; 1994). Newly translated FRA proteins translocate to the nucleus and
heterodimerize with constitutively expressed Jun-related transcription factors (Morgan
and Curran, 1989; Hughes and Dragunow, 1995). These heterodimers bind AP-l
elements (Sheng and Greenberg, 1990) identified in the TH gene promoter (Kumer and
Vrana, 1996) and modulate transactivation rate of various late response genes, including
TH (Icard-Liepkalns et al., 1992; Guo et al., 1998). Accordingly, expression of FRA in
TH-containing perikarya of the ARC has been shown to reflect the activity of TIDA
neurons (Hoffman et al., 1994; Lerant et al., 1996; Lerant and Freeman, 1997; Cheung et
al., 1997), and represents an early neurochemical marker of activity which permits the
identification of populations of chemically similar neurons uniquely responsive to
specific stimuli.

While TIDA neurons share TH-IR and residence within the ARC, there is
anatomical, histochemical and biochemical evidence to divide the ARC into dorsomedial
(DM) and ventrolateral (VL) subdivisions. Perikarya of the DM-ARC have more intense
TH-IR staining than those of the VL-ARC. In addition, there is a discrepancy in TH- and
AAAD-IR between the DM and VL subdivisions of the ARC. The DM-ARC contains
both TH-IR and AAAD-IR, whereas the VL-ARC exhibits only TH-IR (Skagerberg et al.,
1988; Misu et al., 1996). Hence, the DM-ARC contains the full complement of enzymes
necessary to synthesize dopamine, whereas the VL-ARC lacks AAAD and produces only
DOPA (Meister et al., 1988a; Komori et al., 1991). However, since vascular tissue may

express AAAD, it is possible that DOPA released at the median eminence into the

81

hypophysial portal vessels may be converted to dopamine en route to the pituitary
(Bertler et al., 1964; Constantinidis et al., 1969).

Since prolactin activates TIDA neurons after long latency, and changes in
immediate early gene products like FRA, can be used as an index of activity in these
neurons, we hypothesize that prolactin will cause changes in FRA expression in TIDA
neurons. If FRA expression in TIDA neurons is regulated by prolactin, then dual
TH/FRA immunohistochemistry will show differential expression of FRA-IR in TH-IR
neurons in the ARC after prolactin treatment. To this end, initial experiments determined
the effects of haloperidol-induced endogenous hyperprolactinemia on overall numbers of
TH-IR neurons and the percentage of these neurons expressing FRA in the DM and VL
subdivisions of the rostral, middle and caudal ARC (Cheung et al., 1997; Paxinos and
Watson, 1997) of male rats. Follow-up experiments examined the duration of initial
hyperprolactinemia required to elicit an effect, and compared the effects of haloperidol to

those of exogenous prolactin administration in both male and diestrous female rats.

82

B. RESULTS

TH-IR neurons are not homogeneously distributed throughout the rostrocaudal
extent of the ARC. Rather, there exists a rostrocaudal gradient in numbers of TH-IR
neurons within ARC subdivisions (Cheung et al., 1997). In male rats, there are fewer
TH-IR neurons in the DM subdivision as ARC regions extend caudally (figure 4.1). In
contrast, there are successively more TH-IR neurons in the VL subdivision as ARC
regions extend caudally. Hence, numbers of TH-IR neurons in the DM relative to the VL
ARC subdivision vary by rostrocaudal region. In male rats, the ratio of numbers of TH-
IR neurons in the DM-ARC to those of the VL-ARC is greater in the rostral ARC (4:1)
than the middle ARC (3:2) or caudal ARC ( 1:1). Despite the difference in numbers of
TH-IR neurons in ARC subdivisions and rostrocaudal regions, the percentage of TH-IR
neurons expressing FRA-IR nuclei remains uniformly low (3-5%) in all subdivisions and
regions of untreated male rats (figure 4.2).

Haloperidol is a D2 dopamine receptor antagonist that stimulates prolactin
secretion via disinhibition of pituitary lactotrophs. Systemic administration of haloperidol
increased plasma prolactin concentrations to peak levels by 1 h, and maintained
endogenous hyperprolactinemia for 12 h post-injection (figure 4.3). Haloperidol caused
a transient increase at 3 h in the percentage of TH-IR neurons expressing FRA-IR nuclei
in the DM subdivision of all rostrocaudal ARC regions in male rats (figure 4.4).
Furthermore, there was a rostrocaudal gradient with respect to magnitude of response;
effects were greater in the rostral ARC (8-fold increase) than the middle ARC (5-fold
increase) or caudal ARC (3-fold increase). In the VL subdivision of all rostrocaudal ARC

regions haloperidol caused a rapid (by 1.5 h) and prolonged (through 12 h) decrease in the

83

DM-ARC

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

80 j T
2’ T *
Q 60 ~ T
O4
8
z 40 L
E
I“ 20 L
:n:

0

Rostral Middle Caudal
80 -
VL-ARC

%
Q 60 - >1: T
g”; T
L2] 40 ~
E
[_‘ 20 ~
11:1:

0

Rostral Middle Caudal

Figure 4.1 Comparison of the numbers of TH-IR neurons in the DM and VL
subdivisions of rostral, middle and caudal ARC of male rats. After perfusion, brains were
removed, sectioned in the frontal plane (30 um) and immunohistochemically processed
for the detection of TH-IR and FRA-IR in the DM (upper panel) and VL (lower panel)
subdivisions of the rostral, middle and caudal ARC. Approximate distances relative to
bregma for these regions were -2.3 mm for rostral ARC, -2.8 mm for middle ARC and -
3.3 mm for caudal ARC (Paxinos and Watson, 1997). Columns represent means and
vertical lines 1 SEM from 6-8 animals. *, Values that are significantly different (p <
0.05) from the rostral ARC.

84

8 r DM-ARC

 

 

 

 

 

 

 

 

 

 

 

%TH-IR neurons expressing FRA
_i

 

 

 

 

 

 

 

 

 

 

 

0
Rostral Middle Caudal
g 8* VL-ARC
on
.S
a 6,“
8
g l
m 4
t:
e T l
:3
“5 2
E l
E: L
°\o O
Rostral Middle Caudal

Figure 4.2 Comparison of %TH-IR neurons expressing FRA in the DM and VL

subdivisions of rostral, middle and caudal ARC of male rats. Rats were perfused, brains

sectioned and tissue immunohistochemically processed for the detection of TH-IR. 311d
FRA-IR in the DM (upper panel) and VL (lower panel) subdivisions of the rostral, mlddle
and caudal ARC. Approximate distances relative to bregma for these regions were j2-3
mm for rostral ARC, -2.8 mm for middle ARC and -3.3 mm for caudal ARC (PaXIIlOS
and Watson, 1997). Columns represent means and vertical lines 1 SEM from 7'9
animals.

85

 

 

25— *
A
E
m20~
.2 *
.3 *
gs
a *
E10~
O
<
b]
O 5*
Dd
Q4
0 1 l l 1 J
0 1 3 6 12

HOURS AFTER HALOPERIDOL

Figure 4.3 Time course effects of haloperidol on plasma prolactin concentrations in
male rats. Rats were injected with haloperidol (1 mg/kg; sc) and decapitated 1, 3, 6 or 12
h later. Zero time controls were injected with vehicle (0.3% tartaric acid; 1 ml/kg; sc) 3 h
prior to decapitation. Symbols represent the means and vertical lines 1 SEM of prolactin
concentrations in plasma of 7-8 rats. *, Values for haloperidol-treated rats that are
significantly different (p < 0.05) from vehicle-treated controls.

86

DM-ARC

 

 

 

 

 

 

 

§ 10 -

Ln

O

E s _

Cl)

é) a:

g 6 - O Rostral

DJ [3 Middle

%3 A Caudal

E 4'

E 2-

e1

E O b 1 1 I I J

°\° 0.0 1.5 3.0 6.0 12.0
HOURS AFTER HALOPERIDOL

Figure 4.4 Time course effects of haloperidol on %TH-IR neurons expressing FRA in
the DM subdivision of rostral, middle and caudal ARC of male rats. Rats were injected
with haloperidol (filled symbols; 1 mg/kg; sc) and perfused 1.5, 3, 6 or 12 h later. Zero
time controls (open symbols) were injected with vehicle (0.3% tartaric acid; 1 ml/kg; sc)
3 h prior to perfusion. Approximate distances relative to bregma for these regions were
-2.3 mm for rostral ARC (circles), -2.8 mm for middle ARC (squares) and -3.3 mm for
caudal ARC (triangles) (Paxinos and Watson, 1997). Symbols represent means and
vertical lines 1 SEM from 6-8 animals. *, Values for haloperidol-treated rats that are
significantly different (p < 0.05) from vehicle-treated controls.

87

percentage of TH-IR neurons expressing F RA-IR nuclei (figure 4.5). However,
haloperidol had no effect on numbers of TH-IR neurons at any time point in any ARC
subdivision or rostrocaudal region (figure 4.6).

The effects of haloperidol could be mediated by direct action on central dopamine
receptors, or indirectly via endogenous hyperprolactinemia. Accordingly, endogenous
prolactin was immunoneutralized to determine if the effects of haloperidol on FRA
expression in TIDA neurons were mediated by prolactin. Prolactin antiserum was
administered at a dose known to block the stimulatory effects of estrogen on TIDA
neuronal activity (Toney et al., 1992b) and suppress plasma prolactin concentrations for
at least 3 h after haloperidol treatment (figure 3.4). Co-administration of prolactin
antiserum completely blocked the haloperidol-induced increase in percentage of TH-IR
neurons expressing FRA at 3 h in the DM-ARC of all rostrocaudal regions (figure 4.7).
In contrast, prolactin antiserum failed to block the haloperidol-induced decrease in
percentage of TH-IR neurons expressing FRA at 3 h in the VL—ARC of all rostrocaudal
regions (figure 4.8).

Hyperprolactinemia of different durations was investigated to fimher characterize
the stimulus required to induce FRA expression in DM-ARC TH-IR neurons. To this
end, rats were exposed to initial periods of haloperidol-induced hyperprolactinemia of 0,
1 or 3 h duration. Rats that had experienced 3 h of hyperprolactinemia (positive controls)
demonstrated the expected increase in percentage of DM-ARC TH-IR neurons expressing
FRA at 3 h (figure 4.9). In contrast, FRA expression in rats co-administered prolactin
antiserum and haloperidol (antibody and negative controls), or exposed to only 1 h of

hyperprolactinemia, were no different from vehicle-treated controls. This study suggested

88

 

 

 

§ 8_ VL-ARC

”-1

O

E

E3 6—

?? O Rostral

LI.) 4 _ El Middle

% A Caudal

Q

g 2 —

a.

E °‘

°\° 0.0 1.5 3.0 6.0 12.0
HOURS AFTER HALOPERIDOL

Figure 4.5 Time course effects of haloperidol on %TH-IR neurons expressing FRA in
the VL subdivision of rostral, middle and caudal ARC of male rats. Rats were injected
with haloperidol (filled symbols; 1 mg/kg; sc) and perfused 1.5, 3, 6 or 12 h later. Zero
time controls (open symbols) were injected with vehicle (0.3% tartaric acid; 1 ml/kg', 5°)

3 h prior to perfusion. Approximate distances relative to bregma for these regions We“:
-2.3 mm for rostral ARC (circles), -2.8 mm for middle ARC (squares) and -3.3 mm for
caudal ARC (triangles) (Paxinos and Watson, 1997). Symbols represent means and
vertical lines 1 SEM from 6-8 animals. *, Values for haloperidol-treated rats that are
significantly different (p < 0.05) from vehicle-treated controls.

89

DM-ARC

 

 

 

 

 

 

 

 

100 ~ 0 Rostral
D Middle
(2 . A/s:
O
M 80 *
:3
LL]
Z ' W
E v I 4
I 60 r- _ -
E v
:u: .
40 1 1 1 l I
0.0 1.5 3.0 6.0 12.0
80 ~ VL-ARC
(2 .
g 6o . W" A A
E 40 -
E .
:n: 20 b Q/‘\F + ‘
04.0 14.5 3:0 6:0 121.0
HOURS AFTER HALOPERIDOL

Figure 4.6 Time course comparison of the numbers of TH-IR neurons in the DM and
VL subdivisions of rostral, middle and caudal ARC of vehicle- and haloperidol-treated
male rats. Rats were injected with haloperidol (filled symbols; 1 mg/kg; sc) and perfused
1.5, 3, 6 or 12 h later. Zero time controls (open symbols) were injected with vehicle
(0.3% tartaric acid; 1 ml/kg; sc) 3 h prior to perfusion. Approximate distances relative to
bregma for these regions were -2.3 mm for rostral ARC (circles), -2.8 mm for middle
ARC (squares) and -3.3 mm for caudal ARC (triangles) (Paxinos and Watson, 1997).
Symbols represent means and vertical lines 1 SEM from 6-8 animals.

90

DM-ARC

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E 10 V :1 Control
(L; - Haloperidol
a at: PRL-AB + Haloperidol
a 8’ *
m l-
‘2
E ‘3
g 2 l.
E I J T /T T
g 0 % %
Rostral Middle Caudal

Figure 4.7 Effects of prolactin antiserum on %TH-IR neurons expressing FRA in the
DM subdivision of rostral, middle and caudal ARC of haloperidol-treated male rats. Rats
were injected with either haloperidol (filled columns; 1 mg/kg; sc) or its 0.3% tartaric
acid vehicle (open columns; 1 kag; sc), and with either prolactin antiserum (hatched
columns; 200 ul/rat; iv) or vehicle (normal rabbit serum, 200 ul/rat; iv) 3 h prior to
decapitation. Approximate distances relative to bregma for these regions were -2.3 mm
for rostral ARC, -2.8 mm for middle ARC and -3.3 mm for caudal ARC (Paxinos and
Watson, 1997). Columns represent means and vertical lines 1 SEM from 7-9 animals. *,
Values that are significantly different (p < 0.05) from vehicle-treated controls.

91

VL-ARC

:1 Control
T - Haloperidol
PRL-AB + Haloperidol

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2- * * T
1- *7} :1: * *
é % in

 

 

 

 

 

 

%TH-IR NEURONS EXPRESSING FRA
__{

Rostral Middle Caudal

Figure 4.8 Effects of prolactin antiserum on %TH-IR neurons expressing FRA in the VL
subdivision of rostral, middle and caudal ARC of haloperidol-treated male rats. Rats
were injected with either haloperidol (filled columns; 1 mg/kg; sc) or its 0.3% tartaric
acid vehicle (open colmnns; 1 ml/kg; sc), and with either prolactin antiserum (hatched
columns; 200 til/rat; iv) or vehicle (normal rabbit serum, 200 ill/rat; iv) 3 h prior to
decapitation. Approximate distances relative to bregma for these regions were -2.3 mm
for rostral ARC, -2.8 mm for middle ARC and -3.3 mm for caudal ARC (Paxinos and
Watson, 1997). Columns represent means and vertical lines 1 SEM from 7-9 animals. *,
Values that are significantly different (p < 0.05) from vehicle-treated controls.

92

DM-ARC

l::l Vehicle
6 _ - Haloperidol

 

 

 

 

 

 

%TH-IR NEURONS EXPRESSING FRA
A

CONTROL 3 l 0
HYPERPROLACTINEMLA DURATION (h)

Figure 4.9 Effects of duration of hyperprolactinemia on %TH-IR neurons expressing
FRA in the DM-ARC of haloperidol-treated male rats. Rats were injected with
haloperidol (filled columns; 1 mg/kg; sc) 3 h prior to perfusion and with prolactin
antiserum (200 til/rat; iv) 2 h prior to perfusion. Vehicle controls (open column) were
injected with 0.3% tartaric acid (1.0 ml/kg; sc) and with normal rabbit serum (200 ul/rat;
iv) 3 h before perfusion. Positive and negative controls were injected with haloperidol 3 h
prior to perfusion and with either normal rabbit serum or prolactin antiserum 3 h prior to
perfusion, respectively. Columns represent means and vertical lines 1 SEM from 7-9
animals. *, Values that are significantly different (p < 0.05) from vehicle-treated controls.

93

that in male rats, greater than 1 h of haloperidol-induced hyperprolactinemia is required
for increasing FRA expression in DM-ARC TH-IR neurons at the time of peak response
at 3 h.

Similar to the effects of haloperidol injection, central administration of prolactin
to male rats increased the percentage of DM-ARC TH-IR neurons expressing FRA at 3 h
in all rostrocaudal ARC regions (figure 4.10). This effect remained significant through 6
h in the rostral and middle ARC. The magnitude of response exhibited a rostrocaudal
gradient; effects were greater in the rostral and middle ARC (S-fold increases) than the
caudal ARC (4-fold increase). In the VL-ARC of male rats, prolactin caused a transient
increase in the percentage of TH-IR neurons expressing FRA at 1.5 h in all rostrocaudal
ARC regions which returned to control levels by 3 h (figure 4.11). Despite the changes in
immediate early gene expression, prolactin had no effect on numbers of TH-IR neurons at
any time point in any ARC subdivision or rostrocaudal region of male rats (figure 4.12).

By way of summary, figure 4.13 compares representative photomicrographs of
TH-IR neurons expressing FRA-IR nuclei in the rostral DM-ARC of male rats injected
with vehicle, haloperidol or prolactin 3 h prior to perfusion. The percentage of TH-IR
neurons expressing FRA was approximately 1% after vehicle treatment, 6% after
haloperidol treatment, and 25% after prolactin treatment in male rats.

Complementary studies performed in diestrous female rats demonstrated a
descending gradient in numbers of TH-IR neurons in the DM subdivision as the ARC
extends caudally, whereas numbers of these neurons in the VL subdivision remain fairly
constant throughout the rostrocaudal ARC (figure 4.14). Hence, numbers of TH-IR

neurons in the DM relative to the those of the VL ARC subdivision vary by rostrocaudal

94

DM-ARC

   
   

 

 

 

35
E l
2 30 " O Rostral
m l C] Middle
(é) 25 _ A Caudal
‘92 20-
LL]
%’
O 15 ~
Z
E 5 -
m'
[—
.\° 0 ‘ ‘ ‘ ‘ '
0.0 1.5 3.0 6.0 12.0
HOURS AFTER PROLACTIN

Figure 4.10 Time course effects of prolactin on %TH-IR neurons expressing FRA in the
DM subdivision of rostral, middle and caudal ARC of male rats. Rats were injected with
prolactin (filled symbols; 10 ug/rat; icv) and perfused 1.5, 3, 6 or 12 h later. Zero time
controls (open symbols) were injected with water vehicle (3 Ill/fat; icv) 6 h prior to
perfusion. Approximate distances relative to bregma for these regions were -2.3 mm for
rostral ARC (circles), -2.8 mm for middle ARC (squares) and -3.3 mm for caudal ARC
(triangles) (Paxinos and Watson, 1997). Symbols represent means and vertical lines 1
SEM from 6-8 animals. *, Values for prolactin-treated rats that are significantly different
(p < 0.05) from vehicle-treated controls.

95

VL-ARC

 

 

 

g 25 —
LL.
2

20 -
g *
g3 15 — O Rostral
LU Cl Middle
m A Caudal
E 10~
M
:3
LE
a 5*
m' i
i— ‘ ,
o\° 0 ' ‘ 1 '

0.0 1.5 3.0 6.0 12.0
HOURS AFTER PROLACTIN

Figure 4.11 Time course effects of prolactin on %TH-IR neurons expressing FRA in the
VL subdivision of rostral, middle and caudal ARC of male rats. Rats were injected with
prolactin (filled symbols; 10 pyrat; icv) and perfused 1.5, 3, 6 or 12 h later. Zero time
controls (open symbols) were injected with water vehicle (3 til/rat; icv) 6 h prior to
perfusion. Approximate distances relative to bregma for these regions were -2.3 mm for
rostral ARC (circles), -2.8 mm for middle ARC (squares) and -3.3 nun for caudal ARC
(triangles) (Paxinos and Watson, 1997). Symbols represent means and vertical lines 1
SEM from 7-8 animals. *, Values for prolactin-treated rats that are significantly different
(p < 0.05) from vehicle-treated controls.

96

DM-ARC

 

 

 

 

 

 

 

 

 

100 r O Rostral
0 Middle
(2 A Caudal
a 80
D W + 4
g M 4 i
9.4.
I 60 f x
m RV— ;
[—
qt r
40 1 i 1 1 i
0.0 1.5 3.0 6.0 12.0
80 f VL-ARC
:UZJ .
g 60 -L _ _ _ 1
% L (IF—FM— 4
m 40 r
I. _
[-'
#1: 20 “m
ofo is 3:0 610 15.0
HOURS AFTER PROLACTIN

Figure 4.12 Time course comparison of the numbers of TH-IR neurons in the DM and
VL subdivisions of rostral, middle and caudal ARC of vehicle- and prolactin-treated male
rats. Rats were injected with prolactin (filled symbols; 10 pig/rat, icv) and perfused 1.5, 3,
6 or 12 h later. Zero time controls (open symbols) were injected with vehicle (distilled
water; 3 ul/rat, icv) 3 h prior to perfusion. Approximate distances relative to bregma for
these regions were -2.3 mm for rostral ARC (circles), -2.8 mm for middle ARC (squares)
and ~3.3 mm for caudal ARC (triangles) (Paxinos and Watson, 1997). Symbols represent
means and vertical lines 1 SEM from 6-8 animals.

97

 

Figure 4.13 Comparison of digitized photomicrographs of TH-IR neurons expressing
FRA in the rostral DM-ARC of male rats injected with vehicle, haloperidol or prolactin.
Fluorescent TH-IR neurons under dark-field optics (panels A, C, E) and the same images
under bright-field optics (panels B, D, F, respectively) showing blackened FRA-IR nuclei.
Rats injected with either 0.3% tartaric acid (panels A and B; 1 ml/kg; sc), haloperidol
(panels C and D; 1 mg/kg; sc) or prolactin (panels E and F; 10 jig/rat; icv) were perfused
3 h later. Arrowheads represent TH-IR neurons with FRA-IR nuclei.

98

region. In female rats, the ratio of numbers of TH-IR neurons in the DM-ARC to those of
the VL-ARC is greater in the rostral ARC (11:1) than middle ARC (7:1), or caudal ARC
(5:1).

In a comparison of the numbers of TH-IR neurons in the rostral, middle, and
caudal ARC between male and diestrous female rats, there was no gender difference in
numbers of TH-IR neurons in the DM-ARC at any rostrocaudal region (figure 4.15).
Yet, in the VL-ARC there were more TH-IR neurons in male rats than females at all
rostrocaudal regions. Since numbers of TH-IR neurons in the VL-ARC remain relatively
constant in females, yet increase caudally in males, this gender difference is most
pronounced in the caudal ARC where the number of TH-IR neurons in males was
approximately 5 times greater than that in females.

Neurochemical studies have demonstrated that in female rats circulating prolactin
tonically stimulates TIDA neuronal activity to 2-3 times that of males (for review see
Moore and Lookingland, 1995). Accordingly, while the percentage of TH-IR neurons
expressing FRA-IR nuclei did not significantly differ across ARC subdivisions or
rostrocaudal regions within basally active females (figure 4.16), there was a marked
gender difference. In both DM and VL subdivisions of all rostrocaudal ARC regions the
basal percentage of TH-IR neurons expressing FRA-IR nuclei was greater in females than
males (figure 4.17). On average, the percentage of FRA expression in TH-IR neurons in
females was 4-times greater in the DM-ARC, and 2-7 times greater in the VL-ARC as
compared to males. In diestrous female rats the percentage of TH-IR neurons expressing
FRA-IR nuclei in the DM-ARC of all rostrocaudal regions was significantly increased 3 h

after central administration of prolactin (figure 4.18). Greatest effects of prolactin were

99

DM-ARC

 

 

 

 

 

 

 

 

 

 

 

 

 

... [ 7T ..
é //
z 40 — /
a?
E 20 - /
4. /
0 2 % //,
Rostral Middle Caudal
80‘ VL-ARC
E 60 —
E
Z 40 —
a.
E 20 ~
I“: 0 EZZZéZZZJ Eééégééé‘ EEZEEZ
Rostral Middle Caudal

Figure 4.14 Comparison of the numbers of TH-IR neurons in the DM and VL
subdivisions of rostral, middle and caudal ARC of diestrous female rats. After perfusion,
brains were removed, sectioned in the frontal plane (30 um) and immunohistochemically
processed for the detection of TH-IR and FRA-IR in the DM (upper panel) and VL (lower
panel) subdivisions of the rostral, middle and caudal ARC. Approximate distances
relative to bregma for these regions were -2.3 mm for rostral ARC, -2.8 mm for middle
ARC and -3.3 mm for caudal ARC (Paxinos and Watson, 1997). Columns represent
means and vertical lines 1 SEM from 6-8 animals. *, Values that are significantly
different (p < 0.05) from the rostral ARC.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. /T T ,T E2222 Female
.0 o~ / //
... / % %
E 20- / ¢ /
... , / / /
Rostral Middle Caudal
80' VL-ARC
% 60. I
§ T
z 40-
.95
E 20 ~ T *
Rostral Middle Caudal

Figure 4.15 Comparison of numbers of TH-IR neurons in the DM and VL subdivisions
of rostral, middle and caudal ARC of male and diestrous female rats. Male (open
columns) and diestrous female (hatched columns) rats were perfused, brains sectioned
and tissue immunohistochemically processed for the detection of TH-IR in the DM
(upper panel) and VL (lower panel) subdivisions of the rostral, middle and caudal ARC.
Approximate distances relative to bregma for these regions were -2.3 mm for rostral
ARC, -2.8 mm for middle ARC and -3.3 mm for caudal ARC (Paxinos and Watson,
1997). Columns represent means and vertical lines 1 SEM from 7-9 animals. *, Values
for diestrous female rats that are significantly different (p < 0.05) from region-matched
male rats.

101

30-

20

%TH-IR neurons expressing FRA

30-

%TH-IR neurons expressing FRA

25-

DM-ARC

 

 

 

\&

 

 

 

a\\§e

 

 

\\\

 

 

k

 

25~

20-

Rostral Middle C audal

VL-ARC

 

 

 

 

a \\V—‘
a\\\\)—e

 

 

 

 

 

 

 

Rostral Middle Caudal

Figure 4.16 Comparison of %TH-IR neurons expressing FRA in the DM and VL
subdivisions of rostral, middle and caudal ARC of female rats. Rats were perfused, brains
sectioned and tissue immunohistochemically processed for the detection of TH-IR and
FRA-IR in the DM (upper panel) and VL (lower panel) subdivisions of the rostral, middle
and caudal ARC. Approximate distances relative to bregma for these regions were -2.3
mm for rostral ARC, -2.8 mm for middle ARC and -3.3 mm for caudal ARC (Paxinos
and Watson, 1997). Columns represent means and vertical lines 1 SEM from 7-9

animals.

.—
O
N

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E 30_ DM-ARC
E 255 * l 22:22. as...
a: - /

a: 15 - / / *
é ...: ? é /
a _ / / /
a 0’ % a /
°\ Rostral Middle Caudal
g 30f VL-ARC

E” 25 ~ *

g 20: é a:
”i ? J
‘5 10~ / /
a . / /
e .- l % a
°\ Rostral Middle Caudal

Figure 4.17 Comparison of %TH-IR neurons expressing FRA in the DM and VL
subdivisions of rostral, middle and caudal ARC of male and diestrous female rats. Male
(open columns) and diestrous female (hatched columns) rats were perfused, brains
sectioned and tissue immunohistochemically processed for the detection of TH-IR and
FRA-IR in the DM (upper panel) and VL (lower panel) subdivisions of the rostral, middle
and caudal ARC. Approximate distances relative to bregma for these regions were -2.3
mm for rostral ARC, ~2.8 mm for middle ARC and -3.3 mm for caudal ARC (Paxinos
and Watson, 1997). Columns represent means and vertical lines 1 SEM from 7-9
animals. *, Values for diestrous female rats that are significantly different (p < 0.05)
from region-matched male rats.

103

observed in the rostral and middle ARC (40-45% TH-IR neurons expressed FRA) as
compared to the caudal ARC (30% TH-IR neurons expressed FRA). Additionally, in
female rats, prolactin had no effect at 3 h in the VL-ARC of any rostrocaudal region

(figure 4.18).

l04

DM-ARC

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

:2 60 ~
[ZZZ] Vehicle

LED i * * — Prolactin
E 50 -
a
8 40 ~
:2.
35 a:
m 30 .
c:
8 l
B 20 ~ T
c:
m I
'T‘ 10 ~
I .
l—
c.\" o

Rostral Middle Caudal

60 r

g VL-ARC
g? 50 - [22:] Vehicle
'5; — Prolactin
V)
g 40 .
x
a) 30
U, _
8 .
.5, 20 _ r
g 1
"r 10 ~
5: T
F: * i
°\ 0

Rostral Middle Caudal

Figure 4.18 Effects of prolactin on %TH-IR neurons expressing FRA in the DM and VL
subdivisions of rostral, middle and caudal ARC of diestrous female rats. Rats were
injected with prolactin (filled columns; 10 jig/rat; icv) or its water vehicle (open
columns; 3 ul/rat; icv) and perfused 3 h later. Brains were removed, sectioned and tissue
immunohistochemically processed for the detection of TH-IR and FRA-IR in the DM
(upper panel) and VL (lower panel) subdivisions of the rostral, middle and caudal ARC.
Approximate distances relative to bregma for these regions were -2.3 mm for rostral
ARC, -2.8 mm for middle ARC and -3.3 mm for caudal ARC (Paxinos and Watson,
1997). Columns represent means and vertical lines 1 SEM fi'om 7-9 animals. *, Values
for diestrous female rats that are significantly different (p < 0.05) from regionally-
matched vehicle-treated controls.

105

C. DISCUSSION

The discovery that the immediate early gene c-fos is rapidly and transiently
expressed in neurons following sensory stimulation (Morgan and Curran, 1989) has
spurred interest in the role of AP-l transcription factors such as FRA and Jun-related
antigens in mediating adaptive genomic changes in neurons following both acute and
chronic activation (Sheng and Greenberg, 1990; Pennypacker et al., 1995; Harris, 1998).
Moreover, mapping of FRA-IR neurons has led to the localization and neurochemical
identification of neuronal populations within specific brain regions that are uniquely
responsive to a variety of stimuli including seizure, brain injury, pain, and administration
of dopamine agonists (Pennypacker et al., 1995; Chaudhuri, 1997). The aim of the present
study was to determine if prolactin regulates FRA expression in TIDA neurons located in
anatomically distinct subdivisions of the rostral, middle and caudal ARC.

Perikarya of TIDA neurons, originally described as comprising a majority of the
A12 cell group (Dahlstrom and Fuxe, 1964), are distributed throughout the rostrocaudal
extent of the ARC. Two populations of TH-containing neurons have been identified
based on their neurochemical phenotypes, and the size and location of their perikarya in
the DM and VL subdivisions of the ARC (Everitt et al., 1986). TH-IR neurons in the
DM-ARC predominantly innervate the medial portion of the median eminence
(Kobayashi and Matsui, 1969; Rethelyi, 1985; Daikoku et al., 1986) and are generally
believed to be the primary inhibitors of pituitary prolactin release (Andersson et al., 1981;
Reymond et al., 1983; Hoffman et al., 1994; Lee and Voogt, 1999). TH-IR neurons in the
VL-ARC innervate the more lateral aspects of the median eminence (Kobayashi and

Matsui, 1969; Fuxe et al., 1986), but since these neurons lack AAAD and do not

106

synthesize dopamine (Meister et al., 1988a; Komori et al., 1991; Misu et al., 1996;
Skagerberg et al., 1988), their function is unknown. However, since vascular tissue may
express AAAD, it is possible that DOPA released at the median eminence into the
hypophysial portal vessels may be converted to dopamine en route to the pituitary
(Bertler et al., 1964; Constantinidis et al., 1969).

The results of the present study demonstrate an inverse relationship in male rats
between numbers of TH-IR neurons in the DM- and VL-ARC in the rostrocaudal plane;
i.e. with increasing caudal extent numbers of TH-IR neurons in the DM-ARC decrease,
whereas those of the VL-ARC increase. TH-IR neurons projecting to the medial aspect
of the median eminence have been shown to have a bimodal rostrocaudal distribution
(Kawano and Daikoku, 1987); however, it is possible that this rostrocaudal gradient may
be attributed to the inclusion of a small population of tuberohypophysial dopaminergic
perikarya (Lookingland et al., 1985) during quantification of the rostral ARC since these
neurons could not be distinguished from TIDA neurons in this study. The large number of
TH-[R neurons in the caudal VL-ARC of male rats suggests physiological relevance,
however, the role of these neurons remains unknown.

In agreement with a previous study (Cheung et al., 1997), there are different
numbers of TIDA neurons in the DM- and VL-ARC at any given rostrocaudal region.
For example, in male rats there are up to 4 times more TIDA neurons in the rostral DM-
ARC than in the adjacent rostral VL-ARC (Cheung et al., 1997). The preponderance of
TH-IR neurons in the DM-ARC may imply the relative importance of these neurons in

the regulation of prolactin secretion as compared to those of the VL-ARC.

107

Despite differences in numbers of TH-IR neurons in ARC subdivisions and
rostrocaudal regions, the percentage of these neurons expressing FRA-IR nuclei was
uniformly low (3-5%) across all ARC subdivisions and rostrocaudal regions in male rats.
These results presumably reflect the low level of neurotransmitter release and de novo
replenishment in TIDA neurons. Furthermore, these results argue against differential
levels of TIDA neuronal activity in the DM- and VL-ARC of male rats.

Haloperidol is a D2 receptor antagonist that rapidly stimulates prolactin secretion
via disinhibition of the anterior pituitary lactotrophs. The prolonged hyperprolactinemia
that follows feeds back to stimulate TIDA neurons (Moore and Lookingland, 1995).
Haloperidol has no direct acute effects on TIDA neurons. Its latent stimulatory effects can
be blocked by hypophysectomy (Gudelsky et al., 1977) or prolactin antiserum (Gudelsky
and Porter, 1980; Van Loon et al., 1983), demonstrating that the stimulatory effects are
mediated by prolactin. In male rats, this “inductive” activation of TIDA neurons is
characterized by long latency (12 h), increased TH mRN A in the ARC (Arbogast and
Voogt, 1991; Selmanoff et al., 1991) and dependence upon protein synthesis (Johnston et
al., 1980). These features suggest that prolactin induces long-term changes in gene
expression that result in increased capacity of TIDA neurons to synthesize and release
dopamine. Prolactin may elicit these effects through regulation of FRA expression in
TIDA neurons.

Consistent with this hypothesis, the results of the present study reveal that in male
rats prolactin modulates FRA expression in TIDA neurons located in both DM and VL
subdivisions of the ARC without affecting the overall numbers of these neurons. In the

DM-ARC, both haloperidol-induced endogenous hyperprolactinemia and central

108

administration of prolactin cause similar delayed increases in the percentage of TIDA
neurons expressing FRA with peak responses occurring by 3 h. This response persists for
an additional 3 h in prolactin-treated rats. This is likely due to the central route of
administration and the dose of prolactin used in this study which undoubtedly results in
greater access of prolactin to its site of action in the mediobasal hypothalamus (Gudelsky
et al., 1978) and a more prolonged activation of prolactin receptors within this region
(Chiu et al., 1992; Chiu and Wise, 1994).

Since changes in FRA expression in TIDA neurons of the DM-ARC expected at 3
h could be completely blocked by immunoneutralization of circulating prolactin 1 h after
haloperidol-injection, it is possible that there exists a stimulus threshold for the induction
of F RA expression (requiring duration of hyperprolactinemia greater than 1 h), or
continuous hyperprolactinemia is required for maintenance of the response in these
neurons.

In the VL-ARC, the stimulatory effect of prolactin on FRA expression in TIDA
neurons precedes that in the DM-ARC and is transient; i.e. by 3 h after prolactin
administration (when peak responses in TIDA neurons in the DM-ARC were attained)
effects in the VL-ARC were no longer observed. Considering that transcriptional
activation of c-fos and related immediate early genes occurs within minutes, with the
accumulation of nuclear protein reaching peak levels on average about 90—120 min later
(Harris, 1998). Thus temporal differences in the onset of prolactin-induced changes in
expression of F RA by TIDA neurons in the DM- and VL-ARC suggest two possibilities.
Either the mechanisms by which prolactin activates these neurons are different (i.e. acute

direct versus delayed secondary actions), or TIDA neurons in the VL—ARC are more

109

sensitive to the stimulatory actions of prolactin than those of the DM-ARC. Differences
in sensitivities of these neurons to prolactin could be due to differences in the types,
numbers and/or affinities of their prolactin receptors (Kelly et al., 1992; Chiu and Wise,
1994; Arbogast and Voogt, 1997; Bakowska and Morrell, 1997).

In male rats haloperidol suppresses FRA expression in TIDA neurons in the VL-
ARC and blocks the stimulatory effects of prolactin on these neurons via a prolactin-
independent mechanism. Indeed, immunoneutralization of circulating prolactin in
haloperidol-treated rats has no effect on the ability of this D2 dopamine antagonist to
decrease the percentage of TIDA neurons expressing F RA in the VL-ARC, in contrast to
the DM-ARC. The inhibitory effect of haloperidol on F RA expression in TIDA neurons
in the VL-ARC could due to blockade of D2 dopamine receptor-mediated tonic regulation
of immediate early gene expression in these neurons that is not present in TIDA neurons
of the DM-ARC. The significance of this finding remains to be elucidated.

Complementary studies in female rats examined the distribution of TH-IR neurons
in ARC subdivisions across rostrocaudal regions and compared these findings with those
of male rats. Similar to male rats, females exhibit a descending rostrocaudal gradient in
numbers of TH-IR neurons in the DM-ARC subdivision. In the VL-ARC a relatively
small subpopulation of TH-IR neurons is evenly distributed throughout the rostrocaudal
ARC. Since numbers of these neurons remain constant in females, yet increase caudally
in males, a gender difference exists; male rats have 5 times more of these neurons in the
caudal VL-ARC than female rats. While the role of TH—IR neurons of the VL-ARC
remains elusive, the results suggest a gender-related function. In agreement with previous

studies, female rats have several times more TIDA neurons in the DM-ARC than in the

110

VL-ARC at a given rostrocaudal region (Cheung et al., 1997; Lerant and Freeman, 1997).
The preponderance of TH-IR neurons in the DM-ARC, particularly in females, may imply
the relative importance of these neurons in the regulation of prolactin secretion as
compared to those of the VL-ARC.

There are sexual differences in the basal expression of F RA in TIDA neurons in
the DM-ARC (Cheung et al., 1997) that correspond with differences between males and
females in the activities of TIDA neurons associated with neurotransmitter release and its
replenishment via de novo synthesis. Indeed, the synthesis, release and metabolism of
dopamine in the median eminence are all elevated in females as compared with males
(Demarest and Moore, 1981b; Gudelsky and Porter, 1981; Demarest et al., 1984;
Lookingland et al., 1987a), due, in part. to higher levels in females of TH mRNA in
perikarya in the ARC (Arbogast and Voogt, 1990) and TH in terminals in the median
eminence (Porter, 1986). In agreement, the results of present study reveal that the
percentage of TH-IR neurons expressing FRA in both the DM- and VL-ARC of all
rostrocaudal regions is uniformly elevated and several-fold higher in non-stimulated
female rats as compared to males.

Since sexual differences in neurochemical activity of TIDA neurons are partially
due to the feedback effects of circulating prolactin (Moore and Lookingland, 1995) it was
likely that prolactin would also regulate FRA expression in females. Indeed, prolactin
increased the percentage of TIDA neurons expressing FRA in the DM-ARC, but not the
VL-ARC at 3 h in female rats. A comparison of the effects of prolactin on FRA
expression in TH-IR neurons of the DM-ARC in male and female rats shows a gender

difference in absolute magnitude of effect, i.e. females have a greater response to

111

prolactin than males. There is neurochemical evidence that removal of endogenous
circulating prolactin reduces the responsiveness of TIDA neurons to subsequent prolactin
administration (Demarest et al., 1985b). Therefore, it is plausible that the observed gender
difference in magnitude of response could be due to the different endogenous levels of
circulating prolactin and its effects on TIDA neurons. There is a rostrocaudal gradient in
magnitude of response to prolactin at a given dose in both male and female rats, such that
effects of prolactin are consistently greater in the rostral and middle ARC than the caudal
ARC. This observation could be due to differences in the types, numbers and/or affinities
of their prolactin receptors (Kelly et al., 1992; Chiu and Wise, 1994; Arbogast and Voogt,
1997; Bakowska and Morrell, 1997).

In conclusion, these results reveal that: (1) prolactin regulates immediate early
gene expression in TIDA neurons in both male and female rats, (2) there are temporal
differences in the response of TIDA neurons in DM- and VL-ARC subdivisions to
prolactin (i.e. prolactin initially stimulates transient expression of FRA in TIDA neurons
of the VL-ARC, followed by a more prolonged activation of FRA expression in TIDA
neurons of the DM-ARC), and (3) there is a gender difference in magnitude of response
of TIDA neurons to prolactin (i.e. female rats demonstrate a greater response to prolactin
than males), and (4) there are rostrocaudal differences in the magnitude of response of
TIDA neurons to prolactin (i.e. afier correcting for differences in absolute numbers of
TIDA neurons, prolactin elicits its greatest response in the rostral ARC and least response

in the caudal ARC).

112

5. EFFECTS OF PROLACTIN ON FRA EXPRESSION IN
NON-TH-IR CELLS IN THE ARCUATE NUCLEUS

A. INTRODUCTION

Neural activity mapping with inducible transcription factors (e.g. c-fos, FRA,
zif268) has led to the localization and neurochemical identification of neuronal
populations within specific brain regions that are uniquely responsive to a variety of
stimuli including seizure, brain injury, pain, and administration of dopamine agonists
(Sheng and Greenberg, 1990; Pennypacker et al., 1995; Chaudhuri, 1997).

In the previous chapter this method was utilized in the characterization of
prolactin-induced changes in FRA expression in TIDA neurons (Hentschel et al., 2000b).
During the analysis of those experiments, it appeared that changes in FRA expression
were also occurring in non-TH-IR cells of the ARC. Furthermore, these changes seemed
to temporally precede those in TIDA neurons and were localized within cells in the ARC,
consistent with the complete deafferentation study that demonstrated the haloperidol-
induced, prolactin-mediated increase in dopamine turnover in the median eminence
results from actions of prolactin on neurons within the mediobasal hypothalamus
(Gudelsky et al., 1977). Given the 6-12 h latency that precedes the delayed inductive
activation of TIDA neurons (Demarest et al., 1984; 1986), the approximate anatomical
localization of these cells and their apparent responsiveness to prolactin, it is plausible
that these chemically unidentified cells could mediate the effects of prolactin on TIDA
neurons.

Accordingly, the first step in this investigation is to verify the purported prolactin-

induced changes in immediate early gene expression in chemically unidentified cells in

113

the ARC. The working hypothesis is that prolactin regulates FRA expression in subsets
of non-TH-IR cells within this region. If this is the case, then dual FRA/T H
immunohistochemistry will show differential expression of FRA-IR in non-TH-IR cells
in the ARC after prolactin treatment. To this end, initial experiments determined the time
course effects of haloperidol-induced endogenous hyperprolactinemia on numbers of
FRA-IR cells that do not contain TH—IR in the DM and VL subdivisions of the rostral,
middle and caudal ARC (Paxinos and Watson, 1997) of male rats. Follow-up experiments
examined the ability of PRL-AB to block the effects of haloperidol on FRA expression
and compared the effects of haloperidol to those of exogenous prolactin administration in

male rats.

114

B. RESULTS

In the present study changes in FRA expression were reported by ARC
subdivision. As shown in figure 2.9, the ARC was divided into DM and VL subdivisions
by extension of a line dorsolaterally at an angle of 30-50° from the ventrolateral aspect of
the third ventricle. The DM subdivision was further separated into two regions based
upon the presence (magnocellular) or absence (parvocellular) of large TH-IR perikarya.

Basal F RA expression in non-TH-IR cells of the ARC was relatively low ranging
between 5-15 cells per region per section across all ARC subdivisions and rostrocaudal
regions in male rats (figure 5.1). Within a particular ARC subdivision, levels of FRA
expression were uniform throughout the rostrocaudal ARC. However, while FRA
expression in non-TH-IR cells in the magnocellular-UM and ventrolateral ARC were not
statistically different from one another, both regions had significantly greater FRA
expression than that of the parvocellular-DM ARC.

Haloperidol was utilized in these studies to induce endogenous
hyperprolactinemia As established in previous chapters (figures 3.1 and 4.1), systemic
administration of the D2 dopamine receptor antagonist haloperidol increased plasma
prolactin concentrations to peak levels by 1 h, and maintained endogenous
hyperprolactinemia for 12 h post-injection.

Figure 5.2 demonstrates the DM-ARC subdivisions and compares representative
photomicrographs of FRA expression in these areas of the middle ARC 3 h after vehicle
or haloperidol injection in male rats. Tyrosine hydroxylase-IR perikarya that fluoresce
characterize the magnocellular-, but not parvocellular-, DM ARC. Bright-field images

show that FRA expression is modestly increased in the magnocellular-DM ARC

115

[:1 Magnocellular DM-ARC
[ZZZ Parvocellular DM-ARC

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

810- T
2.1 T
3;. 6" ”T“ f
"‘3 4: 7
= 7 /
es. _ g g
522 a /
it 0 / %

 

ROSTRAL MIDDLE

Figure 5.1 Comparison of basal numbers of FRA-IR nuclei in non-TH-IR cells in
subdivisions of the rostral, middle and caudal ARC in male rats. After perfusion, brains
were removed, sectioned in the frontal plane (30 um) and immunohistochemically
processed for the detection of FRA-IR and TH-IR in the magnocellular-DM (open
columns), parvocellular-DM (hatched columns) and VL (cross hatched columns)
subdivisions of the rostral, middle and caudal ARC. Columns represent means and
vertical lines 1 SEM from 7-9 animals. *, Values that are significantly different (p <
0.05) from the magnocellular-DM ARC. #, Values that are significantly different (p <
0.05) from the VL ARC.

116

 

C
a", ‘Ay
o¢,- "
"'.. I.-
.0...
, .
:‘Jto‘h
0 e
‘37 7’" o
- O a“. (
\.‘. ....
9.. I ‘
‘ l ”,0

Figure 5.2 Digitized photomicrographs showing increased FRA expression in the middle
ARC afier haloperidol injection in male rats. Fluorescent TH-IR neurons under dark-field
optics (panels A and C) and the same images under bright-field optics (panels B and D,
respectively) showing blackened FRA-IR nuclei. Rats were injected with either 0.3%

tartaric acid (panels A and B; 1 ml/kg; sc), or haloperidol Qianels C and D; 1 mg/kg; sc)
and perfused 3 h later.

117

and greatly increased in the parvocellular-DM ARC afier haloperidol administration.

Numbers of FRA-IR nuclei in non-TH-IR cells were increased 1.5—3 h after
haloperidol injection in the magnocellular-BM (figure 5.3) and parvocellular-DM (figure
5.4) ARC of all rostrocaudal regions in male rats. In contrast to the DM-ARC
subdivisions, numbers of FRA-IR nuclei in non-TH-IR cells of the VL—ARC subdivision
were decreased 3-6 h after haloperidol injection in all rostrocaudal regions (figure 5.5).
There were no rostrocaudal differences in magnitude of response to haloperidol treatment
within any ARC subdivision.

The effects of haloperidol could be mediated by direct action of this drug on
central dopamine receptors, or indirectly via endogenous hyperprolactinemia.
Accordingly, endogenous prolactin was immunoneutralized to determine if the effects of
haloperidol on FRA expression were mediated by prolactin. Prolactin antiserum was
administered at a dose known to block the stimulatory effects of estrogen on TIDA
neuronal activity (Toney et al., 1992b) and suppress plasma prolactin concentrations for
at least 3 h after haloperidol treatment (figure 3.4). As shown in figure 5.6, cc-
administration of prolactin antiserum completely blocked the haloperidol-induced
increase in numbers of FRA-IR nuclei in non-TH-IR cells (at 3 h) in the magnocellular-
DM ARC of all rostrocaudal regions. In contrast, prolactin antiserum merely blunted the
haloperidol-induced increase in the numbers of FRA-IR nuclei in non-TH-IR cells (at 3 h)
in the parvocellular-DM ARC (figure 5.7), and had no effect on the haloperidol-induced
decrease in the numbers of F RA-IR nuclei in non-TH-IR cells in the VL-ARC (figure

5.8). These observations were consistent across all rostrocaudal ARC regions.

118

Magnocellular-DM ARC

 

 

 

50F
:2
T) a: O Rostral
a: 40 .. [:1 Middle
E A Caudal
‘7‘
8 30 a
c:
:5 * ..
Q)
'25 20 ~ \
3
a:
a: :1:
17 =1g
g 10 e
u. A
at e
O l l I l
0.0 1.5 3.0 6.0 12.0
HOURS AFTER HALOPERIDOL

Figure 5.3 Time course effects of ha10peridol on numbers of FRA-IR nuclei in non-TH-
IR cells in the DM—MC subdivision of rostral (circles), middle (squares) and caudal
(triangles) ARC of male rats. Rats were injected with haloperidol (filled symbols; 1
mg/kg; sc) and perfused 1.5, 3, 6 or 12 h later. Zero time controls (open symbols) were
injected with vehicle (0.3% tartaric acid; 1 ml/kg; sc) 3 h prior to perfusion. Symbols
represent means and vertical lines 1 SEM from 7-9 animals. *, Values for haloperidol-
treated rats that are significantly different (p < 0.05) from vehicle-treated controls.

119

Parvocellular-BM ARC

 

 

 

 

160 _
.52 .
78* 140 _ O Rostral
C] .
FE: Middle
m. 120 r- A Caudal
E-i
é: 100 e
O
C
.5 80 -
'8
g 60 ~
:1
ad 40 -
’7‘
g 20 -
[‘4
k 0 .— l I l l I
0.0 1.5 3.0 6.0 12.0
HOURS AFTER HALOPERIDOL

Figure 5.4 Time course effects of haloperidol on numbers of FRA-IR nuclei in non-TH-
IR cells in the parvocellular-BM subdivision of rostral (circles), middle (squares) and
caudal (triangles) ARC of male rats. Rats were injected with haloperidol (filled symbols;
1 mg/kg; sc) and perfused 1.5, 3, 6 or 12 h later. Zero time controls (open symbols) were
injected with vehicle (0.3% tartaric acid; 1 ml/kg; sc) 3 h prior to perfusion. Symbols
represent means and vertical lines 1 SEM from 7-9 animals. *, Values for haloperidol-
treated rats that are significantly different (p < 0.05) from vehicle-treated controls.

120

VL ARC

20 _
O Rostral
E] Middle
A ll\
15 f \\ A Caudal

\

 

# FRA-IR nuclei in non-TH-IR cells

 

 

'. * ‘*//
* a:
0 r . 1 . a
0.0 1.5 3.0 6.0 12.0
HOURS AFTER HALOPERIDOL

Figure 5.5 Time course effects of haloperidol on numbers of FRA-IR nuclei in non-TH-
IR cells in the VL subdivision of rostral (circles), middle (squares) and caudal (triangles)
ARC of male rats. Rats were injected with haloperidol (filled symbols; 1 mg/kg; sc) and
perfused 1.5, 3, 6 or 12 h later. Zero time controls (open symbols) were injected with
vehicle (0.3% tartaric acid; 1 ml/kg; sc) 3 h prior to perfusion. Symbols represent means
and vertical lines 1 SEM from 7-9 animals. ‘, Values for haloperidol-treated rats that are
significantly different (p < 0.05) from vehicle-treated controls.

12]

Magnocellular—DM ARC

I: Control
— Haloperidol
PRL-AB + Haloperidol

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E2

2 l T }

3% 10 - / I T /

a é % é

a? .- a Z a

a é a é
ROSTRAL MIDDLE CAUDAL

Figure 5.6 Effects of prolactin antiserum on numbers of FRA-IR nuclei in non-TH-IR
cells in the magnocellular-DM subdivision of rostral (circles), middle (squares) and
caudal (triangles) ARC of haloperidol-treated male rats. Rats were injected with either
haloperidol (filled columns; 1 mg/kg; so) or its 0.3% tartaric acid vehicle (open columns;
1 ml/kg; so), and with either prolactin antiserum (PRL-AB; hatched columns; 200 ul/rat;
iv) or vehicle (normal rabbit serum, 200 til/rat; iv) 3 h prior to decapitation. Columns
represent means and vertical lines 1 SEM from 7-9 animals. *, Values that are
significantly different (p < 0.05) from vehicle-treated controls.

122

Parvocellular-BM ARC

:1 Control
_ Haloperidol

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘2‘ 80 F PRL-AB + Haloperidol *

a 60 - * #
E 40 ~ *f; l
'3 , /
a? zo~ ’2‘” % Z
a % Z a
r 0 A fl 4 fl %

ROSTRAL MIDDLE CAUDAL

Figure 5.7 Effects of prolactin antiserum on numbers of FRA-IR nuclei in non-TH-IR
cells in the parvocellular-DM subdivision of rostral (circles), middle (squares) and caudal
(triangles) ARC of haloperidol-treated male rats. Rats were injected with either
haloperidol (filled columns; 1 mg/kg; so) or its 0.3% tartaric acid vehicle (open columns;
1 ml/kg; so), and with either prolactin antiserum (PRL-AB; hatched columns; 200 ul/rat;
iv) or vehicle (normal rabbit serum, 200 ul/rat; iv) 3 h prior to decapitation. Columns
represent means and vertical lines 1 SEM from 7-9 animals. *, Values that are
significantly different (p < 0.05) from vehicle-treated controls. #, Values that are
significantly different (p < 0.05) from haloperidol-treated controls.

123

VL ARC

[:1 Control
- Haloperidol

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

£2

*7 l

[E 8 ~ T

4- / / /

E f % ¢

‘1: 0 % /. A
ROSTRAL MIDDLE CAUDAL

Figure 5.8 Effects of prolactin antiserum on numbers of FRA-IR nuclei in non-TH-IR
cells in the VL subdivision of rostral (circles), middle (squares) and caudal (triangles)
ARC of haloperidol-treated male rats. Rats were injected with either haloperidol (filled
columns; 1 mg/kg; sc) or its 0.3% tartaric acid vehicle (open columns; 1 ml/kg; so), and
with either prolactin antiserum (PRL-AB; hatched columns; 200 ul/rat; iv) or vehicle
(normal rabbit serum, 200 ul/rat; iv) 3 h prior to decapitation. Columns represent means
and vertical lines 1 SEM from 7-9 animals. *, Values that are significantly different (p <
0.05) from vehicle-treated controls.

124

Central administration of prolactin transiently increased numbers of F RA-IR
nuclei in non-TH-IR cells at 3 h in the magnocellular (figure 5.9) and parvocellular
(figure 5.10) subdivisions of the DM-ARC of all rostrocaudal regions in male rats. In the
VL-ARC, prolactin increased numbers of FRA-IR nuclei in non-TH-IR cells 1.5—3 h after
injection in all rostrocaudal regions (figure 5.11). There were no rostrocaudal differences

in magnitude of response to prolactin treatment within any ARC subdivision.

125

Magnocellular—DM ARC

 

 

 

100 —
.59,
T3 * O Rostral

E] M'

a 80 _ rddle
I. A Caudal
[T
8 6O -
G
.S
'5
T, 40 -
5
C“
E
I
E 20 ~
LL!
111:

O I l l l I

0.0 1.5 3.0 6.0 12.0
HOURS AFTER PROLACTIN

Figure 5.9 Time course effects of prolactin on numbers of FRA-IR nuclei in non-TH-IR
cells in the magnocellular-BM subdivision of rostral (circles), middle (squares) and
caudal (triangles) ARC of male rats. Rats were injected with prolactin (filled symbols; 10
jig/rat; icv) and perfused 1.5, 3, 6 or 12 h later. Zero time controls (open symbols) were
injected with vehicle (distilled water; 3 til/rat; icv) 6 h prior to perfusion. Symbols
represent means and vertical lines 1 SEM from 6-8 animals. *, Values for prolactin-
treated rats that are significantly different (p < 0.05) from vehicle-treated controls.

126

Parvocellular-BM ARC

4° F >1:
0 Rostral
[3 Middle
30 ~ A Caudal
2O -

# FRA-IR nuclei in non-TH-IR cells

 

l J I I

0.0 1.5 3.0 6.0 12.0
HOURS AFTER PROLACTIN

Figure 5.10 Time course effects of prolactin on numbers of FRA-IR nuclei in non-TH-
IR cells in the parvocellular-DM subdivision of rostral (circles), middle (squares) and
caudal (triangles) ARC of male rats. Rats were injected with prolactin (filled symbols; 10
jig/rat; icv) and perfused 1.5, 3, 6 or 12 h later. Zero time controls (open symbols) were
injected with vehicle (distilled water; 3 ul/rat; icv) 6 h prior to perfusion. Symbols
represent means and vertical lines 1 SEM from 6-8 animals. *, Values for prolactin-
treated rats that are significantly different (p < 0.05) from vehicle-treated controls.

l27

VL ARC

 

 

 

 

60 ~ :1:
.542 3'?
T) " O Rostral
O .
a 50 - [:1 Middle
m- A Caudal
E7 40 A
r:
O
c:
.8 30 ~
E
0
a 20 _
ed
"T‘
g 10 ~
Ln
1121:
O J l 1 I l
0.0 1.5 3.0 6.0 12.0
HOURS AFTER PROLACTIN

Figure 5.11 Time course effects of prolactin on numbers of FRA-IR nuclei in non-TH-
IR cells in the VL subdivision of rostral (circles), middle (squares) and caudal (triangles)
ARC of male rats. Rats were injected with prolactin (filled symbols; 10 ug/rat; icv) and
perfused 1.5, 3, 6 or 12 h later. Zero time controls (open symbols) were injected with
vehicle (distilled water; 3 til/rat; icv) 6 h prior to perfirsion. Symbols represent means and
vertical lines 1 SEM from 6-8 animals. *, Values for prolactin-treated rats that are
significantly different (p < 0.05) from vehicle-treated controls.

128

C. DISCUSSION:

While it is well known that TIDA neurons regulate prolactin secretion, little
information is available regarding the feedback mechanism underlying the delayed
inductive activation of these neurons by prolactin. In the present study a serendipitous
observation was further investigated via neural activity mapping with inducible
transcription factors.

Neural activity mapping with inducible transcription factors has led to the
localization and identification of neuronal populations within specific brain regions that
are responsive to a variety of stimuli. While this approach is widely utilized and
generally accepted there are two major considerations in this type of study: cell type
expression specificity and stimulus coupling uncertainty (Chaudhuri, 1997; Harris, 1998).
It is now known that not all cells produce all transcription factors, hence the phenomenon
termed cell type expression specificity. Cell type expression specificity demands careful
interpretation of negative results. Another complexity of neural activity mapping using
inducible transcription factors is the convergence/redundancy of multiple signaling
pathways which leads to stimulus-coupling uncertainty. Stimulus-coupling uncertainty
can make a causal link between changes in gene expression and a stimulus difficult to
establish and highlights the importance of appropriate controls.

In the present study, only positive results are reported, and it was known a priori
that the cells were capable of FRA expression. Furthermore, specific causal relationships
are neither tested nor reported, and similarly-handled controls were employed. In this
way, neural activity mapping via immunohistochemical detection of inducible

transcription factors (i.e. FRA proteins) was utilized to determine if prolactin regulates

129

FRA expression in non-TH-IR cells in the DM and VL subdivisions of the ARC in male
rats.

ARC subdivisions had low basal expression of FRA-IR in non-TH-IR cells (5-15
cells/region/section) across all ARC subdivisions and rostrocaudal regions in male rats.
The lower basal numbers of FRA-IR cells in the parvocellular-DM ARC most likely
relates to differences in relative size of ARC subdivisions and hence total cell numbers.
Since AP-l sites are inherent to many genes including “house-keeping genes” these
numbers probably reflect their low level of expression. Naturally, any changes in FRA
expression due to haloperidol-induced hyperprolactinemia must be detected over this
background noise.

Haloperidol rapidly stimulates prolactin secretion via blockade of D2 receptors on
anterior pituitary lactotrophs, and the prolonged hyperprolactinemia that follows feeds
back to stimulate TIDA neurons (Moore and Lookingland, 1995). Haloperidol has no
direct acute effects on TIDA neurons and its latent stimulatory effects can be blocked by
hypophysectomy (Gudelsky et al., 1977) or prolactin antiserum (Gudelsky and Porter,
1980; Van Loon et al., 1983; Hentschel et al, 2000a), demonstrating that the stimulatory
effects are mediated by prolactin. While measuring prolactin-induced changes in F RA
expression in TIDA neurons, it appeared that changes in FRA expression in non-TH-IR
cells of the ARC seemed to temporally precede and/or coincide with those in TIDA
neurons. Furthermore, these cells were localized within the ARC, consistent with the
complete deafferentation study that demonstrated the haloperidol-induced, prolactin-
mediated increase in dopamine turnover in the median eminence results from actions of

prolactin on neurons within the mediobasal hypothalamus (Gudelsky et al., 1977). Given

130

 

 

the 6-12 h latency that precedes the inductive activation of TIDA neurons (Demarest et
al., 1984; 1986), it is plausible that these cells mediate the effects of prolactin on TIDA
neurons. Logically, the first step in this investigation was to verify the purported
prolactin-induced changes in immediate early gene expression. The working hypothesis
was that prolactin regulates FRA expression in subpopulations of non-TH-IR cells in the
ARC.

Consistent with this hypothesis, the results of the present study reveal that in male
rats prolactin differentially modulates F RA expression in non-TH-IR cells located in both
DM and VL subdivisions of the ARC. In the magnocellular-BM ARC, both haloperidol-
induced endogenous hyperprolactinemia and central administration of prolactin cause
similar delayed increases in numbers of non-TH-IR cells expressing FRA with peak
responses occurring by 1.5-3 h. As described in Chapter 4, the (two-fold) greater response
in rats treated with icv prolactin compared to haloperidol is likely dose- and route of
administration-related.

The effects of haloperidol could be mediated by direct action of this drug on
central dopamine receptors, or indirectly via endogenous hyperprolactinemia.
Preliminary experiments examined the ability of PRL-AB to block the effects of
haloperidol on F RA expression in male rats. Prolactin antiserum (200ul/rat; iv)
decreased plasma prolactin to non-detectable levels as measured by Nb2 lymphoma
bioassay (Canon et al., 1991) for up to 3 h in haloperidol-treated male rats.
Radioimmunoasssay analyses of binding specificity for this antiserum revealed that PRL-
AB used in these experiments was selective for prolactin, having no affinity for

luteinizing hormone or growth hormone (Hentschel et al., 2000a). In the present studies,

13]

coadministration of PRL-AB completely blocked the haloperidol-induced changes in
FRA expression expected at 3 h in non-TH-IR cells in the magnocellular-DM ARC,
indicating that these changes are prolactin-dependent. Furthermore, the peak temporal
changes in FRA in non-TH-IR cells of the magnocellular-DM ARC occur 3 h afier
prolactin treatment, and therefore precede or occur concurrently with the changes
observed for regionally matched TIDA neurons (3-6 h; Chapter 4). It is possible that
chemically unidentified cells of the magnocellular-BM ARC may have a role in
mediating the delayed inductive activation of TIDA neurons by prolactin.

As seen in the magnocellular-DM ARC, both haloperidol-induced endogenous
hyperprolactinemia and central administration of prolactin increases the numbers of non-
TH-IR cells expressing FRA in the parvocellular-DM ARC with peak responses
occurring by 1.5-3 h. However, coadministration of PRL-AB merely blunted the
haloperidol-induced changes in FRA expression at 3 h in non-TH—IR cells in the
parvocellular-BM ARC, indicating that these changes are only partially prolactin-
dependent. Additionally, haloperidol-treated rats demonstrate a 4-fold greater response
than prolactin-treated rats which is likely due to D2 dopamine receptor mediated effects in
the parvocellular-BM ARC. These results suggest that the cells of the parvocellular-BM
ARC are tonically modulated through a D2 dopamine receptor-mediated mechanism,
unlike the cells of the magnocellular-DM ARC.

In the VL-ARC, the onset of the stimulatory effects of prolactin on F RA
expression in non-TH-IR cells precedes that of both regions of the DM-ARC.
Considering that transcriptional activation of c-fos and related immediate early genes

occurs within minutes, with the accumulation of nuclear protein reaching peak levels on

132

average about 90—120 min later (Harris, 1998), temporal differences in the onset of
prolactin-induced changes in expression of FRA in the DM- and VL-ARC suggest two
possibilities, either the mechanisms by which prolactin effects these cells are different
(i.e. acute direct versus delayed secondary actions), or those non-TH cells of the VL-ARC
are more sensitive to prolactin than those of the DM-ARC. Differences in sensitivities of
these cells to prolactin could be due to differences in expression of prolactin receptor
isoforms and their respective differential densities. Temporal changes in FRA in non-TH-
IR cells of the VL-ARC are coincident with the changes in F RA expression within TIDA
neurons of the VL-ARC. However, since the changes in FRA expression in both TH-IR
and non-TH-IR cells alike of the VL-ARC precede those of all other ARC subdivisions, it
is possible that these cells may initiate the prolactin-induced inductive activation of TIDA
neurons.

In male rats haloperidol suppresses FRA expression in non-TH-IR cells in the VL-
ARC and blocks the stimulatory effects of prolactin on these cells via a prolactin-
independent mechanism. Indeed, immunoneutralization of circulating prolactin in
haloperidol-treated rats has no effect on the ability of this D2 dopamine antagonist to
decrease the number of non-TH-IR cells expressing FRA in the VL-ARC, in contrast to
the magnocellular-DM ARC. The inhibitory effect of haloperidol on F RA expression in
non-TH-IR cells of the VL-ARC could due to blockade of D2 dopamine receptor-
mediated tonic regulation of immediate early gene expression in these neurons that is not
present in cells of the magnocellular-BM ARC. The significance of this finding remains

to be elucidated.

133

In conclusion, these results reveal that prolactin indeed regulates immediate early
gene expression in non-TH-IR cells in the ARC, and that the cells of different ARC
subdivisions are temporally differentially responsive to prolactin administration (i.e.
prolactin initially stimulates transient expression of FRA in cells of the VL-ARC,
followed by FRA expression in cells of the DM-ARC). In the chapters to follow the

identity of these prolactin-responsive, non-TH-IR cells is investigated.

I34

6. ROLE OF NEUROTENSIN IN PROLACTIN-
INDUCED ACTIVATION OF TIDA NEURONS

A. INTRODUCTION

The feedback activation of TIDA neurons by prolactin consists of tonic and
inductive components (Demarest et al., 1984; 1986). The tonic component of prolactin
feedback, observed exclusively in females, determines the basal rate of prolactin secretion
by regulating short-term changes in TIDA neuronal activity in response to acute changes
in concentrations of circulating prolactin. The set point of this neuroendocrine feedback
system is higher in females, due in part, to higher circulating prolactin levels caused by a
direct stimulatory effect of estrogen on prolactin secretion (Moore and Lookingland,
1995)

The inductive component of prolactin feedback, common to both genders,
involves a change in the capacity of TIDA neurons to respond to prolonged periods (3-6 h
or more) of hyperprolactinemia (Moore et al., 1987a; Hentschel et al., 2000a), and does
not occur until 6-12 h following systemic (Demarest et al., 1984; Demarest et al., 1986;
Hokfelt and Fuxe, 1972; Selmanoff, 1981) or central (Annunziato and Moore, 1978)
administration of exogenous prolactin, or after experimental manipulations which result
in endogenous hyperprolactinemia (Moore et al., 1985). In Chapter 4 we proposed that
prolactin induces long-term changes in gene expression that result in increased capacity
of TIDA neurons to synthesize and release dopamine, and demonstrated changes in
immediate early gene expression in TIDA neurons 3-6 h after prolactin administration. In
Chapter 5 neural activity mapping with inducible transcription factors led to the novel

identification of non-TH-IR cells in the ARC that are prolactin-responsive. Furthermore,

135

the changes in immediate early gene expression induced by prolactin in these cells
temporally precedes or coincides with those of TIDA neurons. The localization of these
cells in the ARC is consistent with mediobasal hypothalamic deafferentation studies
(Gudelsky et al., 1978), and suggests that these unidentified cells could be mediators of
the delayed inductive activation of TIDA neurons by prolactin.

Certain aspects of prolactin-induced activation of TIDA neurons suggest a role for
a peptide mediator. For example, there is temporal dependence in the inductive
component of prolactin feedback, i.e. activation of TIDA neurons does not occur until 6-
12 h following hyperprolactinemia. Furthermore, the inductive actions of prolactin are
protein synthesis-dependent, i.e. activation of TIDA neurons by prolactin can be blocked
by cyclohexirnide (Johnston et al., 1980). These characteristics suggest that the delay in
the ability of prolactin to activate TIDA neurons may be due to the time required for de
novo synthesis, transport and release of a peptide neurotransmitter.

Neurotensin is a tridecapeptide neurotransmitter which fits several criteria as a
possible mediator of the effects of prolactin on TIDA neurons. For example, neurotensin-
containing neurons are located in the ARC (Emson et al., 1982; Merchanthaler and
Lennard, 1991) and TIDA neurons encode mRNA for high-affinity neurotensin receptors
(Alexander, 1997). Furthermore, administration of neurotensin produces dose-dependent
activation of TIDA neurons both in vitro (Herbison et al., 1986; Lin and Pan, 1993) and
in viva (Widerlov et al., 1982; Gudelsky et al., 1989; Fan et al., 1992).

The purpose of the studies described in this chapter is to determine if neurotensin
mediates the delayed inductive activation of TIDA neurons by prolactin. If this is the

case, then the non-peptide neurotensin receptor antagonist SR-48692 (Gully et al., 1993)

I36

will block the stimulatory effects of prolactin on TIDA neuronal activity as estimated by
measuring concentrations of DOPAC in termini of these neurons in the median eminence
(Lookingland et al., 1987a). In these studies TIDA neuronal responses will be compared
to those of NSDA and MLDA neurons.

While NSDA and MLDA systems are not prolactin-responsive, these neurons are
activated by haloperidol. Furthermore, there are the many associations between NSDA
and MLDA neurons and neurotensin. For example, neurotensin receptors are localized
on perikarya and termini of NSDA and MLDA neurons (Young and Kuhar, 1981) and
neurotensin administration activates NSDA (Pinnock, 1985) and MLDA neurons
(Widerlov et al., 1982) and facilitates basal and potassium-induced dopamine release in
these neurons (Okuma and Osumi, 1982). In addition, neurotensin is often colocalized
with dopamine in perikarya of NSDA and MLDA neurons (Hokfelt et al., 1984). Hence,
this study will consider the effects of a neurotensin receptor antagonist on basal and
activated NSDA and MLDA neurons by measuring concentrations of DOPAC in the
striatum and nucleus accumbens, regions containing termini of these neurons,

respectively. These studies were performed in male and diestrous female rats.

137

B. RESULTS

SR-48692 is a selective non-peptide neurotensin receptor antagonist which
competitively inhibits binding of ['251]NT to high-affinity neurotensin receptors in vitro
(Gully et al., 1993), and in vivo reverses turning behavior induced by intrastriatal
injection of neurotensin in both mice and rats (Gully et al., 1993; Steinburg et al., 1994).
SR-48692 is devoid of intrinsic agonist activity in vivo, and significant antagonistic
effects have been demonstrated by 30 min (with maximal effects observed between 1 and
2 h) following peripheral administration of doses between 80 jig/kg (p.o.) (Gully et al.,
1993) and 1,000 ug/kg (i.p.; Castagliuolo et al., 1996). As shown in table 6.1, l h
following its systemic injection, SR-48692 (1,000 ug/kg; i.p.) had no effect on DOPAC
or dopamine concentrations in the median eminence, striatum or nucleus accumbens of
either male or female rats.

Haloperidol is a D2 dopamine receptor antagonist which induces
hyperprolactinemia (figures 3.1 or 4.3, top panels) via disinhibition of dopamine
suppression of prolactin secretion from adenohypophysial lactotrophs; the resultant
hyperprolactinemia, in turn, activates TIDA neurons (Moore and Lookingland, 1995). As
shown in figure 6.1, 12 h afier treatment of male and female rats with haloperidol (1
mg/kg; s.c.) DOPAC concentrations in the median eminence were significantly increased
as compared with vehicle-treated controls. The stimulatory effect of haloperidol on
median eminence DOPAC concentrations was reversed by SR-48692 (1,000 jig/kg, i.p.; 1
h) in both male and female rats. Injections of SR-48692 (10 - 1,000 jig/kg, i.p.; 1 h) had
no effect on dopamine concentrations in the median eminence of either vehicle- or

haloperidol-treated male and female rats.

138

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE 6.1 Effects of SR-48692 on DOPAC and dopamine concentrations in median
eminence, striatum and nucleus accumbens of male and female rats.
MEDIAN EMINENCE
GROUP GENDER DOPAC :t SEM DA :h SEM N
Control Male 15.2 d: 1.0 125.8 :1: 3.8 7
SR-48692 Male 17.2 i 1.4 124.5 :h 2.5 9
Control Female 18.0 3: 1.6 100.9 i 6.7 9
SR—48692 Female 18.6 :t 1.3 105.7 at 5.6 11
STRIATUM
GROUP GENDER DOPAC :1: SEM DA :1: SEM N
Control Male 27.3 :h 1.5 102.5 i 5.8 8
SR-48692 Male 30.0 i 2.2 104.5 :h 3.3 9
Control Female 24.3 d: 0.8 105.2 :1: 6.2 9
SR-48692 Female 27.7 i 1.1 106.9 :t 5.9 9
NUCLEUS ACCUMBENS
GROUP GENDER DOPAC :t SEM DA :1: SEM N
Control Male 26.0 d: 1.1 100.3 :1: 4.8 7
SR-48692 Male 26.1 :t 0.9 105.8 i 4.9 8
Control Female 23.2 at 1.4 91.4 at 6.5 10
SR-48692 Female 22.5 i 1.4 87.2 :1: 5.1 10

 

 

 

Rats were injected with either SR-48692 (1,000 jig/kg; i.p.) or its 10% DMSO in 0.9%
saline vehicle (4 ml/kg; i.p.) 1 h prior to decapitation. Values represent means i 1 SEM
of DOPAC and dopamine concentrations (ng/mg protein) in the median eminence,
striatum and nucleus accumbens.

139

275:8 385.2629 Soc Acosta xufiomiwfi 8m 35 38 ngbAoutoaoB: com 82? ”mod v a ... .38 CEO—M:
25:8 98 Chm—d £9: mo 85580 562: 05 E €685 wEEE 3235588 2:838 mo 2mm _ H £82: 53058 83m> ”ZOEOm
355:5 5:58 05 E @805 9:35 3235588 Oak—on mo 2mm _ 8:: ~89?» was 3.88 232%: £2:on i2. dosflamoov
8 SE a a 2.2 Mmesa 3 2029 2:9. $3 a 82m $2 ...: s 2: new: 83 3 2: .o: 8335 .220 5_; 2a 55388
8 SE. a N_ Ada %th _ ”mass—8 :33 26.29» Box 2353 $3.6 m: 8 69m ”wimp: ~ $5228 3:5 Become—m; 856 5_; 88%: 225
Sam .38 038$ 98 2S: BHSboncomo—m: me 3550 $6.2: 05 5 3235588 ogaon Ba 0559 :o mammal—m we soobm fie 83E

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  
 
  

m 3H 32 we»: 82 o 32.2: mew: 82
2 33.? mi: 02 a @332 ER: 8.
a $3.02 9%: 2 w 0332 $332
a E H mos Eh: o w 3.. H 32 we»: 0
m 0.0 H 2.: 6.38 o 3 H 22 .828
a 2% H 252 Naewvim mmoa : 2% a 72m: 8335 mmoo
mama Egon Rams mzuzfiom
Ema 833% $8
82 2: 200 82 2:
o o
. m . m
G G
. 2 O . o_ 0
.d .d
V V
..u U
F ..M
. 8 w . om w
mw me
. mm . mm
* * _ovtoqo_am I
* . cm 20Eo> H . cm

 

 

aflgmm mfl<2

140

Haloperidol activates NSDA and MLDA neurons via blockade of dopaminergic
autoreceptors (Nowycky and Roth, 1978) in the striatum and nucleus accumbens,
respectively. As shown in figure 6.2, 12 h after treatment of male rats with haloperidol
(1 mg/kg; s.c.) DOPAC concentrations in the striatum and nucleus accumbens were
significantly increased as compared with vehicle-treated controls. The stimulatory effects
of haloperidol on striatum and nucleus accumbens DOPAC concentrations were not
reversed by SR-48692 (10 - 1,000 rig/kg, i.p.; 1 h), and injections of SR—48692 had no
effect on dopamine concentrations in the striatum or nucleus accumbens of either vehicle-
or haloperidol-treated male rats.

Results depicted in figure 6.] were obtained using an experimental model which
employed haloperidol to induce endogenous hyperprolactinemia as a means of indirectly
stimulating TIDA neurons. One possible explanation for the dose-related reversal of
haloperidol-induced DOPAC concentrations by SR-48692 is that this compound prevents
the ability of haloperidol to stimulate prolactin secretion. To test for this, the effects of
SR-48692 (1,000 jig/kg, i.p.; 1 h) on haloperidol-induced hyperprolactinemia were
investigated in male and female rats. As shown in figure 6.3, haloperidol (1 mg/kg, s.c.;
1 h) significantly increased plasma prolactin concentrations in both male and female rats,
and SR-48692 (1 ,000 ug/kg, i.p.; 1 h) had no effect on plasma prolactin concentrations in
either vehicle- or haloperidol-treated male or female rats.

As shown in figure 6.4, 12 h following central administration of prolactin (10
rig/rat, i.c.v.), concentrations of DOPAC in the median eminence were increased in both
male and female rats, and these effects were reversed by administration of SR-48692

(1,000 jig/kg, i.p.; 1 h). Injection of SR-48692 had no effect on dopamine concentrations

141

.3828 3385.225? 20¢ “Shag—o bananas Ba 35 38 38:._ovtomo_£ com 82? ”mod v a ..., .AEOEV 353233
363:: 28 C235 EBmEm 05 2 $2205 959: 3.2835250 05238 mo 2mm _ H 382: E8052 8:13 ”EOCOm .338:on
353:: 28 EBmEm 05 E A5805 wE\w5 36588856 Hus—0Q .«0 2mm _ mos: 329.3 98 €on 258.52 3.8200 “no... doufiiwoov
2 58 g _ i: m9:5 s 2029 253 22 a 82m .82 3 s E: 381 83 8 O2 5: 8835. use 5_; as £828.88
8 SE a 2 69m mmvSE _ ”3:528 :38 22:? Box 2.3.5 {and 3m 8 Add Mme—RE _ ”355—8 3:5 Butane—2 35.0 59> @900? 225 3g
.32 22: 838348536: no 33238 3.22.: Ea E3325 2: 2 2283588 02238 28 Gammon co 82:53 me 33cm N6 «Sui

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
  
 
  

 

m 3.2.8 3%: 82 m 3H $2 3%: 82
A 3 H 28 33: 2: m 3H 3: win 82
w EH05 3:8 2 w 3H3: 3:8 2
w on H 0.8 33: o a 3 H 32 3%: c
A. ms H v.0: 32:00 x Wm H om: 65:00
: Ema H Ewe/H Noowvbmw mmOQ : Ema H 24% NoowaHm MmOQ
@585 Egon A885 mag/Egon
Age 8838 80m
82 o2 zoo 82 2: 28
2 3
. Om . On
Q G
H 0 A 0
. 3. W . a. W
O 3
\u; \u/
. 8 fl . 8 fl
w w
arm at
* * . mm. . wk.
*
*
1 oo . co

 

38822 I

mzmmZDUU< mDm—AUDZ 2225 U EDHSMPm

142

2888 385.202? 52m 25% agar—mm..." 8a 35 $8 3235
-8583 5 829 ”3.8 v a . “was“ :A 58 3:83 835 28855 @835 we saw _ 88 35> 2a 885 8828. 8.5.8
83881 3 SE a _ G: mmass s 2033 8:9. $2. a 82m $2 an s 2: 3m: 88.: Naowvxm has 55 8a 85888
3 88 @158 a 2 8 Easy .. _ 8 vaSe _ Mm2:28 883 2029 28 255 82 m: 5 A3. ”flue _ $538 3:8 85qu
was 5? 860.8 a»; sum .32 033. ea 29: 38:._8cao§ a §u§§=8 .58.qu sea 8 8835 «o 388m 3 8&8

 

 

 

 

 

 

 

 

 

 

 

 

 

 
 
 

 

 

NoewYMm AOMHZOU NmemYMm ADM—H200
o _ _ o
H 4 2 - 4 2
H .d .d
. 8 w A 8 w
1
.8 W . 8 V
.....m A;
. 8 W) * * . 8 W
m m
* . 8 Evie—um I . . 8
... A 2 2029 NH 8 a

”ma—aazfim Hd<2

143

.2988 388.0882 :58 8888 888838 2a 85 38 8388.88.28 .2
8:3, ”8.8 v a . .32 9:28 08:5. 85 Ema one 8 855:8 582.. 05 a 8088 885 9888880 8888 8 2mm 2 a 88.:
gag 88> ”22.8m— .8855 88:. as a 332.. 9:85 9589805 9&8 8 2mm _ $8 38> 85 885 83282 8.580
22. 88588.88 3 88 a _ 8.2 ”9:? 3 2083 283 $3 a 8:5 .82 2 8 Q: “981 88.: 8835 888 5_; Ba 83888
9 88 a 2 «>3 85.. m M8828 83 0823 883 8:88 2 8 2.3 we? 2 “858 3:2 8888 E 858 5_; 38.8 22»
sum .38 0858 ea one 3838228 8 8885 882: 2: a 8038888 08:88.8 85 938 8 882m 8 388m «.8 2&2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
 
 
 

2 m.m H 93 «momYMm + 5828.— » n6 H Qua 33.1% + quad—Em
3 v.0 H v.3 Nowmvbmm + 2020> n ma. H 3% «331% + £2.85
2 1n H fium £03“; + $8305 a «6 H how 2023» + 5820.5
2 En H Qua o_o£o> + o_oEo> 5 WV H 5.3 o_oEu> + o_oEo>
: 2mm H Z<m2 SOME a 2mm H 24m: SOMO
mans 8:828 mama 28828
8838 88.28 8838 86.28
c o
. m . m
. o— .d . 3 ..Va
w a
m P m,
H . 2 fl . 2 fl
2 m 2 m
6 G
. 0N . 0N
* Egon—0.5 I *
mm 3029» Fllll. % nu

 

 

magma Hag

I44

in the median eminence of either vehicle- or prolactin-treated male and female rats. As
shown in figure 6.5, DOPAC and dopamine concentrations in the striatum and nucleus
accumbens were not effected by either central administration of prolactin (10 pg/rat,

i.c.v., 12 h), or systemic injection of SR—48692 (1,000 ug/kg, i.p.; 1 h) in male rats.

145

65095 83:88“ mac—cs: Ea Emd 53818 05 E 380$ map—35 8093588 03833 mo 2mm _ H 888 882%.. mos—Er
”SPF—Cm 8:38:88 man—oz: 85 838.5" 05 5 £805 mafia «8888:3800 U<mon mo Sum _ 8:: Hana?» 28 E88 “confine mafia—cu
H.5... 888888 3 88 a _ 2: m9:8 8 0888 288 .888 8 82a .82 a 8 Ea 381 28.: 83.1% 888 88, 88 8888888
8 8.5“ a 2 :68 38:: m “8533 :25 £033 “883 3:96 a: co A.>.o._ £831 2 ”mafia—8 3:5 £63on “8 556 5m? 8883 0.83 88M
.38 29: 8383.:u8—8q mo 835588 3223 v.3 8.58ch 05 5 80388880 2:832. 98 U<mOQ no Naome—m mo 300mm m6 95$...—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
 

 

 

 

 

 

 

 

 

 

 
 

 

 

 

» «N H 3”: «$345 + 58205 2 in H mdfl noomvém + 58.05
m o6 H a2: «$238 + o_oEo> o «d H «62 mommYMm + 202;
a _.v H Eng 2o£o> + Ewan—ohm m m6 H NH: 2028/ + Egon—ohm
w m4 H v.02 o_oEo> + 2023» m ad H :2 o_oEo> + o_oEo>
: 2mm H 728% BOMO : Ema H 7282 SOMO
NwEhav %mOQ 8:35 Eamon
NaowYMm AOMHZOU NaowYMm 405200
c o
L O— . A:
m . m
V V
. 8 b 8 m
m m
w w
an 8
L OM ( H F l on (
H. L
. ov . ov

 

 

flow—Gun.— I
o—oEo> H

mZHmEDUU< mDflAUDZ SPHERE

146

C. DISCUSSION

The selective neurotensin receptor antagonist SR-48692 binds both high- and low-
affinity forms of the neurotensin receptor (Gully et al., 1993; Betancur et al., 1995; Le et
al., 1996) and readily crosses the blood-brain barrier following its systemic
administration, reaching pharmacologically active levels between 30 min and 2-3 h
following doses of between 80 and 1,000 ug/kg (Gully et al., 1993; Castagliuolo et al.,
1996). The results of the present study demonstrate that administration of SR-48692
(1,000 ug/kg; i.p.) has no effect on basal DOPAC and dopamine concentrations in the
median eminence of either male or female rats. This lack of effect was not due to the
dose or time of administration of SR-48692 since using the same parameters this
compound was effective in blocking elevated median eminence DOPAC concentrations
in both haloperidol- and prolactin-treated rats. Rather, the inability of this compound to
alter median eminence DOPAC and dopamine concentrations is probably due to the
absence of intrinsic endogenous agonistic activity at neurotensin receptors under basal
conditions when the activity of TIDA neurons in females is tonically regulated by acute
changes in circulating prolactin.

On the other hand, blockade of neurotensin receptors with SR—48692 prevents the
delayed stimulatory effects of both haloperidol and prolactin on median eminence
DOPAC concentrations, without altering the ability of haloperidol to increase prolactin
secretion. Taken together, these results suggest that hyperprolactinemia stimulates the
release of endogenous neurotensin, which acts via neurotensin receptors to activate TIDA

neurons. Hence the inductive activation of TIDA neurons by prolactin in both males and

147

females is neurotensin-dependent and, in this respect, different from the mechanism
underlying the tonic regulation of these neurons in females rats.

The inductive component of prolactin feedback, common to both genders,
involves a change in the capacity of TIDA neurons to respond to prolonged periods (3-6 h
or more; Hentschel et al., 2000a) of hyperprolactinemia (Moore et al., 1987a) and is
characterized by long latency. Prolactin activates TIDA neurons in such a manner that
the response is protein synthesis dependent (Johnston et al., 1980). It is possible that this
protein synthesis-dependent period corresponds to the induction and de novo synthesis of
neurotensin. In the present study, blockade of neurotensin receptors 11 h after
haloperidol-induced hyperprolactinemia or central prolactin administration prevented the
activation of TIDA neurons, a temporal period appropriate in duration for de novo
synthesis and release of a neurotensin (Rostene and Alexander, 1997).

The source of neurotensin that mediates the stimulatory effect of prolactin on
TIDA neurons is located within the mediobasal hypothalamus since complete surgical
deafferentation of this region from the rest of the brain does not alter the ability of
hyperprolactinemia to activate TIDA neurons (Gudelsky et al., 1978). Since neurotensin
mRN A and neurotensin-IR have been localized to neurons within the ARC (including
those containing TH-IR; Rostene and Alexander, 1997), prolactin-induced neurotensin
synthesis and release could occur in afferent neuropeptidergic neurons (Merchanthaler
and Lennard, 1991; Marcos et al., 1996a) and/or in subpopulations of TIDA neurons
themselves (Hokfelt et al., 1984; Bachelet et al., 1997). Furthermore, it is plausible that

the acutely prolactin-responsive, chemically unidentified cells in the ARC (highlighted in

148

Chapter 5) are neurotensinergic neurons involved in mediating the delayed inductive
activation of TIDA neurons.

The finding that SR-48692 has no effect on basal activity of NSDA and MLDA
neurons is consistent with a previous report (Pugsley et al., 1995) and is not dose- or
time-dependent since injection of SR-48692 (8O ug/kg; i.p.; 0.5 - 2 h) reversed the
contralateral turning behavior induced by a 10 pg intrastriatal injection of neurotensin
(Gully et al., 1993). These results suggest that (like TIDA neurons) endogenous
neurotensin does not participate in regulation of the basal activity of NSDA or MLDA
neurons in male and female rats.

Haloperidol is reported to effect NSDA and MLDA neuronal activity in a variety
of ways. The predominant effect is stimulatory via blockade of inhibitory presynaptic D2
autoreceptors on terminals in the striatum and nucleus accumbens, respectively (N owycky
and Roth, 197 8). Haloperidol may also indirectly stimulate these systems through
disinhibition as follows. D2 dopamine receptor antagonism blocks the tonic inhibition of
neurotensin neurons (Fuxe et al., 1992; Tanganelli et al., 1993; Brun et al., 1995) thereby
increasing activity and release of neurotensin in the striatum and nucleus accumbens (as
evidenced by increased extracellular neurotensin content in microdialysates in terminal
regions; Wagstaff et al., 1996). Neurotensin, in turn, may bind its receptors localized on
terminals of NSDA and MLDA neurons (Quirion et al., 1982; Quirion, 1983; Herve et al.,
1986; Kitabgi et al., 1987; Delle Donne et al., 1996) to decrease the sensitivity of D2
autoreceptors to dopamine (Fuxe et al., 1992; Tanganelli et al., 1989) through an
undefined mechanism. Consequently, robust activation is the overall effect of haloperidol

on NSDA and MLDA systems.

149

In the present study, neurotensin receptor blockade by SR-48692 had no effect on
NSDA or MLDA neurons. It is possible that SR-48692 had no effect on haloperidol-
induced activation of NSDA and MLDA neurons because the effects of endogenous
neurotensin are mediated through modulation of the binding constant of NSDA and
MLDA autoreceptors which are already blocked by haloperidol or have a differential time
course. More likely, regulation of NSDA and MLDA neuronal activity is predominantly
mediated via D2 autoreceptors and neurotensin receptors do not have a significant role.

In conclusion, the results of the present studies are consistent with the hypothesis
that neurotensin mediates the delayed inductive activation of TIDA neurons by prolactin
in both male and female rats, whereas the tonic activation of TIDA neurons by circulating
prolactin in female rats is neurotensin-independent. It remains plausible that the acutely
prolactin-responsive, chemically unidentified cells in the ARC (highlighted in Chapter 5)
are neurotensinergic neurons which mediate the delayed inductive activation of TIDA
neurons in male and female rats. In the following chapter double immunohistochemistry

will be utilized to determine if these cells are indeed neurotensinergic.

150

7. EFFECTS OF PROLACTIN ON EXPRESSION OF
FRA IN NEUROTENSIN—IR NEURONS IN THE ARC

A. INTRODUCTION

The inductive component of prolactin feedback involves a change in the capacity
of TIDA neurons to respond to prolonged periods of hyperprolactinemia (Moore et al.,
1987a; Hentschel et al., 2000a), and does not occur until 6-12 h following systemic
(Hokfelt and Fuxe, 1972; Selmanoff, 1981; Demarest et al., 1984, 1986) or central
(Annunziato and Moore, 1978; Hentschel et al., 2000a) administration of exogenous
prolactin. In Chapter 4 we proposed that prolactin induces long-term changes in gene
expression that result in increased capacity of TIDA neurons to synthesize and release
dopamine, and demonstrated changes in immediate early gene expression in TIDA
neurons 3-6 h after prolactin administration (Hentschel et al., 2000b).

In Chapter 5 neural activity mapping with inducible transcription factors led to the
novel identification of non-TH-IR cells in the ARC that are prolactin-responsive.
Furthermore, the changes in immediate early gene expression induced by prolactin in
these cells temporally precedes or coincides with those of TIDA neurons. The presence
of these cells in the ARC is consistent with deafferentation studies that localized the
actions of prolactin within the medial basal hypothalamus (Gudelsky et al., 1978), and
suggests that these chemically unidentified cells could be mediators of the inductive
activation of TIDA neurons.

In Chapter 6, the role of neurotensin in prolactin-induced activation of TIDA
neurons was investigated since several characteristics suggested a role for a neuropeptide

mediator. The delayed inductive activation of TIDA neurons by prolactin has a 6-12 h

151

latency and is protein synthesis-dependent (Johnston et al., 1980). Neurotensin is a
neuropeptide that activates TIDA neurons acutely (Widerlov et al., 1982; Herbison et al.,
1986; Gudelsky et al., 1989; Pan et al., 1992 Lin and Pan, 1993) and is localized within
the ARC (Emson et al., 1982; Merchanthaler and Lennard, 1991). Thus, it was proposed
that neurotensin mediates the delayed inductive activation of TIDA neurons by prolactin,
and consistent with this hypothesis, the non-peptide neurotensin receptor antagonist SR-
48692 (Gully et al., 1993) blocked the activation of TIDA neurons in neurochemical
studies (Hentschel et al., 1998).

It remains plausible that the acutely prolactin-responsive, chemically unidentified
cells in the ARC are neurotensin neurons, and that these neurons mediate the delayed
inductive activation of TIDA neurons by prolactin. Logically, if these neurotensin
neurons mediate the effects of prolactin, then they should be responsive to
hyperprolactinemia. In this series of experiments we tested the hypothesis that
neurotensin neurons in the ARC are prolactin-responsive using neural activity mapping
with inducible transcription factors. This technique has not been previously described for
use in neurotensin neurons. However, several lines of evidence suggest that FRAs play a
predominant role in the regulation of neurotensin gene expression. For example, the rat
neurotensin gene contains an AP-l consensus sequence in its promoter (Kislauskis et al.,
1988; Rostene and Alexander, 1997), antisense phosphothioate oligodeoxynucleotides to
c-fos attenuate induction of neurotensin gene expression (Merchant, 1994; Robertson et
al., 1995; Svennigsson et al., 1997) and mutations in the AP-l site of this gene decreases

responses to all inducers of neurotensin gene expression by at least an order of magnitude

152

(Harrison et al., 1995). Therefore, FRA proteins are reasonable transcription factors to
target for investigation of this hypothesis.

If neurotensin neurons in the ARC are prolactin-responsive, then prolactin will
induce changes in immediate early gene expression in these neurons. To determine if this
is the case, initial experiments examined the time course of effects of haloperidol-induced
endogenous hyperprolactinemia on F RA expression in neurotensin neurons in DM and
VL subdivisions of the rostral, middle and caudal ARC in male rats as estimated by dual
immunohistochemistry for the detection of FRA- and neurotensin-IR. Subdivisions of the
DM ARC (i.e. magnocellular and parvocellular regions) were not analyzed separately in
these studies since these regions are defined by TH-IR perikarya and can not be reliably
estimated in the absence of TH staining. Follow-up experiments compared the time
course effects of haloperidol to those of exogenous prolactin, and determined the

localization of neurotensin-IR neurons relative to TH-IR neurons in the ARC in male rats.

153

B. RESULTS

Neurotensin-IR neurons are not homogeneously distributed throughout the ARC.
Rather, there exists a rostrocaudal gradient in numbers of neurotensin-IR neurons within
ARC subdivisions. As shown in figure 7.1, there are successively more neurotensin-IR
neurons in the DM and VL subdivisions as ARC regions extend caudally. Furthermore,
for any rostrocaudal ARC region there were nearly twice as many neurotensin-IR neurons
in the VL-ARC as compared to the DM-ARC (figure 7.1).

In order to test the hypothesis that neurotensin-IR neurons are responsive to
hyperprolactinemia, the time course of effects of hyperprolactinemia on expression of
FRA in neurotensin-IR neurons was examined using dual immunohistochemistry.
Logically, any hyperprolactinemia-induced changes in immediate early gene expression
must be detected over basal levels in these neurons. The basal FRA expression in
neurotensin-IR cells of the ARC was relatively low with the highest percentage of
neurotensin-IR neurons expressing FRA in the rostral VL-ARC. As shown in figure 7.2,
the percentage of neurotensin-IR neurons expressing FRA in the VL-ARC decreased from
6% to 2% as ARC regions extended caudally. For all rostrocaudal ARC regions basal
F RA expression in neurotensin-IR cells in the VL-ARC was greater than that in the DM-
ARC (figure 7.2).

Figure 7.3 compares representative photomicrographs of FRA expression in
neurotensin-IR neurons of the VL subdivision in the middle ARC 3 h after vehicle or
haloperidol injection in male rats. Bright-field images show that FRA expression in
neurotensin-IR neurons is increased in the VL-ARC after haloperidol administration. The

percentage of neurotensin-IR neurons expressing FRA was increased afier haloperidol

154

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

20 ' [:1 ROSTRAL
m MIDDLE
m CAUDAL *,#
m 15 - # T
6 * //T
E T r /
. I a
*2 / /
.. 5 _ T % %
0 4. 4.
DM-ARC VL-ARC

Figure 7.1 Comparison of numbers of NT-IR neurons in the DM and VL subdivisions of
rostral, middle and caudal ARC of male rats. After perfusion, brains were removed,
sectioned in the frontal plane (30 um) and immunohistochemically processed for the
detection of NT-IR in the DM and VL subdivisions of the rostral (open columns), middle
(hatched columns) and caudal (cross-hatched columns) ARC. Columns represent means
and vertical lines 1 SEM from 6-8 animals. *, Values within an ARC subdivision that are
significantly different (p < 0.05) from the rostral ARC. #, Values within a rostrocaudal
region that are significantly different (p < 0.05) from the DM-ARC.

155

j—I
O
l

[:1 ROSTRAL
m MIDDLE
m CAUDAL

OO
__4

 

 

 

% NT-IR neurons expressing FRA

 

 

 

 

 

 

 

 

DM-ARC VL-ARC

Figure 7.2 Comparison of percentages of NT-IR neurons expressing FRA-IR nuclei in
the DM and VL subdivisions of the rostral, middle and caudal ARC in male rats. After
perfusion, brains were removed, sectioned in the frontal plane (30 um) and
immunohistochemically processed for the detection of FRA-IR and NT-IR in the DM and
VL subdivisions of the rostral (open columns), middle (hatched columns) and caudal
(cross-hatched columns) ARC. Columns represent means and vertical lines 1 SEM from
6-8 animals. *, Values within an ARC subdivision that are significantly different (p <
0.05) from the rostral ARC. #, Values within a rostrocaudal region that are significantly
different (p < 0.05) from the DM-ARC.

156

 

Figure 7.3 Digitized photomicrographs showing increased FRA expression in NT-IR
neurons in the middle ARC afier haloperidol injection in male rats. Animals were
injected with either vehicle (TOP PANEL; 0.3% tartaric acid, 1 ml/kg; sc), or haloperidol
(BOTTOM PANEL; 1 mg/kg; sc) and perfused 3 h later. After perfusion, brains were
removed, sectioned in the frontal plane (30 um) and immunohistochemically processed
for the detection of FRA-IR and NT-IR. Neurons with blackened FRA-IR nuclei and
brown NT-IR perikarya are indicated with an arrowhead.

157

injection for 1.5—6 h in the DM-ARC (figure 7.4) and for 1.5-3 h in the VL-ARC (figure
7.5) in all rostrocaudal regions in male rats. There were no differences in the magnitude
of the response to haloperidol with regard to either ARC subdivision or rostrocaudal ARC
region.

Figure 7.6 compares representative photomicrographs of numbers of neurotensin-
IR neurons in the middle ARC afier vehicle or haloperidol injection in male rats. Bright-
field images show that numbers of neurotensin-IR neurons were increased after
haloperidol administration. Haloperidol increased numbers of neurotensin-IR neurons in
all rostrocaudal regions in male rats 12 h after injection in the DM-ARC (figure 7.7, top
panel) and 6-12 h after injection in the VL-ARC (figure 7.7, bottom panel). The
haloperidol-induced increase in total numbers of neurotensin-IR neurons was greatest in
middle to caudal VL—ARC.

Central administration of prolactin increased the percentage of neurotensin-IR
neurons expressing FRA at all time points (i.e. 1.5-12 h) in the DM (figure 7.8) and VL
(figure 7.9) subdivisions of all rostrocaudal ARC regions in male rats. Moreover, there
were no differences in the magnitude of response to prolactin with regard to ARC
subdivision or rostrocaudal ARC region. Prolactin increased numbers of neurotensin-IR
neurons in all rostrocaudal regions in male rats 12 h after injection in the DM-ARC
(figure 7.10, top panel) and 6-12 h after injection in the VL-ARC (figure 7.10, bottom
panel). The prolactin-induced increases in numbers of neurotensin-IR neurons were
greatest in VL subdivisions and caudal ARC regions.

To determine if the increased expression of neurotensin-IR following prolactin

treatment (figure 7.10) was occurring within dopamine neurons in the ARC, dual

158

DM-ARC

 

 

 

35 -
E O Rostral
3O - ,
LL. C] Middle
DD
.5 25 - A Caudal
3
8 20 r
E?
o
m 15 _
8
a 10-
t:
a s-
E
0 _
o\°
0.0 1.5 3.0 6.0 12.0
HOURS AFTER HALOPERIDOL

Figure 7.4 Time course effects of haloperidol on percentages of NT-IR neurons
expressing FRA in the DM subdivision of rostral (circles), middle (squares) and caudal
(triangles) ARC of male rats. Rats were injected with haloperidol (filled symbols; 1
mg/kg; sc) and perfused 1.5, 3, 6 or 12 h later. Zero time controls (open symbols) were
injected with vehicle (0.3% tartaric acid; 1 ml/kg; sc) 3 h prior to perfusion. After
perfusion, brains were removed, sectioned in the frontal plane (30 um) and
immunohistochemically processed for the detection of FRA-IR and NT-IR in the DM-
ARC. Symbols represent means and vertical lines 1 SEM from 6-8 animals. *, Values for
haloperidol-treated rats that are significantly different (p < 0.05) from vehicle—treated
controls.

159

VL-ARC

 

 

 

 

 

 

3O -
E O Rostral
LI-t 25 e D Middle
%0 A Caudal
(I)
8 20 ~
H
Q
X
a)
V1 15 -
8
§
{:1 10 - /
9.4.
E 5-
o\°

0 l I l l J

0.0 1.5 3.0 6.0 12.0
HOURS AFTER HALOPERIDOL

Figure 7.5 Time course effects of haloperidol on percentages of NT-IR neurons
expressing FRA in the VL subdivision of rostral (circles), middle (squares) and caudal
(triangles) ARC of male rats. Rats were injected with haloperidol (filled symbols; 1
mg/kg; sc) and perfused 1.5, 3, 6 or 12 h later. Zero time controls (open symbols) were
injected with vehicle (0.3% tartaric acid; 1 ml/kg; sc) 3 h prior to perfusion. After
perfusion, brains were removed, sectioned in the frontal plane (30 um) and
immunohistochemically processed for the detection of FRA-IR and NT-IR in the VL-
ARC. Symbols represent means and vertical lines 1 SEM from 6-8 animals. *, Values for

haloperidol-treated rats that are significantly different (p < 0.05) from vehicle-treated
controls.

160

 

Figure 7.6 Digitized photomicrographs showing increased numbers of NT-IR neurons in
the middle ARC after haloperidol injection in male rats. Animals were injected with
either vehicle (TOP PANEL; 0.3% tartaric acid, 1 ml/kg; sc), or haloperidol (BOTTOM
PANEL; 1 mg/kg; sc) and perfused 3 h later. After perfusion, brains were removed,
sectioned in the fiontal plane (30 um) and immunohistochemically processed for the
detection of NT-IR. Neurons with NT-IR perikarya are indicated with an arrowhead.

161

35 _ DM-ARC
O Rostral
30 . Cl Middle
A Caudal
25 -

20. *

# NT-IR NEURONS
8 3
m

 

 

 

 

 

(I)
z
E
E
EL
2
Eu:
5 .
0 J l I J
0.0 1.5 3.0 6.0 . 12.0
HOURS AFTER HALOPERIDOL

Figure 7.7 Time course effects of haloperidol on numbers of NT-IR neurons in the DM
and VL subdivisions of rostral (circles), middle (squares) and caudal (triangles) ARC in
male rats. Rats were injected with haloperidol (filled symbols; 1 mg/kg; sc) and perfused
1.5, 3, 6 or 12 h later. Zero time controls (open symbols) were injected with vehicle
(0.3% tartaric acid; 1 ml/kg; sc) 3 h prior to perfusion. After perfusion, brains were
removed, sectioned in the frontal plane (30 um) and immunohistochemically processed
for the detection of FRA-IR and NT-lR in the DM-ARC (TOP PANEL) and VL-ARC
(BOTTOM PANEL). Symbols represent means and vertical lines 1 SEM from 6-8
animals. *, Values for haloperidol-treated rats that are significantly different (p < 0.05)
from vehicle-treated controls.

162

DM-ARC

 

 

 

50 r
g =1: 0 Rostral
”-1 40 - D Middle
on E
.5 * /\ A Caudal
V)
8 3o - \ *
ii a:
a: \ .. 3;
m 20 r /
g a:
S
Q)
r: 10 _
.95
[E o - 3
o\°

0.0 1.5 3.0 6.0 12.0
HOURS AFTER PROLACTIN

Figure 7.8 Time course effects of prolactin on percentages of NT-IR neurons expressing
FRA in the DM subdivision of rostral (circles), middle (squares) and caudal (triangles)
ARC of male rats. Rats were injected with prolactin (filled symbols; 10 rig/rat; icv) and
perfused 1.5, 3, 6 or 12 h later. Zero time controls (open symbols) were injected with
vehicle (water; 3 ul/rat; icv) 3 h prior to perfusion. Afier perfusion, brains were removed,
sectioned in the frontal plane (30 um) and immunohistochemically processed for the
detection of FRA-IR and NT-IR. Symbols represent means and vertical lines 1 SEM from
6—8 animals. ‘, Values for prolactin-treated rats that are significantly different (p < 0.05)
from vehicle-treated controls.

163

VL-ARC

50—

* O Rostral
40 _ l 1:] Middle

* A Caudal J
a:
30 ~ * * 4A a:
a: * _l*

 

 

% NT-IR neurons expressing FRA

 

 

20 ~ *
a:
10 ~
'1
O _ A
0.0 1.5 3.0 6.0 12.0
HOURS AFTER PROLACTIN

Figure 7.9 Time course effects of prolactin on percentages of NT-IR neurons expressing
FRA in the VL subdivision of rostral (circles), middle (squares) and caudal (triangles)
ARC of male rats. Rats were injected with prolactin (filled symbols; 10 rig/rat; icv) and
perfused 1.5, 3, 6 or 12 h later. Zero time controls (open symbols) were injected with
vehicle (water; 3 Ill/fat; icv) 3 h prior to perfirsion. After perfusion, brains were removed,
sectioned in the frontal plane (30 um) and immunohistochemically processed for the
detection of FRA-IR and NT-IR. Symbols represent means and vertical lines 1 SEM from
6-8 animals. *, Values for prolactin-treated rats that are significantly different (p < 0.05)
from vehicle-treated controls.

164

DM-ARC

35'

 

 

 

 

 

 

 

O Rostral
m 30 * [:1 Middle
E 25 _ A Caudal
ad
:3
LI.)
Z
9%
l-‘
Z
:a:
(I)
Z
é
LLJ
Z
9%
E—
Z
:4:
5 a
0 4 1 J 1 Q
0.0 1.5 3.0 6.0 - 12.0
HOURS AFTER PROLACTIN

Figure 7.10 Time course effects of prolactin on numbers of NT-IR neurons in the DM
and VL subdivisions of rostral (circles), middle (squares) and caudal (triangles) ARC in
male rats. Rats were injected with prolactin (filled symbols; 10 ug/rat; icv) and perfused
1.5, 3, 6 or 12 h later. Zero time controls (open symbols) were injected with vehicle
(water; 3 til/rat; icv) 3 h prior to perfusion. After perfusion, brains were removed,
sectioned in the frontal plane (30 um) and immunohistochemically processed for the
detection of NT-IR in the DM-ARC (TOP PANEL) and VL-ARC (BOTTOM PANEL).
Symbols represent means and vertical lines 1 SEM from 6-8 animals. *, Values for
prolactin-treated rats that are significantly different (p < 0.05) from vehicle-treated
controls.

165

immunohistochemistry for the detection of TH and neurotensin was performed on tissue
sections derived from the prolactin time course. Figure 7.11 depicts the localization of
neurotensin-IR neurons relative to TH-IR neurons in the VL-ARC in male rats. Under
bright-field optics the outlines of neurotensin—IR neurons stained black were traced and
superimposed upon the fluorescent images of TH-IR neurons in the same section under
dark-field optics. Immunoreactivity for neurotensin and TH was very rarely colocalized
within neurons in the ARC. However, neurotensin-IR neurons were regularly adjacent to
TH-IR neurons in the DM-ARC and VL-ARC. As shown in figure 7.12, the numbers of
neurotensin-IR neurons in the middle DM-ARC were increased 12 h afier prolactin
injection. However, no increase in colocalization of neurotensin and TH was observed at
any time after prolactin injection. Similarly, in the middle VL-ARC, numbers of
neurotensin-IR neurons were increased 6-12 h after prolactin injection while no increase

in colocalization of neurotensin and TH was observed through 12 h (figure 7.13).

I66

50mm

 

Figure 7.11 Digitized photomicrographs depicting the localization of NT-IR neurons
relative to TH-IR neurons in the VL-ARC in male rats. Brains were perfused, removed,
sectioned in the frontal plane (30 um) and immunohistochemically processed for the
detection of NT—IR and TH-IR. Black NT-IR neurons under bright-field optics (LEFT
PANEL) were traced and superimposed on the same image under dark-field optics which
shows fluorescent TH-IR neurons (LEFT PANEL).

167

DM-ARC

5-

$15 *
4* o

%10

LL]
3- 2

E55

[—

z

:tt:

 

# Dual NT-IR and TH-IR neurons

 

J_ l l l l

0.0 1.5 3.0 6.0 12.0
HOURS AFTER PROLACTIN

Figure 7.12 Time course effects of prolactin on numbers of dual NT-IR and TH-IR
neurons, and numbers of NT-IR neurons (INSET) in the DM subdivision of the middle
ARC in male rats. Rats were injected with prolactin (filled symbols and columns; 10
pyrat; icv) and perfused 1.5, 3, 6 or 12 h later. Zero time controls (open symbol and
column) were injected with vehicle (water; 3 ul/rat; icv) 3 h prior to perfiision. After
perfusion, brains were removed, sectioned in the frontal plane (30 um) and
immunohistochemically processed for the detection of NT-IR and TH-IR. Symbols and
columns represent means and vertical lines 1 SEM from 6-7 animals. *, Values for
prolactin-treated rats that are significantly different (p < 0.05) from vehicle-treated
controls.

168

 

 

 

# Dual NT-IR and TH-IR neurons
N

 

 

1 r—
0 _ 0’5 1' -fi a
0.0 1.5 3.0 6.0 12.0
HOURS AFTER PROLACTIN

Figure 7.13 Time course effects of prolactin on numbers of dual NT-IR and TH-IR
neurons, and numbers of NT-IR neurons (INSET) in the VL subdivision of the middle
ARC in male rats. Rats were injected with prolactin (filled symbols and columns; 10
rig/rat; icv) and perfused 1.5, 3, 6 or 12 h later. Zero time controls (open symbol and
column) were injected with vehicle (water; 3 til/rat; icv) 3 h prior to perfusion. After
perfusion, brains were removed, sectioned in the frontal plane (30 um) and
immunohistochemically processed for the detection of NT-IR and TH-IR. Symbols and
columns represent means and vertical lines 1 SEM from 6-7 animals. *, Values for
prolactin-treated rats that are significantly different (p < 0.05) from vehicle-treated
controls.

169

C. DISCUSSION:

In the present study neural activity mapping using inducible transcription factors
was applied to neurotensin-IR neurons in the ARC to further characterize the role of
neurotensin in the delayed inductive activation of TIDA neurons by prolactin.

Perikarya of neurotensin neurons are heterogeneously distributed throughout the
ARC. Two populations of neurotensin-containing neurons have been identified in the
ARC based on location of perikarya and termini. Neurotensin neurons in the DM-ARC
terminate within the ARC, whereas neurotensin neurons in the VL-ARC terminate
predominantly in the median eminence (Meister et al., 1989). The results of the present
study indicate that the majority of neurotensin neurons are localized in the VL-ARC
consistent with a previous report (Lantos et al., 1995) and that there exists a rostrocaudal
gradient in numbers of neurotensin neurons such that there are more neurotensin neurons
in subdivisions in the caudal extent of the ARC. Since TIDA neurons of the DM-ARC
have been shown to have neurotensin receptor mRN A (Alexander, 1997; Rostene and
Alexander, 1997) and neurotensin neurons in this ARC subdivision terminate locally
(Meister et al., 1989), it is possible that these neurotensin neurons are involved in the
feedback activation of TIDA neurons by prolactin.

Neurotensin neurons in the VL-ARC terminate in the external zone of the median
eminence, but neurotensin is not secreted into the hypophysial portal blood (Clarke et al.,
1993). Thus it is possible that these neurons are involved in axon-axonal interactions
with TIDA neurons to facilitate the secretion of dopamine. However, the role of these

neurons and the significance of their anatomy remains to be elucidated.

170

Despite differences in numbers of neurotensin neurons in ARC subdivisions and
rostrocaudal regions, the percentage of these neurons expressing FRA-IR nuclei was low
(<8%) across all ARC subdivisions and rostrocaudal regions in male rats. In agreement
with previous neurochemical studies (Chapter 6), data suggest that the majority of
neurotensin neurons in the ARC are not tonically active.

Consistent with the hypothesis that prolactin induces immediate early gene (i.e.
FRA) expression in neurotensin neurons in the ARC, both haloperidol-induced
endogenous hyperprolactinemia and central administration of prolactin cause rapid
increases (by 1.5 h) in the percentage of neurotensin neurons expressing F RA in the DM
and VL subdivisions of the ARC in all rostrocaudal regions. This response persists 3-6 h
in haloperidol-treated animals and 12 h in prolactin-treated rats. The longer response in
icv prolactin-treated rats is likely related to the pharmacokinetics of central dosing (as
discussed in Chapter 4).

Changes in immediate early gene expression in neurotensin neurons are followed
by delayed increases in total numbers of neurotensin neurons. In all rostrocaudal regions
the numbers of neurotensin-IR neurons in the DM-ARC are increased 12 h after
injections of haloperidol or prolactin. Similarly, total numbers of neurotensin-IR neurons
in the VL-ARC are increased 612 h after injections of haloperidol or prolactin. The
discrepancy between the DM-ARC and VL-ARC in the temporal characteristics of
neurotensin expression may be related to the shorter latency to peak changes in
immediate early gene expression in the VL-ARC (1.5 h) compared to that of the DM-
ARC (3 h), or to differential post-translational processing of proneurotensin (Carraway

and Mitra, 1990; Checler et al., 1991; Woulfe et al., 1994) in the VL-ARC as compared

171

to the DM-ARC. The increased capacity of TIDA neurons that temporally coincides with
the increase in numbers of neurotensin neurons in the ARC could be due to the effects of
this neuropeptide as a neuromodulator of dopaminergic neurotransmission. For example,
neurotensin has been shown to facilitate dopamine release in the NSDA system (Battaini
etal., 1986; Tanganelli et a1, 1993).

Activation of TIDA neurons is blocked following administration of a neurotensin
receptor antagonist 11 h after induction of hyperprolactinemia, suggesting that the long
protein synthesis-dependent period could correspond to the de novo synthesis and release
of neurotensin. In the present study, neurotensin neurons were localized within the ARC
in close proximity to TIDA neurons. Immediate early gene expression in these
neurotensin neurons is induced by hyperprolactinemia and the time course of these
changes precede those in TIDA neurons (i.e. numbers of neurotensin cells are increased
6-12 h following hyperprolactinemia). These data support the hypothesis that delayed
inductive activation of TIDA neurons by prolactin is mediated by neurotensin neurons in
the ARC.

Since proneurotensin mRN A (Kiyarna and Emson 1991) and neurotensin-IR
(Chronwall, 1985; Palkovits, 1992) have been localized in neurons within the ARC
including those containing TH-IR (Ibata et al., 1983; Halasz et al., 1985), prolactin-
induced neurotensin synthesis could occur in afferent neurotensin neurons in the ARC
(Merchanthaler and Lennard, 1991; Marcos et al., 1996a) and/or in subpopulations of
TIDA neurons themselves (Hokfelt et al., 1984; Bachelet et al., 1997). Dual immuno-
histochemistry for the detection of neurotensin-IR and TH-IR was utilized to determine if

the prolactin-induced synthesis of neurotensin occurs in TIDA neurons. Consistent with

172

one report (Bachelet et al., 1997) yet contrasting others (Kahn et al., 1980; Ibata et al.,
1983; Hokfelt et al., 1984; Ciofi et al., 1993), neurotensin-IR is rarely colocalized with
TH-IR neurons in the ARC. Moreover, no increase in colocalization is observed at any
time following prolactin administration for up to 12 h. Instead, neurotensin-IR neurons
are often located adjacent to TH-IR neurons in the DM-ARC and VL-ARC. Thus, it is
more likely that neurotensin mediates prolactin-induced activation of TIDA neurons via
local afferent input within the mediobasal hypothalamus.

The immunohistochemical detection of neurotensin-IR perikarya in the ARC has
inherent technical challenges. Neurotensin is synthesized de novo as a proneuropeptide
and is proteolytically processed to its active form during axonal transport (Hook et al.,
1990; Rostene and Alexander, 1997). Furthermore, neurotensin-IR fibers heavily ramify
over the entire ARC (Kahn et al., 1980; Jennes et al., 1982). Consequently, the staining
of neurotensin-IR perikarya may be weak and must be detected over the high background
of neurotensin-IR fibers. To evade this technical difficulty many investigators pretreat
animals with colchicine, an inhibitor of axonal transport. Accordingly, colchicine
increases neuropeptide content in perikarya and decreases fiber background (Kahn et al.,
1980). However, recent studies demonstrate that colchicine treatment per se rapidly and
strongly induces expression of some peptides, including that of neurotensin, in certain
cell populations in the vicinity of the ventricles (Kiyama and Emson, 1991; Alexander
and Leeman, 1994; Rostene and Alexander, 1997). Consequently, experimental
interpretations may be complicated/confounded by the use of colchicine. Thus reports of
extensive colocalization of neurotensin and TH in studies utilizing colchicine may have

been overestimated.

173

In conclusion, the results of this study reveal that prolactin acutely regulates
immediate early gene expression in neurotensin neurons in the ARC, and induces a
delayed increase in neurotensin synthesis in neurons in the ARC that is not colocalized in
TIDA neurons. These results provide additional evidence that the delayed inductive
activation of TIDA neurons by prolactin is indeed mediated by local afferent neurotensin

neurons in the ARC.

174

8. SUMMARY AND DISCUSSION
A. SUMMARY

It has been more than ten years since the discovery of the delayed-inductive
activation of TIDA neurons by prolactin. However, the mechanisms underlying the
effects of prolactin remain largely unknown. The overall aim of the experiments
described in this dissertation was to gain insight into these mechanisms. To this end,
studies herein detailed: the temporal effects of prolactin on neurochemical estimates of
TIDA neuronal activity, the temporal effects of prolactin on immediate early gene
expression in cells of the ARC including TIDA neurons, and the role of neurotensin as a
mediator of the effects of prolactin. The major points fi'om these studies are summarized
below.

A single injection of haloperidol elicits an early (by 1 h) and prolonged (through
12 h) state of hyperprolactinemia. Hyperprolactinemia subsequently activates TIDA
neurons afier a delay of six or more h. Irnmunoneutralization of circulating prolactin
blocks the delayed activation of TIDA neurons by haloperidol, confirming that prolactin
mediates the stimulatory effects of haloperidol on these neurons.

In the temporal characterization of the delayed response, experiments demonstrate
that the activation of TIDA neurons is dependent upon a period of sustained
hyperprolactinemia for initiation of the response. A period greater than 3 h, but less than
6 h of hyperprolactinemia is required for its initiation. Furthermore, hyperprolactinemia
is required to maintain the response as immunoneutralization of circulating prolactin

blocks the activation of TIDA neurons in delayed response-initiated animals.

175

Prolactin regulates the expression of FRA in TH-IR neurons in the ARC. In
female rats, the basal expression of FRA in TH-IR neurons in the ARC is 4 times greater
than that in males, in agreement with the gender difference established by neurochemical
estimates. Hyperprolactinemia causes transient increases in the percentage of TH-IR
neurons expressing FRA after 1.5 h in the VL-ARC and 3 h in the DM-ARC (figure 8.1).
Immunoneutralization studies indicate that these changes are dependent upon a sustained
period of hyperprolactinemia greater than 1 h for induction.

Prolactin also regulates the expression of FRA in chemically unidentified neurons
in the ARC. In non-catecholaminergic cells, prolactin causes transient increases in
numbers of FRA-IR nuclei after 1.5-3 h in the VL—ARC, and 3 h in the DM-ARC (figure
8.1). Additionally, the changes in FRA expression occur in both magnocellular and
parvocellular subdivisions of the DM-ARC.

Prolactin regulates the expression of F RA in neurotensin-IR neurons in the ARC.
Prolactin increases the percentage of neurotensin-IR neurons expressing FRA from 1.5-12
h in the ARC, and causes a delayed increase in the total number of neurotensin-IR
neurons after 6 h in the VL-ARC and 12 h in the DM-ARC (figure 8.1).
Immunohistochemistry demonstrates that the neurotensin-IR is only rarely colocalized
with TH-IR, and that the incidence is not affected by up to 12 h of hyperprolactinemia.
However, the same studies also demonstrate that neurotensin neurons are often closely
associated with TIDA neurons.

The neurotensin receptor antagonist SR-48692 has no effect on neurochemical
estimates of basal TIDA neuronal activity or plasma prolactin concentrations. However,

this antagonist blocks the delayed stimulatory effects of prolactin on dopamine release

176

1.5 HOURS 3 HOURS

 

. NT-IR
O Non-TH-IR
o FRA-IR

 

 

 

 

Figure 8.1 Schematic summary of the temporal effects of prolactin on FRA expression
in cells, and recruitment of neurotensin neurons in ARC subdivisions during the inductive
activation of TIDA neurons. Abbreviations: DA, dopamine; FRA, Fos-related antigens;
IR, immunoreactive; NT, neurotensin; TH, tyrosine hydroxylase.

177

from TIDA neurons. The results suggest that neurotensin mediates the delayed, but not

tonic stimulatory effects of prolactin on TIDA neurons.

178

B. CONCLUDING DISCUSSION

To review, the inductive component of TIDA neuronal activation by prolactin
involves a change in the capacity to respond to prolonged periods of hyperprolactinemia
(Moore et al., 1987a) and is observed in both genders. It stimulates TIDA neuronal
activity to levels 2-3 times basal (Demarest et al., 1985a; Selmanoff, 1985) 6—12 h
following systemic (Hokfelt and Fuxe, 1972; Selmanoff, 1981; Demarest et al., 1984;
1986) or central (Annunziato and Moore, 1978) administration of exogenous prolactin, or
after manipulations which elicit endogenous hyperprolactinemia (Moore et al., 1985;
Hentschel et al., 2000a). This component of prolactin feedback is protein synthesis-
dependent (Johnston et al., 1980) and is associated with increased gene expression [i.e.
TH mRNA] in TIDA neurons (Arbogast and Voogt, 1991; Selmanoff et al., 1991).

Prolactin exerts multiple effects on TIDA and other non-catecholaminergic
neurons in the ARC to elicit the change in capacity and increased responsiveness of TIDA
neurons. Data from the experiments in this dissertation have provided insight into the
mechanisms underlying the delayed-inductive activation of TIDA neurons by prolactin.
The data summarized early in this chapter are interpreted and mechanisms leading to the
inductive activation of TIDA neurons in the 6th to 12th hour are temporally presented as
they theoretically occur.

Following 1.5 h of elevated prolactin, changes are occurring in gene expression in
neurotensin neurons throughout the ARC. Perhaps these neurons begin to upregulate
synthetic enzymes for preproneurotensin. However, it is unlikely that these neurons

directly stimulate TIDA neurons (at this time) as TIDA neurons are acutely activated by

179

central injections of neurotensin within 1 h which is inconsistent with the neurochemical
time course of TIDA neuronal activation.

At the same time, prolactin selectively elicits changes in gene expression in
catecholaminergic neurons in the VL-ARC. These neurons may be involved in the
stimulation of the TIDA neurons in the DM-ARC, and/or may be recruited and induced to
synthesize AAAD.

After 3 h of elevated prolactin, there is sustained immediate early gene expression
in neurotensin neurons throughout the ARC. Presumably these neurons are molecularly
active as they continue to synthesize neuropeptide and perhaps begin to recruit other
neurons to synthesize neurotensin.

TIDA neurons of the DM-ARC show transient changes in immediate early gene
expression after 3 h of elevated prolactin. These changes are dependent on sustained
hyperprolactinemia for initiation. The changes likely involve modulation of late response
genes such as the rate limiting enzyme, TH which could ultimately contribute to the
increased capacity of this system. Since data show that the total number of
catecholaminergic neurons in the ARC is constant, it is likely that the level of activity of
existing neurons is elevated. This idea is consistent with data from investigators who note
increased TH mRNA in the ARC after chronic hyperprolactinemia (Selmanoff, 1991;
Arbogast and Voogt, 1991).

The increased immediate early gene expression in TIDA neurons of the DM-ARC
at 3 h is unlikely representative of increased neurotransmitter release since it is
inconsistent with the neurochemical time course of activation. Increases in immediate

early gene expression may also be elicited from afferent neuronal input that modulate

180

second messenger systems, protein kinase activity and the expression of constitutive
transcription factors or their localization. For example, it has been demonstrated that
some growth factors and neuropeptides may affect levels of second messengers and
induce changes in immediate early genes, like FRA, without affecting neurotransmitter
release (Hoffinan and Murphy, 2000).

Following 6 h of elevated prolactin, there is sustained elevation of immediate
early gene expression in neurotensin neurons throughout the ARC as well as evidence of
recruitment of neurotensin neurons in the VL-ARC. In addition, neurochemistry
demonstrates the onset of activation of TIDA neurons that is mediated through
neurotensin receptors (Hentschel et al., 1998). The data suggest that neurotensin neurons
have a role proximal to TIDA neurons in the circuitry of the inductive component (figure
8.2). In agreement with the well-known facilitatory role of neurotensin neurons in
dopaminergic systems, neurotensin could act to enhance dopamine release from TIDA
neurons.

After 12 h of increased prolactin levels, immediate early gene expression in
neurotensin neurons remains elevated and total numbers of neurotensin-IR neurons
throughout the ARC are increased. The data are consistent with recruitment of
neurotensin neurons into an active pool and ongoing neuropeptide synthesis. Given the
intimate proximity of neurotensin and TIDA neurons in the ARC, it is plausible that
stimulatory input from neurotensin neurons to TIDA neurons is maintained by elevated
prolactin and mediated through neurotensin receptors (figure 8.2). The increased capacity
of TIDA neurons could be from several mechanisms: (1) increased synthetic capacity of

existing TIDA neurons by induction of synthetic enzymes such as the rate limiting

181

 

 

D :

ME

Figure 8.2 Schematic representation of the theoretical distal circuitry of the inductive
activation of TIDA neurons by prolactin. Abbreviations: 3V, third ventricle; ARC,
arcuate nucleus; DM, dorsomedial; MC, magnocellular; ME, median eminence; NT,
neurotensin neuron; NTR, neurotensin receptor; PC, parvocellular; TIDA, tubero-
infundibular dopamine neuron; VL, ventrolateral.

182

enzyme, TH; (2) the facilitative effects of neurotensin neuromodulation of TIDA neuronal
activity; and/or (3) the recruitment of the TH-IR neurons of the VL-ARC to synthesize
AAAD, and release dopamine. Such hypotheses could be tested using in situ
hybridization histochemistry.

In situ hybridization histochemistry (ISHH) could be utilized to test for increased
synthetic capacity of TIDA neurons via induction of TH and other synthetic enzymes by
examining the time course effects of central prolactin administration on synthetic enzyme
mRNA expression in TIDA neurons. Similarly, if the increased capacity of TIDA neurons
is due to the recruitment of TH-IR neurons of the VL-ARC to synthesize AAAD and
release dopamine, then the time course effects of central prolactin administration should
demonstrate the induction of AAAD mRN A expression in neurons in the VL—ARC
containing TH mRNA as detected using dual ISHH. These data should next be verified
using dual immunohistochemistry for the translated (protein) products.

The function of the inductive activation of TIDA neurons by prolactin is not
known. However, since prolactin exerts actions on reproduction, fluid balance, growth,
metabolism and immune function, the regulation of its secretion from the pituitary is very
important. TIDA neurons are the primary regulators of prolactin secretion. It’s possible
that the inductive component as a conserved mechanism for control of pathologic
hypersecretion of prolactin. The characteristics of the inductive component are consistent
with this: (1) the inductive component acts as the gain of a system, increasing the capacity
of the TIDA neuronal system to control prolactin secretion, if necessary, (2) prolonged
hyperprolactinemia is required for initiation of the response, and (3) sustained

hyperprolactinemia is required to sustain the response. The inductive component appears

183

to act as a fail-safe system for the regulation of the secretion of prolactin and hence

maintaining normal reproductive, growth, metabolic and immunologic functions.

184

BIBLIOGRAPHY

185

Adams MR, Ward RP and Dorsa DM: Dopamine receptor modulation of gene expression
in the brain. The Dopamine Receptors, 1 Edition. Edited by Neve KA and Neve RL.
Totowa, New Jersey, Humana Press, 1996, pp 305-342.

Alexander M]: Estrogen—regulated synthesis of neurotensin in neurosecretory cells of the
hypothalamic arcuate nucleus in the female rat. Endocrinology 1993; 133: 1809-1816.

Alexander MJ and Leeman SE: Estrogen-inducible neurotensin immunoreactivity in the
preoptic area of the female rat. Journal of Comparative Neurology 1994; 345: 496-
509.

Alexander, Mark J. Dopamine neurons in the adult rat diencephalon express mRNA
encoding high-affinity neurotensin receptor. Society For Neuroscience Abstracts
1997; 23(2): 1488-1488.

Alonso A, F aure MP and Beaudet A: Neurotensin promotes oscillatory bursting behavior

and is internalized in basal forebrain cholinergic neurons. Journal of Neuroscience
1994; 14: 5778-5792.

Andersson K, Fuxe K, Eneroth P, Nyberg F and R005 P: Rat prolactin and hypothalamic
catecholarnine nerve terminal systems. Evidence for rapid and discrete increases in
dopamine and noradrenaline turnover in the hypophysectomized male rat. European
Journal of Pharmacology 1981; 76: 261-265.

Annunziato L and Moore KE: Prolactin in csf selectively increases dopamine turnover in
the median eminence. Life Sciences 1978; 2037-2042.

Arbogast LA and Voogt JL: Sex-related alterations in hypothalamic tyrosine hydroxylase
after neonatal monosodium glutamate treatment. Neuroendocrinology 1990; 460-467.

Arbogast LA and Voogt JL: Hyperprolactinemia increases and hypoprolactinemia
decreases tyrosine hydroxylase messenger ribonucleic acid levels in the arcuate
nuclei, but not the substantia nigra or zona incerta. Endocrinology 1991; 128: 997-
1005.

Arbogast LA and Voogt J L: Prolactin receptors are colocalized in dopaminergic neurons
in fetal hypothalamic cell cultures: Effect of prolactin on tyrosine hydroxylase
activity. Endocrinology 1997; 138: 3016-3023.

Bachelet CM, Scarceriaux V, Rostene W and Pelaprat D: Evidence for neurotensin
autoreceptors and relationship of neurotensin and its receptors with tyrosine
hydroxylase-positive neurons in rat primary hypothalamic cultures. European Journal
of Neuroscience 1997; 9: 1483-1487.

Baldino F, Jr. and Wolfson B: Postsynaptic actions of neurotensin on preoptic-anterior
hypothalamic neurons in vitro. Brain Research 1985a; 325: 161-170.

186

Baldino F, Jr., Davis LG and Wolfson B: Structure-activity studies with carboxy- and

amino-terminal fragments of neurotensin on hypothalamic neurons in vitro. Brain
Research 1985b; 342: 266-272.

Bakowska JC and Morrell JI: Atlas of the neurons that express mRNA for the long form

of the prolactin receptor in the forebrain of the female rat. Journal of Comparative
Neurology 1997; 386: 161-177.

Battaini F, Govoni S, Di Giovine S and Trabucchi M: Neurotensin effect on dopamine
release and calcium transport in rat striatum: interactions with diphenylalkylarnine

calcium antagonists. Naunyn-Schmiedeberg's Archives of Pharmacology 1986; 332:
267-270.

Beauregard M, Ferron A and Descarries L: Opposite effects of neurotensin on dopamine
inhibition in different regions of the rat brain: An iontophoretic study. Neuroscience
1992; 47: 613-619.

Ben-Jonathan N: Dopamine: a prolactin-inhibiting hormone. Endocrine Reviews 1985; 6:
564-589.

Bertler A, Falck B and Rosengren E: The direct demonstration of a barrier mechanism in
the brain capillaries. Acta Pharmacologia 1964; 20: 317-321 .

Berry SA and Gudelsky GA: D1 receptors function to inhibit the activation of
tuberoinfundibular dopamine neurons. Journal of Pharmacology and Experimental
Therapeutics 1990; 254: 677-682.

Berry SA and Gudelsky GA: Effect of D2 dopamine agonists on tuberoinfirndibular
dopamine neurons. Neuropharrnacology 1991; 30: 961-965.

Berry SA and Gudelsky GA: Evidence for protein kinase-C meditation of the
neurotensin-induced activation of tyrosine hydroxylase in tuberoinfundibular
dopaminergic neurons. Endocrinology 1992; 131: 1207-1211.

Betancur C, Canton M, Gully D, Vela G, Pelaprat D and Rostene W: Characterization
and distribution of binding sites for a new neurotensin receptor antagonist ligand,
[3H]SR 48692, in the guinea pig brain. Journal of Pharmacology and Experimental
Therapeutics 1995; 273: 1450-1458.

Boudin H, Pelaprat D, Rostene W and Beaudet A: Cellular distribution of neurotensin
receptors in rat brain: immunohistochemical study using an anti-peptide antibody
against the cloned high affinity receptor. Journal of Comparative Neurology 1996;
373: 76-89.

Brun P, Steinberg R, LeFur G and Soubrie P: Blockade of neurotensin receptor by SR-
48692 potentiates the facilitatory effect of haloperidol on the evoked in viva

dopamine release in the rat nucleus accumbens. Journal of Neuroehemistryl 1995;
2073-2079.

187

Bullock BP, McNeil GP and Dobner PR: Synergistic induction of neurotensin gene
transcription in PC12 cells parallels changes in AP-l activity. Molecular Brain
Research 1994; 27: 232-242.

Callahan P, Baumann MH and Rabii J: Inhibition of tuberoinfundibular dopaminergic
neural activity during suckling: involvement of mu and kappa opiate receptor
subtypes. Journal of Neuroendocrinology 1996; 8: 771-776.

Cannon M, Hu L, Ye J and Lawson DM: A comparison of plasma prolactin levels in
young female Long-Evans and Holtzrnan rats as measured by Nb2 lymphoma
bioassay and radioimmunoassay. Proceedings of the Society for Experimental
Biology and Medicine 1991; 197: 471-476.

Carraway R and Leeman SE: The isolation of a new hypotensive peptide, neurotensin,
from bovine hypothalami. Journal of Biological Chemistry 1973; 248: 6854-6861 .

Carraway RE and Leeman SE: The amino acid sequence of a hypothalamic peptide,
neurotensin. Journal of Biological Chemistry 1975; 250: 1907-1911.

Carraway R and Leeman SE: Characterization of radioimmunoassayable neurotensin in
the rat. Its differential distribution in the central nervous system, small intestine, and
stomach. Journal of Biological Chemistry 1976; 251: 7045-7052.

Carraway RE and Mitra SP: Differential processing of neurotensin/neuromedin N
precursor(s) in canine brain and intestine. Journal of Biological Chemistry 1990; 265:
8627-8631.

Carraway RE and Leeman SE: The synthesis of neurotensin. Journal of Biological
Chemistry 1995; 250: 1912-1918.

Castagliuolo I, Leeman SE, Bartolak SE, Nikulasson S, Qiu B, Carraway RE and
Pothoulakis C: A neurotensin antagonist, SR-48692, inhibits colonic responses to
immobilization stress in rats. Proceedings of the National Academy of Sciences 1996;
93: 12611-12615.

Chalon P, Vita N, Kaghad M, Guillemot M, Bonnin J, Delpech B, Le Fur G, Ferrara P
and Caput D: Molecular cloning of a levocabastine-sensitive neurotensin binding site.
FEBS Letters 1996; 386: 91-94.

Chapin DS, Lookingland K] and Moore KE. Effects of LC mobile phase composition on
retention times for biogenic amines, and their precursors and metabolites. Current
Separations 1986; 7(3): 68-70.

Chase TN and Oh JD: Striatal mechanisms and pathogenesis of parkinsonian signs and
motor complications. Annals of Neurology 2000; 47 (4 Suppl 1): S122-130.

Chaudhuri A: Neural activity mapping with inducible transcription factors. NeuroReport
1997; 8: v-ix.

188

Checler F, Dauch P, Masuo Y and Vincent J-P: Neurotensin and neuromedin N are
differently metabolized in ventral tegmental area and nucleus accumbens. Journal of
Neurochemistry 1991; 56: 1320-1328.

Chen J, Kelz MB, Hope BT, Nakabeppu Y and Nestler EJ: Chronic fos-related antigens:
stable variants of delta-fosB induced in brain by chronic treatments. Journal of
Neuroscience 1997; 17: 4933-4941.

Cheung S, Will YM, Hentschel K, Moore KE, Lookingland KJ: Role of gonadal steroids
in determining sexual differences in expression of fos-related antigens in tyrosine
hydroxylase-immunoreactive neurons in subdivisions of the hypothalamic arcuate
nucleus. Endocrinology 1997; 138: 3804-3810.

Chiu S, Koos RD and Wise PM: Detection of prolactin receptor (PRL-R) mRNA in the
rat hypothalamus and pituitary gland. Endocrinology 1992; 130: 1747-1749.

Chiu S and Wise PM: Prolactin receptor mRNA localization in the hypothalamus by in
situ hybridization. Journal of Neuroendocrinology 1994; 191-199.

Chronwall BM: Anatomy and physiology of the neuroendocrine arcuate nucleus.
Peptides 1985; 6: 1-11.Demarest KT, McKay DW, Riegle GD and Moore KE: Sexual
differences in tuberoinfundibular dopaminergic nerve activity induced by neonatal
androgen exposure. Neuroendocrinology 1981; 32: 108-113.

Ciofi P, Crowley WR, Pillez A, Schrnued LL, Tramu G and Mazzuca M: Plasticity in
expression of immunoreactivity for neuropeptide Y, enkephalins and neurotensin in
the hypothalamic tubero-infundibular dopaminergic system during lactation in mice.
Journal of Neuroendocrinology 1993; 5: 599-602.

Constantinidis J, De La Torre JC, Tissot R and Geissbuhler F: [The capillary barrier for
DOPA in brain and different organs]. Psychopharmacologia 1969; 15: 75-87.

Crisostomo S, Guemene D, Garreau-Mills M and Zadwomy D: Prevention of the
expression of incubation behaviour using passive immunisation against prolactin in
turkey hens (Meleagris gallopavo). Reproduction, Nutrition, Development 1997; 37:
253-266.

Curran T and Franza BR: Fos and Jun: The AP-l Connection. Cell 1988; 55: 395-397.

Daikoku S, Kawano H, Okamura Y, Tokuzen M and Nagatsu I: Ontogenesis of
immunoreactive tyrosine hydroxylase-containing neurons in rat hypothalamus.
Developmental Brain Research 1986; 28: 85-98.

Dana C, Vial M, Leonard K, Beauregard A, Kitabgi P, Vincent JP, Rostene W and
Beaudet A: Electron microscopic localization of neurotensin binding sites in the
midbrain tegrnenturn of the rat. I. Ventral tegmental area and the interfascicular
nucleus. Journal of Neuroscience 1989; 9: 2247-2257.

189

Dragunow M, Robertson GS, Faull RL, Robertson HA, Jansen K: D2 dopamine receptor

antagonists induce fos and related proteins in rat striatal neurons. Neuroscience 1990;
37: 287-294.

Delle Donne KT, Sesack SR and Pickel VM: Ultrastructural immunocytochemical
localization of neurotensin and the dopamine D2 receptor in the rat nucleus
accumbens. Journal of Comparative Neurology 1996; 371: 552-566.

Demarest KT and Moore KE: Comparison of dopamine synthesis regulation in the
terminals of nigrostriatal, mesolimbic, tuberoinfundibular and tuberohypophyseal
neurons. Journal of Neural Transmission 1979; 46: 263-277.

Demarest KT and Moore KE: Disruption of 5-hydroxytryptaminergic neuronal function
blocks the action of morphine on tuberoinfundibular dopaminergic neurons. Life
Sciences 1981a; 28: 1345-1351.

Demarest KT and Moore KE: Sexual differences in the sensitivity of tuberoinfundibular

dopamine neurons to the actions of prolactin. Neuroendocrinology 1981b; 33: 230-
234.

Demarest KT, McKay DW, Riegle GD, Moore KE: Biochemical indices of
tuberoinfundibular dopaminergic neuronal activity during lactation: A lack of
response to prolactin. Neuroendocrinology 1983; 36: 130-137.

Demarest KT, Riegle GD and Moore KE: Prolactin-induced activation of
tuberoinfundibular dopaminergic neurons: Evidence for both a rapid 'tonic' and a
delayed 'induction' component. Neuroendocrinology 1984; 38: 467-475.

Demarest KT, Riegle GD and Moore KB: The interrelationship between the rapid "tonic"
and the delayed "induction" components of the prolactin-induced activation of
tuberoinfundibular dopamine neurons. Prolactin, Basic and clinical correlates. Edited
by MacLeod RM, Thorner MO and Scapagnini U. Padova, Liviana Press, 1985a, pp
533-542.

Demarest KT, Riegle GD and Moore KE: Hypoprolactinemia induced by
hypophysectomy and long-term bromocriptine treatment decreases tuberoinfimdibular
dopaminergic neuronal activity and the responsiveness of these neurons to prolactin.
Neuroendocrinology 1985b; 40: 369-376.

Demarest KT, Riegle GD and Moore KB: The rapid 'tonic' and the delayed 'induction'
components of the prolactin-induced activation of tuberoinfundibular dopaminergic

neurons following the systemic administration of prolactin. Neuroendocrinology
1986; 43: 291-299.

Dobner PR, Barber DL, Villa-Komaroff L and McKiernan C: Cloning and sequence

analysis of cDNA for the canine neurotensin/neuromedin N precursor. Proceedings of
the National Academy of Sciences 1987; 84: 3516-3520.

190

Durham RA, Johnson JD, Moore KB and Lookingland KJ: Evidence that D2 receptor-
mediated activation of hypothalamic tuberoinfundibular dopaminergic neurons in the

male rat occurs via inhibition of tonically active afferent dynorphinergic neurons.
Brain Research 1996; 732: 113-120.

Durham RA, Eaton MJ, Moore KE, Lookingland KJ: Effects of selective activation of
dopamine D2 and D3 receptors on prolactin secretion and the activity of

tuberoinfundibular dopamine neurons. European Journal of Pharmacology 1997; 335:
37-42.

Durham RA, Johnson JD, Eaton MJ, Moore KB and Lookingland KJ: Opposing roles for
D1 and D2 receptors in the regulation of hypothalamic tuberoinfundibular dopamine
neurons. European Journal of Pharmacology 1998; 355: 141-147.

Eaton MJ, Gopalan C, Kim B, Lookingland KJ and Moore KE: Comparison of the effects
of the dopamine D2 agonist quinelorane on tuberoinfimdibular dopaminergic neuronal
activity in male and female rats. Brain Research 1993a; 629: 53-58.

Eaton MJ, Moore KB and Lookingland KJ: Differential effects of the D2 receptor agonist

quinelorane on the secretion of prolactin and a-melanocyte-stimulating hormone.
Life Sciences 1993b; 53: 107-112.

Emson PC, Goedert M, Horsfield P, Rioux F and St.Pierre S: The regional distribution
and chromatographic characterisation of neurotensin-like immunoreactivity in the rat
central nervous system. Journal of Neurochemistry 1982; 38: 992-999.

Everitt BJ, Meister B, Hokfelt T, Melander T, Terenius L, Rokaeus A, Theodorsson-
Norheirn E, Dockray G, Edwardson J, Cuello AC, Elde R, Goldstein M, Hemmings
H, Ouimet C, Walaas I, Greengard P, Vale W, Weber E, Wu J-Y and Chang K-J: The
hypothalamic arcuate nucleus-median eminence complex: Immunohistochemistry of
transmitters, peptides and DARPP-32 with special reference to coexistence in
dopamine neurons. Brain Research Reviews 1986; 11: 97-155.

Evers BM, Wang X, Zhou Z, Townsend CM, Jr., McNeil GP and Dobner PR:
Characterization of promoter elements required for cell-specific expression of the

neurotensin/neuromedin N gene in a human endocrine cell line. Molecular and
Cellular Biology 1995; 15: 3870-3881.

Farkas RH, Nakajima S and Nakaj ima Y: Neurotensin excites basal forebrain cholinergic
neurons: Ionic and signal transduction mechanisms. Proceedings of the National
Academy of Sciences 1994; 91: 2853-2857.

Frawley LS, Porter TE and Kineman RD: Effects of prolactin on target cells.

Neuroendocrine Perspectives. Edited by Muller EE, MacLeod RM. New York,
Springer-Verlag, 1990, pp 39-75.

191

Fuxe K, Agnati LF, Harfstrand A, Eneroth P, Kitayama I, Farabegoli C, Tesini M, Toni R
and Goldstein M: New vistas on the morphofirnctional organization of the endocrine
hypothalamus. Focus on aging and induced changes in the arcuate nucleus and in the
medianosomes of the rat, Neuroendocrine System and Aging. Edited by Vezzadini P,
Facchini A, Labo G. Bologna, EURAGE, 1986, pp 13-25.

Fuxe K, O'Connor WT, Antonelli T, Osborne PG, Tanganelli S, Agnati LF and
Ungerstedt U: Evidence for a substrate of neuronal plasticity based on pre- and

postsynaptic neurotensin-dopamine receptor interactions in the neostriatum.
Proceedings of the National Academy of Sciences 1992; 89: 5591-5595.

Gaspar P, Berger B and Febvret A: Neurotensin innervation of the human cerebral cortex:
lack of colocalization with catecholamines. Brain Research 1990; 530: 181-195.

Goedert M, Pittaway K, Williams BJ and Emson PC: Specific binding of tritiated
neurotensin to rat brain membranes: Characterization and regional distribution. Brain
Research 1984; 304: 71-81.

Goedert M, Lightrnan SL, Mantyh PW, Hunt SP and Emson PC: Neurotensin-like
immunoreactivity and neurotensin receptors in the rat hypothalamus and in the
neurointerrnediate lobe of the pituitary gland. Brain Research 1985; 358: 59-69.

Gopalan C, Tian Y, Moore KB and Lookingland KJ: Neurochemical evidence that the
inhibitory effect of galanin on tuberoinfundibular dopamine neurons is activity
dependent Neuroendocrinology 1993; 58: 287-293.

Gout PW, Beer CT and Noble RL: Prolactin-stimulated growth of cell cultures

established from malignant Nb rat lymphomas. Cancer Research 1980; 40: 2433-
2436.

Granier C, van Rietschoten J, Kitabgi P, Poustis C and Freychet P: Synthesis and
characterization of neurotensin analogues for structure/activity relationship studies.
Acetyl-neurotensin-(8--l3) is the shortest analogue with full binding and
pharmacological activities. European Journal of Biochemistry 1982; 124: 117-124.

Gudelsky GA and Moore KE: Differential drug effects on dopamine concentrations and
rates of turnover in the median eminence, olfactory tubercle and corpus striatum.
Journal of Neural Transmission 1976; 38: 95-105.

Gudelsky GA and Moore KB: A comparison of the effects of haloperidol on dopamine
turnover in the striatum, olfactory tubercle and median eminence. Journal of
Pharmacology and Experimental Therapeutics 1977; 202: 149-156.

Gudelsky GA, Annunziato L and Moore KE: Localization of the site of the haloperidol-
induced, prolactin-mediated increase of dopamine turnover in the median eminence:
studies in rats with complete hypothalamic deafferentations. Journal of Neural
Transmission 1978; 181-192.

192

Gudelsky GA and Porter JC: Release of dopamine from tuberoinfundibular neurons into
pituitary stalk blood after prolactin or haloperidol administration. Endocrinology
1980; 106: 526-529.

Gudelsky GA and Porter JC: Sex-related difference in the release of dopamine into
hypophysial portal blood. Endocrinology 1981; 109: 1394-1398.

Gudelsky GA, Berry SA and Meltzer HY: Neurotensin activates tuberoinfimdibular
dopamine neurons and increases serum corticosterone concentrations in the rat.
Neuroendocrinology 1989; 49: 604-609.

Gully D, Canton M, Boigegrain R, Jeanjean F, Molimard J-C, Poncelet M, Gueudet C,
Heaulme M, Brouard A, Pelaprat D, Labbe-Jullie C, Mazella J, Soubrie P, Maffrand
J -P, Rostene W, Kitabgi P and LeFur G: Biochemical and pharmacological profile of
a potent and selective non-peptide antagonist of the neurotensin receptor. Proceedings
of the National Academy of Sciences 1993; 90: 65-69.

Gully D, Jeanjean F, Poncelet M, Steinberg R, Soubrie P, LeFur G and Maffrand J-P:
Neuropharmacological profile of non-peptide neurotensin antagonists. Fundamentals
of Clinical Pharmacology 1995; 9: 513-521.

Guo H-F, Tian J, Wang X, Fang Y, Hou Y and Han J: Brain substrates activated by
electroacupuncture (EA) of different frequencies (11): role of F os/Jun proteins in EA-

induced transcription of preproenkephalin and preprodynorphin genes. Molecular
Brain Research 1996; 43: 167-173.

Halasz B, Fuxe K, Agnati LF, Kalia M, Goldstein M, Andersson K, Harfstrand A and
Clark BJ: The Dopaminergic System. Berlin, Springer-Verlag, 1985, pp 1-39.

Harris JA: Using c-fos as a neural marker of pain. Brain Research Bulletin 1998; 45: 1-8.

Harrison RJ, McNeil GP and Dobner PR: Synergistic activation of neurotensin/
neuromedin N gene expression by c-Jun and glucocorticoids: novel effects of Fos
family proteins. Molecular Endocrinology1995; 9: 981-993.

Hentschel K, Cheung S, Moore KB and Lookingland KJ: Pharmacological evidence that
neurotensin mediates prolactin-induced activation of tuberoinfundibular dopamine
neurons. Neuroendocrinology 1998; 68: 71-76.

Hentschel K, Fleckenstein AE, Toney TW, Lawson DM, Moore KB and Lookingland KJ:
Prolactin regulation of tuberoinfundibular dopaminergic neurons: immunoneutraliza-
tion studies. Brain Research 2000a; 852: 28-36.

Hentschel K, Moore KB and Lookingland KJ: Effects of prolactin on expression of Fos-

related antigens in tyrosine hydroxylase-immunoreactive neurons in subdivisions of
the arcuate nucleus. Brain Research 2000b; 857: 110-118.

193

Herbison AE, Hubbard J1 and Sirett NE: Neurotensin excites neurons in the arcuate
nucleus of the rat hypothalamus in vitro. Brain Research 1986; 364: 391-395.

Herve D, Tassin JP, Studler JM, Dana C, Kitabgi P, Vincent J-P, Glowinski J and
Rostene W: Dopaminergic control of 125I-labeled neurotensin binding site density in

corticolimbic structures of the rat brain. Proceedings of the National Academy of
Sciences 1986; 83: 6203-6207.

Hesketh R: FOS, FRAs and JUN, The Oncogene Facts Book. London, Academic Press,
1995, pp 105-160.

Hoffman GE, Smith MS and Fitzsimmons MD: Detecting steriodal effects on immediate
early gene expression in the hypothalamus. Neuroprotocols 1992; 1: 52-66.

Hoffman GE, Smith MS and Verbalis JG: c-Fos and related immediate early gene
products as markers of activity in neuroendocrine systems. Frontiers in
Neuroendocrinology 1993a; 14: 173-213.

Hoffman GE, Lee W-S, Smith MS, Abbud R, Roberts MM, Robinson AG and Verbalis
JG: c-Fos and F os-related antigens as markers for neuronal activity: perspectives
from neuroendocrine systems. NIDA Research Monograph 1993b; 125: 117-133.

Hoffman GE, Le W-W, Abbud R, Lee W-S and Smith MS: Use of fos-related antigens
(FRAs) as markers of neuronal activity: FRA changes in dopamine neurons during ‘
proestrus, pregnancy and lactation. Brain Research 1994; 654: 207-215.

Hoffman GE and Murphy AZ: Anatomical markers of activity in hypothalamic systems,
Neuroendocrinology in Physiology and Medicine. Totowa, Humana Press Inc, 2000,
pp 541-552.

Hokfelt T and F uxe K: Effects of prolactin and ergot alkaloids on the tubero-infundibular
dopamine (DA) neurons. Neuroendocrinology 1972; 9: 100-122.

Hokfelt T, Everitt BJ, Theodorsson-Norheim E and Goldstein M: Occurrence of
neurotensin-like immunoreactivity in subpopulations of hypothalamic,

mesencephalic, and medullary catecholarnine neurons. J oumal of Comparative
Neurology 1984; 222: 543-559.

Hughes P and Dragunow M: Induction of irnmediate-early genes and the control of
neurotransmitter-regulated gene expression within the nervous system.
Pharmacological Reviews 1995; 47: 133-178.

Hsu S-M, Raine L and Fanger H: Use of avidin-biotin-peroxidase complex (ABC) in

immunoperoxidase techniques: a comparison between ABC and unabaled antibody
(PAP) procedures. Journal of Histochemistry and Cytochemistry 1981; 29: 577-580.

194

Hsu S-M and Soban E: Color modification of diaminobenzidine (DAB) precipitation by
metallic ions and its application for double immunohistochemistry. Journal of
Histochemistry and Cytochemistry 1982; 30: 1079-1082.

Ibata Y, F ukui K, Okamura H, Kawakami T, Tanaka M, Obata HL, Tsuto T, Terubayashi
H, Yanaihara C and Yanaihara N: Coexistence of dopamine and neurotensin in

hypothalamic arcuate and periventricular neurons. Brain Research 1983; 269: 177-
179.

Ibata Y, Kawakami F, Fukui K, Okamura H, Obata-Tsuto HL, Tsuto T, Terubayashi H:
Morphological survey of neurotensin-like immunoreactive neurons in the
hypothalamus. Peptides 1984; 5 Suppl 1:109-20: 109-120.

Icard-Liepkalns C, Biguet NF, Vyas S, Robert JJ, Sassone-Corsi P and Mallet J: AP-l
complex and c-fos transcription are involved in TPA provoked and trans-synaptic
inductions of the tyrosine hydroxylase gene: Insights into long-term regulatory
mechanisms. Journal of Neuroscience Research 1992; 32: 290-298.

J ennes L, Stumpf WE and Kalivas PW: Neurotensin: Topographical distribution in rat

brain by immunohistochemistry. Journal of Comparative Neurology 1982; 210: 211-
224.

Johnston CA, Demarest KT and Moore KE: Cyclohexirnide disrupts the prolactin-
mediated stimulation of dopamine synthesis in tuberoinfundibular neurons. Brain
Research 1980; 195: 236-240.

Kahn D, Abrams GM, Zimmerman EA, Carraway RE and Leeman SE: Neurotensin

neurons in the rat hypothalamus: An immunocytochemical study. Endocrinology
1980; 107: 47-54.

Kawano H and Daikoku S: Functional topography of the rat hypothalamic dopamine
neuron systems: retrograde tracing and immunohistochemical study. Journal of
Comparative Neurology 1987; 265: 242-253.

Kelly PA, Djiane J and Edery M: Different forms of the prolactin receptor: Insights into
the mechanism of prolactin action. Trends in Endocrinology and Metabolism 1992; 3:
54-59.

Kislauskis E, Bullock BP, McNeil S and Dobner PR: The rat gene encoding neurotensin
and neuromedin N. Journal of Biological Chemistry 1988; 263: 4963-4968.

Kislauskis E and Dobner PR: Mutually dependent response elements in the cis-regulatory
region of the neurotensin/neuromedin N gene integrate environmental stimuli in PC12
cells. Neuron 1990; 4: 783-795.

Kiss A, Palkovits M, Antoni F, Eskay RL and Skirboll LR: Neurotensin in the rat median

eminence: The possible sources of neurotensin-like fibers and varicosities in the
external layer. Brain Research 1987; 416: 129-135.

195

Kitabgi P, Checler F, Mazella J and Vincent JP: Pharmacology and biochemistry of

neurotensin receptors. Reviews in Clinical and Basic Pharmacology 1985; 5: 397-
486.

Kitabgi P, Rostene W, Dussaillant M, Schotte A, Laduron PM and Vincent J-P: Two
populations of neurotensin binding sites in murine brain: discrimination by the
antihistamine levocabastine reveals markedly different radioautographic distribution.
European Journal of Pharmacology 1987; 140: 285-293.

Kiyama H and Emson PC: Colchicine-induced expression of proneurotensin mRNA in
rat striatum and hypothalamus. Molecular Brain Research 1991; 9: 353-358.

Kobayashi H and Matsui T: Fine structure of the median eminence and its functional
significance. Frontiers in Neuroendocrinology. Edited by Martini L, Ganong WF.
New York, Raven Press, 1969, pp 3-46.

Komori K, Fujii T and Nagatsu I: Do some tyrosine hydroxylase-immunoreactive
neurons in the human ventrolateral arcuate nucleus and globus pallidus produce only
L-DOPA? Neuroscience Letters 1991; 133: 203-206.

Kumer SC and Vrana KE: Intricate regulation of tyrosine hydroxylase activity and gene
expression. Journal of Neurochemistry 1996; 67: 443-462.

Lantos TA, Gorcs TJ and Palkovits M: Irnmunohistochemical mapping of neuropeptides

in the premamillary region of the hypothalamus in rats. Brain Research Reviews
1995; 20: 209-249.

Lawson DM, Sensui N, Haisenleder DH and Gala RR: Rat lymphoma cell bioassay for
prolactin: Observations on its use and comparison with radioimmunoassay. Life
Sciences 1982; 31: 3063-3070.

Le F, Cusack B and Richelson E: The neurotensin receptor: is there more than one
subtype? Trends in Pharmacological Sciences 1996; 17: 1-3.

Lee Y and Voogt J L: Feedback effects of placental lactogens on prolactin levels and fos-
related antigen immunoreactivity of tuberoinfundibular dopaminergic neurons in the
arcuate nucleus during pregnancy in the rat. Endocrinology 1999; 140: 2159-2166.

Lerant A, Herman ME and Freeman ME: Dopaminergic neurons of periventricular and
arcuate nuclei of pseudopregnant rats: Semicircadian rhythm in fos-related antigens

immunoreactivities and in dopamine concentration. Endocrinology 1996; 137: 3621-
3628.

Lerant A and Freeman ME: Dopaminergic neurons in periventricular and arcuate nuclei

of proestrous and ovariectomized rats: Endogenous diurnal rhythm of fos-related
antigens expression. Neuroendocrinology 1997; 65: 436-445.

196

Lin J-Y and Pan J-T: Bombesin and neurotensin excite neurons in hypothalamic arcuate
nucleus brain slices: an extracellular single-unit study. Brain Research Bulletin 1993;
30: 177-180.

Lindley SE, Gunnet JW, Lookingland KJ and Moore KE: Effects of alterations in the
activity of tuberohypophysial dopaminergic neurons on the secretion of a-melanocyte

stimulating hormone. Proceedings of the Society for Experimental Biology and
Medicine 1988; 188: 282-286.

Lindley SE, Lookingland K1 and Moore KE: Activation of tuberoinfundibular but not
tuberohypophysial dopaminergic neurons following intracerebroventricular

administration of a-melanocyte-stimulating hormone. Neuroendocrinology 1990a;
51: 394-399.

Lindley SE, Gunnet JW, Lookingland KJ and Moore KE: 3,4-Dihydroxyphenylacetic
acid concentrations in the intermediate lobe and neural lobe of the posterior pituitary

gland as an index of tuberohypophysial dopaminergic neuronal activity. Brain
Research 1990b; 506: 133-138.

Lookingland KJ and Moore KE: Dopamine receptor-mediated regulation of

incertohypothalamic dopaminergic neurons in the male rat. Brain Research 1984;
304: 329-338.

Lookingland KJ, Farah JM, Lovell KL and Moore KE: Differential regulation of
tuberohypophysial dopaminergic neurons terminating in the intermediate lobe and in
the neural lobe of the rat pituitary gland. Neuroendocrinology 1985; 40: 145-151.

Lookingland KJ, Jarry HD and Moore KB: The metabolism of dopamine in the median
eminence reflects the activity of tuberoinfundibular neurons. Brain Research 1987a;
419(1-2): 303-310.

Lookingland KJ, Gunnet J W and Moore KE: Electrical stimulation of the arcuate nucleus
increases the metabolism of dopamine in terminals of tuberoinfundibular neurons in
the median eminence. Brain Research 1987b; 436: 161-164.

Lookingland K1 and Moore KE: Activation of tuberoinfundibular and tuberohypophysial
dopamine neurons following intra-cerebroventricular administration of bombesin.
Brain Research 1991; 565: 142-147.

Lowry OH, Rosebrough NJ, Farr AL and Randall RJ: Protein measurement with the folin
phenol reagent. Journal of Biological Chemistry 1951; 153: 265-275.

Manzanares J, Lookingland KJ and Moore KE: Kappa opioid receptor-mediated

regulation of dopaminergic neurons in the rat brain. Journal of Pharmacology and
Experimental Therapeutics 1991; 256: 500-505.

197

Manzanares J, Wagner EJ, Lookingland KJ and Moore KE: Effects of
immunoneutralization of dynorphinl.” and dynorphin”; on the activity of central
dopaminergic neurons in the male rat. Brain Research 1992; 587: 301-305.

Marcos P, Corio M, Covenas R and Tramu G: Neurotensin and tyrosine hydroxylase in
the tuberoinfundibular system of the guinea pig hypothalamus with special emphasis
to lactation status. Peptides 1996a; 17: 139-146.

Marcos P, Corio M, Dubourg P and Tramu G: Reciprocal synaptic connections between
neurotensin- and tyrosine hydroxylase-immunoreactive neurons in the mediobasal
hypothalamus of the guinea pig. Brain Research 1996b; 715: 63-70.

Marsault R and Frelin C: Activation by nitric oxide of guanylate cyclase in endothelial
cells from brain capillaries. Journal of Neurochemistry 1992; 59: 942-945.

Mazella J, Poustis C, Labbe C, Checler F, Kitabgi P, Granier C, Van Rietschoten J and
Vincent J-P: Monoiodo-[Trpl 1]neurotensin, a highly radioactive ligand of
neurotensin receptors. Journal of Biological Chemistry 1983; 258: 3476-3481.

Mazella J, Kitabgi P and Vincent J -P: Molecular properties of neurotensin receptors in rat
brain. Journal of Biological Chemistry 1985; 260: 508-514.

McDermott JR, Virrnani MA, Turner JD and Kidd AM: Peptidases involved in the
catabolism of neurotensin: Inhibitor studies using superfused rat hypothalamic slices.
Peptides 1986; 7: 225-230.

Meister B, Hokfelt T, Steinbusch HWM, Skagerberg G, Lindvall O, Geffard M, Joh TH,
Cuello AC and Goldstein M: Do tyrosine hydroxylase-immunoreactive neurons in the
ventrolateral arcuate nucleus produce dopamine or only L-DOPA? Journal of
Chemical Neuroanatomy 1988a; 1: 59-64.

Meister B and Hokfelt T: Peptide- and transmitter-containing neurons in the mediobasal
hypothalamus and their relation to GABAergic systems: possible roles in control of
prolactin and growth hormone secretion. Synapse 1988b; 2: 585-605.

Meister B, Ceccatelli S, Hokfelt T, Anden NE, Anden M and Theodorsson E:
NeurotranSmitters, neuropeptides and binding sites in the rat mediobasal
hypothalamus: effects of monosodium glutamate (MSG) lesions. Experimental Brain
Research 1989; 76: 343-368.

Merchant KM: C-fos Antisense oligonucleotide specifically attenuates haloperidol-
induced increases in neurotensin/neuromedin N mRNA expression in rat dorsal
striatum. Molecular and Cellular Neurosciences 1994; 3: 336-344.

Merchenthaler I and Lennard DE: The hypophysiotropic neurotensin-immunoreactive
neuronal system of the rat brain. Endocrinology 1991; 129: 2875-2880.

198

 

Milenkovic L, Parlow AF and McCann SM: Physiological significance of the negative
short-loop feedback of prolactin. Neuroendocrinology 1990; 52: 389-392.

Miller JC: Induction of c-fos mRNA expression in rat striatum by neuroleptic drugs.
Journal of Neurochemistry 1990; 54: 1453-1455.

Mills DE and Ward RP: Effect of prolactin on blood pressure and cardiovascular
responsiveness in the rat. Proceedings of the Society for Experimental Biology and
Medicine 1986; 181: 3-8.

Misu Y, Goshima Y, Ueda H and Okamura H: Neurobiology of L-DOPAergic systems.
Progress in Neurobiology 1996; 49: 415—454.

Moore KE, Demarest KT and Johnston CA: Influence of prolactin on dopaminergic
neuronal systems in the hypothalamus. FASEB- Federation Proceedings 1980; 39:
2912-2916.

Moore KE, Riegle GD and Demarest KT: Effects of long-term treatment with estradiol
and haloperidol on serum concentrations of prolactin and tuberoinfundibular
dopaminergic neuronal activity. Prolactin, Basic and clinical correlates. Edited by
MacLeod RM, Thorner MO and Scapagnini U. Padova, Liviana Press, 1985, pp 543-
549.

Moore KE, Demarest KT and Lookingland. KJ: Stress, prolactin and hypothalamic
dopaminergic neurons. Neuropharrnacology 1987a; 26: 801-808.

Moore KE: Hypothalamic Dopaminergic Neuronal Systems, Psychopharmacology: The
Third Generation of Progress. Edited by Meltzer HY. New York, Raven Press, 1987b,
pp 127-139.

Moore KE: Differential Regulation of Dopaminergic Neurons in the Mammalian Brain:
A Brief Review. Chinese Journal of Physiology 1992; 35: 67-76.

Moore KB and Lookingland KJ: Dopaminergic neuronal systems in the hypothalamus,
Psychopharrnacology: The Fourth Generation of Progress. Edited by Bloom FE,
Kupfer DJ. New York, Raven Press, 1995, pp 245-256.

Morgan J1 and Curran T: Stimulus-transcription coupling in neurons: role of cellular
immediate-early genes. Trends in Neuroscience 1989; 12: 459-462.

Nguyen TV, Kosofsky BE, Birnbaum R, Cohen BM and Hyman SE: Differential
expression of c-fos and zif268 in rat striatum after haloperidol, clozapine, and
amphetamine. Proceedings of the National Academy of Sciences 1992; 89: 4270-
4274.

199

Nicot A, Bidard JN, Kitabgi P, Lhiaubet AM, Masuo Y, Palkovits M, Rostene W and
Berod A: Neurotensin and neuromedin N brain levels after fornix transection:

evidence for an efficient neurotensin precursor processing in subicular neurons. Brain
Research 1995; 702: 279-283.

Niimi M, Takahara J, Sato M and Kawanishi K: Neurotensin and growth horrnone-
releasing factor-containing neurons projecting to the median eminence of the rat: a

combined retrograde tracing and immunohistochemical study. Neuroscience Letters
1991;133:183-186.

Niswender GD, Midgley AR, Jr., Monroe SE, Reichert LE, Jr.: Radioimmunoassay for
rat luteinizing hormone with antiovine LH serum and ovine LH-ml. Proceedings of
the Society for Experimental Biology and Medicine 1968; 128: 807-811.

Noble RL, Beer CT and Gout PW: Evidence in vivo and in vitro of a role for the pituitary

in the growth of malignant lymphomas in Nb rats. Cancer Research 1980; 40: 2437-
2440.

Nowycky M and Roth RH: Presynaptic dopamine receptors. Development of
supersensitivity following treatment with fluphenazine decanoate. Naunyn-
Schmiedeberg's Archives of Pharmacology 1977; 300: 247-254.

Okamura H, Kitahama K, Nagatsu I and Geffard M: Comparative topography of
dopamine- and tyrosine hydroxylase-immunoreactive neurons in the rat arcuate
nucleus. Neuroscience Letters 1988a; 95: 347-353.

Okamura H, Kitahama K, Mons N, Ibata Y, Jouvet M and Geffard M: L-DOPA-

immunoreactive neurons in the rat hypothalamic tuberal region. Neuroscience Letters
1988b; 95: 42-46.

Okuma Y and Osumi Y: Neurotensin-induced release of endogenous noradrenaline from
rat hypothalamic slices. Life Sciences 1982; 30: 77-84.

Palkovits M: Isolated removal of hypothalamic or other brain nuclei of the rat. Brain
Research 1973; 59: 449-450.

Palkovits M: Isolated removal of hypothalamic nuclei for neuroendocrinological and
neurochemical studies. Anatomical Neuroendocrinology 1974; 60: 72-80.

Palkovits M and Brownstein MJ: Microdissection of brain areas by the punch technique.
Brain Microdissection Techniques. Edited by Cuello AC. New York, John Wiley &
Sons, 1983, pp 1-36.

Palkovits M: Peptidergic neurotransmitters in the endocrine hypothalamus. Functional

Anatomy of the Neuroendocrine Hypothalarnus. Edited by Chadwick DJ and Marsh
JL. New York, John Wiley & Sons Inc., 1992, pp 3-15.

200

Pan J-T, Tian Y, Lookingland K1 and Moore KE: Neurotensin-induced activation of
hypothalamic dopaminergic neurons is accompanied by a decrease in pituitary
secretion of prolactin and a-melanocyte-stimulating hormone. Life Sciences 1992;
50: 2011-2017.

Pape J-R and Tramu G: Suckling-induced changes in neuropeptide Y and
proopiomelanocortin gene expression in the arcuate nucleus of the rat: Evaluation of
a putative intervention of prolactin. Neuroendocrinology 1996; 63: 540-549.

Parker SL, Armstrong WE, Sladek CD, Grosvenor CE and Crowley WR: Prolactin
stimulates the release of oxytocin in lactating rats: evidence for a physiological role
via an action at the neural lobe. Neuroendocrinology 1991; 53: 503-510.

Paxinos G and Watson C: The Rat Brain in Stereotaxic Coordinates, 2'"l Edition. New
York, Academic Press, 1986.

Pennypacker KR, Hong J-S, McMillian MK: Implications of prolonged expression of
fos-related antigens. Trends in Pharmacological Sciences 1995; 16: 317-322.

Pinnock RD: Neurotensin depolarizes substantia nigra dopamine neurones. Brain
Research 1985; 338: 151-154.

Poncelet M, Souilhac J, Gueudet C, Terranova JP, Gully D, LeFur G and Soubrie P:
Effects of SR-48692, a selective non-peptide neurotensin receptor antagonist, on two

dopamine-dependent behavioural responses in mice and rats. PsychOpharrnacology
1994; 116: 237-241.

Porter J C: Relationship of age, sex and reproductive status to the quantity of tyrosine

hydroxylase in the median eminence and superior cervical ganglion of the rat.
Endocrinology 1986; 118: 1426-1432.

Porter JC, Kedzierski W, Aguila-Mansilla N, Jorquera BA and Gonzalez HA: The
tuberoinfundibular dopaminergic neurons of the brain: hormonal regulation.
Advances in Experimental Medicine and Biology 1990; 274: 1-23.

Pugsley TA, Akunne HC, Whetzel SZ, Demattos SB, Corbin AE, Wiley JN, Wustrow
DJ, Wise LD and Heffner TG: Differential effects of the non-peptide neurotensin
antagonist, SR-48692, on the pharmacological effects of neurotensin agonists.
Peptides 1995; 16: 37-44.

Quirion R, Gaudreau P, St.Pierre S, Rioux F and Pert CB: Autoradiographic distribution
of [3H]neurotensin receptors in rat brain: visualization by tritium-sensitive film.
Peptides 1982; 3: 757-763.

Quirion R: Interactions between neurotensin and dopamine in the brain: an overview.
Peptides 1983; 4: 609-615.

201

Rethelyi M: Dendritic arborization and axon trajectory of neurons in the hypothalamic
arcuate nucleus of the rat - updated. Neuroscience 1985; 16: 323-331.

Reymond MJ and Porter JC: Hypothalamic secretion of dopamine after inhibition of
aromatic L-amino acid decarboxylase activity. Endocrinology 1982; 111: 1051-1056.

Reymond MJ, Speciale SG and Porter JC: Dopamine in plasma of lateral and medial
hypophysial portal vessels: evidence for regional variation in the release of

hypothalamic dopamine into hypophysial portal blood. Endocrinology 1983; 112:
1958-1963.

Robertson GS, Tetzlaff W, Bedard A, St Jean M and Wigle N: C-fos mediates
antipsychotic-induced neurotensin gene expression in the rodent striatum.
Neuroscience 1995; 67: 325-344.

Roky R, Obal F, Jr., Valatx J -L, Bredow S, Fang J, Pagano LP and Krueger JM: Prolactin
and rapid eye movement sleep regulation. Sleep 1995; 18: 536-542.

Roky R, Paut-Pagano L, Goffin V, Kitahama K, Valatx J-L, Kelly PA and Jouvet M:

Distribution of prolactin receptors in the rat forebrain. Neuroendocrinology 1996; 63:
422-429.

Rostene W, Alexander MJ: Neurotensin and neuroendocrine regulation. Frontiers in
Neuroendocrinology 1997; 18: 115-173.

Schotte A, Rostene W and Laudron PM: Different subcellular localization of neurotensin-
receptor and neurotensin-acceptor sites in the rat brain dopaminergic system. Journal
of Neurochemistry 1988; 50: 1026-1031.

Selmanoff M: The lateral and medial median eminence: distribution of dopamine,
norepinephrine, and luteinizing hormone-releasing hormone and the effect of
prolactin on catecholarnine turnover. Endocrinology 1981; 108: 1716-1722.

Selmanoff M: Rapid effects of hyperprolactinemia on basal prolactin secretion and
dopamine turnover in the medial and lateral median eminence. Endocrinology 1985;
116: 1943-1952.

Selmanoff M, Shu C, Hartman RD, Barraclough CA and Petersen SL: Tyrosine
hydroxylase and POMC mRN A in the arcuate region are increased by castration and
hyperprolactinemia. Molecular Brain Research 1991; 10: 277-281.

Sheng M and Greenberg ME: The regulation and function of c-fos and other immediate
early genes in the nervous system. Neuron 1990; 4: 477-485.

202

 

Skagerberg G, Meister B, Hokfelt T, Lindvall O, Goldstein M, Joh TH and Cuello AC:
Studies on dopamine-, tyrosine hydroxylase- and aromatic L-amino acid
decarboxylase-containing cells in the rat diencephalon: Comparison between

forrnaldehyde-induced histofluorescence and immunofluorescence. Neuroscience
1988; 24: 605-620.

Steinberg R, Brun P, Fournier M, Souilhac J, Rodier D, Mons G, Terranova JP, LeFur G
and Soubrie P: SR-48692, A non-peptide neurotensin receptor antagonist
differentially affects neurotensin-induced behaviour and changes in dopaminergic
transmission. Neuroscience 1994; 59: 921-929.

Svenningsson P, Georgieva J, Kontny E, Heilig M and Fredholm BB: Involvement of a c-
fos-dependent mechanism in caffeine-induced expression of the preprotachykinin A
and neurotensin/neuromedin N genes in rat striatum. European Journal of
Neuroscience 1997; 9: 2135-2141.

Tanaka T, Shiu RPC, Gout PW, Beer CT, Noble RL and Friesen HG: A new sensitive
and specific bioassay for lactogenic hormones: measurement of prolactin and grth

hormone in human serum. Journal of Clinical Endocrinology and Metabolism 1980;
51: 1058-1063.

Tanganelli 8, von Euler G, F uxe K, Agnati LF and Ungerstedt U: Neurotensin
counteracts apomorphine-induced inhibition of dopamine release as studied by
microdialysis in rat neostriatum. Brain Research 1989; 502: 319-324.

Tanganelli S, Li X-M, Ferraro L, Von Euler G, O'Connor WT, Bianchi C, Beani L and
Fuxe K: Neurotensin and cholecystokinin octapeptide control synergistically
dopamine release and dopamine D2 receptor affinity in rat neostriatum. European
Journal of Pharmacology 1993; 230: 159-166.

Tian Y, Lookingland KJ and Moore KE: Contribution of noradrenergic neurons to 3,4-
dihydroxyphenylacetic acid concentrations in the regions of the rat brain containing
incertohypothalamic dopaminergic neurons. Brain Research 1991; 555: 135-140.

Toney TW, Lookingland KJ and Moore KE: Role of testosterone in the regulation of
tuberoinfundibular dopaminergic neurons in the male rat. Neuroendocrinology 1991;
54: 23-29.

Toney TW, Manzanares J, Moore KB and Lookingland KJ: Sexual differences in the
stimulatory effects of bombesin on tuberoinfundibular dopaminergic neurons. Brain
Research 1992a; 598: 279-285.

Toney TW, Pawsat DE, Fleckenstein AE, Lookingland K1 and Moore KE: Evidence that
prolactin mediates the stimulatory effects of estrogen on tuberoinfundibular dopamine
neurons in female rats. Neuroendocrinology 1992b; 55(3): 282-289.

203

Uhl GR and Snyder SH: Regional and subcellular distributions of brain neurotensin. Life
Sciences 1976; 19: 1827-1832.

Uhl GR, Kuhar MJ and Snyder SH: Neurotensin: Immunohistochemical localization in

rat central nervous system. Proceedings of the National Academy of Sciences 1977;
74: 4059-4063.

Van Der Beek EM, Swarts HJ and Wiegant VM: Central administration of antiserum to
vasoactive intestinal peptide delays and reduces luteinizing hormone and prolactin

surges in ovariectomized, estrogen-treated rats. Neuroendocrinology 1999; 69: 227-
237.

Van Loon GR, Shum A, George SR and Shin SH: Prolactin increases the activity of
tuberoinfundibular and nigroneostriatal dopamine neurons: prolactin antiserum
inhibits the haloperidol-induced increases in dopamine synthesis rates in median
eminence and striatum of rats. Brain Research Bulletin 1983; 10: 539-545.

Watson RE, Wiegand SJ, Clough RW and Hoffman GE: Use of cryoprotectant to

maintain long-term peptide immunoreactivity and tissue morphology. Peptides 1986;
7: 155-159.

Wagner EJ, Moore KB and Lookingland KJ: Sexual differences in N-methyl-D-aspartate
receptor-mediated regulation of tuberoinfundibular dopaminergic neurons in the rat.
Brain Research 1993; 611: 139-146.

Wagner EJ, Moore KB and Lookingland KJ: Neurochemical evidence that AMPA
receptor-mediated tonic inhibition of hypothalamic dopaminergic neurons occurs via
activation of inhibitory intemeurons. Brain Research 1994a; 660: 319-322.

Wagner EJ, Moore KE and Lookingland KJ: Non-NMDA receptor-mediated regulation
of hypothalamic dopaminergic neurons in the rat. European Journal of Pharmacology
1994b; 254: 105-112.

Wagner EJ, Goudreau JL, Moore KB and Lookingland KJ: GABAergic regulation of
tuberoinfundibular dopaminergic neurons in the male rat. Brain Research 1994c; 659:
194-200.

Wagstaff JD, Gibb J W and Hanson GR: Dopamine D2-receptors regulate neurotensin
release from nucleus accumbens and striatum as measured by in vivo microdialysis.
Brain Research 1996; 721: 196-203.

Watson MA, Yamada M, Yamada M, Cusack B, Veverka K, Bolden-Watson C and
Richelson E: The rat neurotensin receptor expressed in chinese hamster ovary cells

mediates the release of inositol phosphates. Journal of Neurochemistry 1992; 59:
1967-1970.

204

Widerlov E, Kilts CD, Mailman RB, Nemeroff CB, McCown TJ, Prange AJ, Jr. and
Breese GR: Increase in dopamine metabolites in rat brain by neurotensin. Journal of
Pharmacology and Experimental Therapeutics 1982; 222: 1-6.

Woulfe J, Lafortune L, de Nadai F, Kitabgi P, Beaudet A: Post-translational processing
of the neurotensin/neuromedin N precursor in the central nervous system of the ra --

II. Immunohistochemical localization of maturation products. Neuroscience 1994; 60:
167-1 81 .

Yamada M, Yamada M, Watson MA and Richelson E: Neurotensin stimulates cyclic
AMP formation in CHO-rNTR-IO cells expressing the cloned rat neurotensin
receptor. European Journal of Pharmacology 1993; 244: 99-101.

Young 111 SW and Kuhar MJ: Neurotensin receptor localization by light microscopic
autoradiography in rat brain. Brain Research 1981; 206: 273-285.

205

"Illllllllllllllllllf