.nohnfl

h...

. it".

.3

, .3.

'uv‘l 2?.‘1
65...?!

f.
I 331.134.}
9. I]: (so-.33..”

...

z. <

..\ Jam‘s;
“flaw v.3...”
....;..3hiz,..u.
z huffinfin

4
l 1.5:: I ll.“ I

:y .r ..: V 5111..qu , . V
73%..mwg .ugafigfi. y. . ‘ .l. {.1}, igm'. l o .

. ., . a. 1:... , a: . :0 xx.’ .
.2: .21.. . gr!!! .3 _

4;; .na

 

This is to certify that the
dissertation entitled

NEURAL SUBSTRATES OF DIURNALITY IN THE NILE
GRASS RAT, ARVICANTHIS NILOTICUS

presented by

Michael David Schwartz

has been accepted towards fulfillment
of the requirements for the

Doctoral degree in Neuroscience

 

 

// // , /

. v 'V ,4”: T't/W/ V 77% a w//
Major Professor’s Signature

I -’ / n ’f

i ’7 /./ .v ’1"

'I » IV .I A" L, ’
l " =

Date

 

MSU is an Affirmative Action/Equal Opportunity Institution

w__.___—__7

LIBRARY
Michigan State
University

 

 

 

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 p:/ClRC/DateDue.indd-p.1

 

 

NEURAL SUBSTRATES OF DIURNALITY IN THE NILE GRASS RAT,
ARVICANTHIS NILOTICUS

BY

Michael David Schwartz

A DISSERTATION
Submitted to
Michigan State University
In partial fulfillment of the requirements
For the degree of
DOCTOR OF PHILOSOPHY

Neuroscience Program

2006

ABSTRACT

NEURAL SUBSTRATES OF DIURNALITY IN THE NILE GRASS RAT,
ARVICANTHIS NILOTICUS

By
Michael David Schwartz

Circadian rhythms in behavior and physiology differ dramatically in diurnal
and nocturnal rodents. However, the neural basis for these differences is
unknown. A pacemaker located in the suprachiasmatic nucleus of the
hypothalamus (SCN) is responsible for generating and maintaining circadian
rhythms in mammals, but current research suggests that cellular and molecular
rhythms in the SCN of diurnal rodents are similar to those of nocturnal species.
The neural substrates of diurnal or nocturnal phase preference are thus likely to
reside downstream of the SCN. The experiments in this dissertation address this
issue by examining outputs of the circadian system in a diurnal rodent, the Nile
grass rat Arvicanthis niloticus. Specifically, experiments assessed the anatomy
and function of a major SCN target, the lower subparaventricular zone (LSPV).

In the first set of experiments, expression of the immediate early gene
cFos was found to be rhythmic in the LSPV of grass rats and nocturnal lab rats
housed in a light-dark cycle. However, the rising phase of this rhythm was
advanced by nearly 8 hours in grass rats compared to lab rats, and was
endogenous in grass rats, but not lab rats, when kept in constant darkness.
Differences in the timing and the persistence of rhythmic activity in the grass rat

LSPV suggest that this region could modulate circadian signals differently in

diurnal and nocturnal rodents.

In the second set of experiments, the efferent neuroanatomy of the grass
rat SCN and LSPV were characterized using microinjections of anterograde
tracers. The distribution of efferent fibers from the LSPV was nearly identical to
that of the SCN, as has been previously reported in nocturnal rodents.
Furthermore, both the SCN and LSPV projected directly onto gonadotropin-
releasing hormone neurons, whose rhythmic activity is reversed in grass rats and
lab rats. Thus, the efferent projections of the LSPV are consistent with a role in
modulating circadian signals. The similarity of SCN and LSPV projections also
suggests that both areas work together to communicate circadian information.

Finally, LSPV function was assessed by monitoring general activity
rhythms following neurotoxic lesions of the LSPV. Destruction of the LSPV
caused profound disturbances in free-running activity rhythms and in entrainment
of those rhythms, suggesting that signals from the grass rat LSPV contribute to
rhythmic precision and stability in grass rats. However, neither activity rhythms
nor the ability to entrain were abolished, indicating that the LSPV is not
necessary for expression of either a diurnal phase preference or activity rhythms.

In summary, the LSPV exhibits the basic functional and anatomical
characteristics necessary to communicate circadian signals originating in the
SCN. Functional circadian rhythms are thus likely to depend on contributions
from multiple oscillators that are ultimately coordinated by the master pacemaker,

and the LSPV is very likely to be one of those oscillators.

Copyright by
MICHAEL DAVID SCHWARTZ

2006

For my mother and father.

Acknowledgements

First and foremost, I want to thank my advisor and mentor, Dr. Laura
Smale, for her curiosity, enthusiasm, support, and her critical comments at every
step of the way. Thanks also to the other members of my guidance committee,
Dr. Antonio Nunez, Dr. Cheryl Sisk and Dr. John Johnson, whose perspectives
have shaped both my personal and professional growth.

Much of this work would not have been possible without my labmates in
the Smale and Nunez labs. Anna Baumgras, Joel Breen, Alexandra Castillo-
Ruiz, Andrea Kaiser, Dr. Gladys Martinez, Terri McElhinny, Dr. Chidambaram
Ramanathan, Jessica Schrader, Nicole Timm and Dr. Russell van Horn all have
a hand in this research, and I am lucky to have worked with them all. Special
thanks to Drs. Megan Mahoney and Josh Nixon, whose advice, encouragement
and friendship over the last 6 years has been, and continues to be, invaluable.

Of course, not all training occurs in the lab. From day one, the
Neuroscience Program welcomed me into a truly collaborative training
community, and its students, postdocs and faculty continue to be a major part of
my development. Thanks to David McFarlane for technical assistance, Dr.
Sharleen Sakai for tract-tracing advice, and Dr. Henryk Urbanski for providing the
gonadotropin-releasing hormone antibody.

Financial support for my research and training came from the National
Institutes of Health (RO1-MH53433, awarded to LS, and F31-MH070087,

awarded to MDS) and the National Science Foundation (lBN-0130977, awarded

vi

to L8). Additional support was provided by the Neuroscience Program and the
Graduate School at Michigan State University.
Thanks to my family and friends who have supported me over the years,

and thanks especially to Julia, for her love and her strength.

vii

TABLE OF CONTENTS

LIST OF TABLES ............................................................................. xi
LIST OF FIGURES .......................................................................... xii
CHAPTER 1 Introduction ................................................................. 1
General introduction .......................................................................................... 1
Cellular and molecular basis of mammalian circadian rhythms ......................... 1
The SCN is similar in diurnal and nocturnal rodents .......................................... 3
Outputs of the SCN ........................................................................................... 6
The lower subparaventricular zone .................................................................... 7
Overview of chapters ......................................................................................... 9

CHAPTER 2 Differences in the suprachiasmatic nucleus and
lower subparaventricular zone of diurnal and nocturnal rodents

........................................................................................................ 13
Introduction ...................................................................................................... 13
Experimental Procedures: ............................................................................... 16

Animals: ....................................................................................................... 16
General immunocytochemical procedure: .................................................... 16
cFos rhythms in LD: ..................................................................................... 18
cFos Rhythms in DD: ................................................................................... 20
Statistical analyses: ..................................................................................... 21
Anatomical characterization of the LSPV region: ......................................... 21
Results: ........................................................................................................... 23
cFos rhythms in LD: ..................................................................................... 23
cFos and CalB rhythms in DD: ..................................................................... 25
Distributions of cFos, VP, VIP and CalB in the SCN/LSPV: ......................... 27
Discussion: ...................................................................................................... 33
Neuroanatomy of the grass rat LSPV: ......................................................... 33
cFos rhythms in the SCN: ............................................................................ 35
cFos rhythms in the LSPV: ........................................................................... 37
Conclusions: ................................................................................................ 40

CHAPTER 3 Efferent projections of the suprachiasmatic nucleus
and lower subparaventricular zone in the Nile grass rat

(Arvicanthis niloticus) ................................................................... 43
Introduction ...................................................................................................... 43
Experimental Procedures ................................................................................ 45

Animals: ....................................................................................................... 45
Surgical procedures and tissue processing: ................................................. 45

viii

Microscopy: .................................................................................................. 47

Results ............................................................................................................ 49
Injection sites: .............................................................................................. 49
Injections centered in the SCN: .................................................................... 53

Rostral Paths: ........................................................................................... 53
Midline thalamic paths: ............................................................................. 56
Lateral Path: ............................................................................................. 58
Caudal paths: ........................................................................................... 58
Injections centered in the LSPV: .................................................................. 64
Rostral paths: ........................................................................................... 64
Midline thalamic paths: ............................................................................. 65
Lateral paths: ............................................................................................ 66
Caudal paths: ........................................................................................... 67

Discussion ....................................................................................................... 70
Common output pathways of the SCN and LSPV: ....................................... 71
Differences in efferents of the SCN and LSPV: ............................................ 74
Functional Implications: ............................................................................... 75
Conclusions: ................................................................................................ 77

CHAPTER 4 The suprachiasmatic nucleus and the lower
subparaventricular zone project onto gonadotropin-releasing
hormone neurons in the Nile grass rat (Arvicanthis niloticus) ..79

Introduction: ..................................................................................................... 79
Materials and Methods: ................................................................................... 82
Animals ........................................................................................................ 82
Surgical Procedures: .................................................................................... 83
Light-level analyses: .................................................................................... 84
Confocal analysis: ........................................................................................ 85
Results: ........................................................................................................... 86
The distribution of GnRH-ir cells: ................................................................. 86
BOA injection sites: ...................................................................................... 87
Appositions between GnRH-ir neurons and BDA-ir fibers: ........................... 88
Confocal imaging: ........................................................................................ 92
Discussion ....................................................................................................... 95

CHAPTER 5 The effects of chemical lesions of the lower
subparaventricular zone on general activity rhythms in the Nile

grass rat ....................................................................................... 101
Introduction .................................................................................................... 101
Experimental Procedures .............................................................................. 103

Animals: ..................................................................................................... 103
Surgical procedures: .................................................................................. 104
Assessment of lesions: .............................................................................. 105
Experimental Design: ................................................................................. 107
Data analysis: ............................................................................................ 109
Results: ......................................................................................................... 112

Histology: ................................................................................................... 1 12

Activity rhythms .......................................................................................... 117
Discussion: .................................................................................................... 126
Histology: ................................................................................................... 126
Free-running rhythms in LSPV-Iesioned animals: ...................................... 127
Entrainment in LSPV-lesioned animals: ..................................................... 129
Conclusions: .............................................................................................. 131
CHAPTER 6 Conclusions ............................................................ 133
References ................................................................................... 141

LIST OF TABLES

Table 4.1: Percentages of GnRH-ir neurons that formed appositions with BDA-
labeled fibers. Values represent the percentage of the total number of GnRH
neurons ipsilateral or contralateral to the BDA injection site with appositions in
each of three rostrocaudal regions. ..................................................................... 93

Table 5.1: Total levels of activity, percentage of activity in the daytime, variance
in daily onset and variance in daily offset in each stage for LSPVx, MISS and
SHAM animals. * = significant main effect of stage; 1' = significant main effect of
group. ................................................................................................................ 116

Table 5.2: Correlations between the number of cFos-immunoreactive cells in the
SCN and LSPV and total levels of activity, amplitude of activity rhythms, and
percentage of activity in the daytime in each stage. * = significant correlation.123

xi

LIST OF FIGURES

Figure 2.1: Photomicrograph of the SCN and LSPV of a grass rat perfused at ZT
22 and stained for cFos-ir and Nissl. The SCN is outlined on the left, and the
sampling region used for LSPV counts is drawn to scale and placed above the
SCN. The section depicted here corresponds to the map of cFos-ir in the mid-
SCN of the series depicted in Fig. 6. 3v: third ventricle; ocx: optic chiasm. Scale
bar: 200 um. ...................................................................................................... 19

Figure 2.2: Numbers of cFos-positive cells in (A) the SCN, and (B) the LSPV of
grass rats (open circles) and lab rats (filled circles) kept in a 12:12 LD cycle.
Asterisks indicate time points where numbers of cFos-positive cells are
significantly elevated relative to other time points in (*) grass rats, (**) lab rats. .22

Figure 2.3: Photomicrograph of the SCN and LSPV of representative grass rats
(top row) and lab rats (bottom row) perfused at ZT 1 (left), ZT 13 (middle) orZT
23 (right) and processed for cFos-ir. 3v: third ventricle; ocx: optic chiasm. Scale
bar = 100 um. ...................................................................................................... 24

Figure 2.4: Counts of cells positive for cFos only (grey bars), CalB only (white
bars), and cells positive for both cFos and CaIB (black bars) in the SCN of A)
grass rats, B) lab rats, after 2-3 weeks in DD. Asterisks indicate time points
where numbers of cFos-positive cells (*) are significantly elevated relative to
other time points, p=.0015 ................................................................................... 26

Figure 2.5: Counts of cells positive for cFos only (grey bars), CalB only (white
bars), and cells positive for both cFos and CalB (black bars) in the SCN of A)
grass rats, B) lab rats, after 2-3 weeks in DD. Asterisks indicate time points
where numbers of cFos-positive cells (*) are significantly elevated relative to
other time points, p=.0015 ................................................................................... 28

Figure 2.6: Distributions of VP, VIP and cFos-ir through sections rostral to the
SCN, in the SCN, and caudal to the SCN of a grass rat perfused at ZT 22. 3v:

third ventricle; ocx: optic chiasm. Scale bar = 100 um. ...................................... 30

Figure 2.7: A) Photomicrograph of the SCN and LSPV of a grass rat perfused at
ZT 22 showing the distributions of cFos and CalB-ir; scale bar = 200 um. B)
Detail of the caudal SCN in the same animal. The section pictured here is the
one from which the map of cFos and CalB-ir in the caudal SCN are drawn in Fig.
8. Scale bar = 200 um. C) Double-labeled cells in the LSPV of the same animal
shown in (A) and (B) taken at 100x, showing cFos-ir (arrowheads), CalB-ir
(arrows), and double-labeled cells (asterisks); scale bar = 20 um. 3v: third
ventricle; ocx: optic chiasm. ................................................................................ 32

Figure 2.8: Distributions of cFos- and CaIB-ir in the SCN and LSPV of a grass rat
perfused at ZT 22. 3v: third ventricle; ocx: optic chiasm. Scale bar = 100 um. .34

xii

Figure 3.1: Line drawings depicting the location of biotinylated dextran amine
(BDA) injections in all cases described in the present study. See pages xv-xvi for
abbreviations ....................................................................................................... 48

Figure 3.2: Line drawings depicting the distribution of BDA- labeled fibers
following an injection centered in the suprachiasmatic nucleus (SCN; case 74).
Panels are arranged rostral-to-caudal, and were traced from sections as
described in text. See pages xv-xvi for abbreviations. ....................................... 50

Figure 3.3: Photomicrographs of BDA- labeled fibers in the vascular organ of the
lamina terminalis (OVLT; A, B) and anteroventral periventricular nucleus (AVPV;
C, D) of cases 74 (A, C) and 61 (B, D). * = 3rd ventricle. Scale bar = 100 pm. ..54

Figure 3.4: Photomicrographs of BDA- labeled fibers in the lateral septum (LS;
AB) and bed nuclei of the stria terminalis (BNST; CD) of cases 74 (A,C) and 61
(B, D). * = lateral ventricle. Scale bar = 100 pm. ............................................... 55

Figure 3.5: Photomicrographs of BDA- labeled fibers in the paraventricular
thalamic nucleus (PVT; A, B) and the centers of the injection sites (C, D) of cases
74 (A, C) and 61 (B, D). * = 3rd ventricle. Scale bar = 100 pm. ......................... 57

Figure 3.6: Photomicrographs of BDA- labeled fibers in the intergeniculate leaflet
of cases 74 (A) and 34 (B). Scale bar = 100 pm. ............................................... 59

Figure 3.7: Line drawings depicting the distribution of BDA- labeled fibers
following an injection centered in the lower subparaventricular zone (LSPV; case
34). Panels are arranged rostral-to-caudal, and were traced from sections as
described in text. See pages xv-xvi for abbreviations. ....................................... 61

Figure 3.8: Photomicrographs of BDA- labeled fibers in the paraventricular
hypothalamic nucleus (PVN; A, B), dorsal subparaventricular zone (sPVZ; C, D),
and dorsomedial hypothalamic nucleus (DMH; E-H) of cases 74 (A, C, E, G) and
61 (B, D, F, H). A detail of the DMH pars compacta (DMHpc) is pictured in
panels G and H. * = 3ml ventricle. Scale bar = 200 ,um (A-F), 100 pm (G, H). ...68

Figure 3.9: Photomicrographs of BDA- labeled fibers in the ventromedial
hypothalamic nucleus(VMH; AB) and adjacent ventral tuberal region (VTU; CD)
of cases 74 (A,C) and 34 (3,0). * = 3'd ventricle. Scale bar = 300 pm (A, B), 100
pm (C, D). ........................................................................................................... 69

Figure 4.1: Line drawings depicting the location of biotinylated dextran amine
(BDA) injections in all cases described in the present study. See pages xv-xvi for
abbreviations. ...................................................................................................... 89

Figure 4.2: Photomicrographs of appositions between BDA-ir fibers (black) and
GnRH neurons (brown; A - D) and processes (E) following injections centered in
the SCN. BDA—ir fiber in (E) is not the same fiber pictured in (D). All images
were taken using a 100x oil objective. Scale bar = 20 pm. ................................ 91

xiii

Figure 4.3: Photomicrographs of appositions between BDA-ir fibers (black) and
GnRH-ir neurons (brown; A — C) and processes (D) following injections centered
in the LSPV. GnRH neuron in (B) is the same neuron depicted in bottom center
of (A). All images were taken using a 100x oil objective. Scale bar = 20 um. ...94

Figure 4.4: Confocal laser fluorescent images of appositions between BDA-ir
fibers (green) and GnRH-ir cell bodies (red) following injections centered in the
LSPV. A - C: projections of serial optical sections. D: single optical section from
the cell depicted in (C). All images were taken using a 63x oil objective. Scale
bar = 20 um ........................................................................................................ 97

Figure 5.1: Photomicrographs of the SCN of a grass rat with a complete LSPV
lesion that was perfused at CT 22. Alternate sections are stained for (A) Nissl,
(B) cFos and (C) VP. Scale bar = 200 um. ...................................................... 108

Figure 5.2: Photomicrographs of the SCN and LSPV of a representative SHAM
(A, C) and LSPVx (B, D) grass rat perfused at ZT 1- ZT 2. Alternate sections are
stained for (A, B) Nissl and (C, D) cFos. Scale bar = 200 um. ......................... 111

Figure 5.3: Schematic diagrams illustrating lesion placement in animals classified
as LSPVx (A) and MISS (B). Atlas adapted from {Paxinos, 2005 #961}. ......... 114

Figure 5.4: Representative double-plotted actograms of a SHAM (A) and MISS
(B) grass rat. LD cycles are indicated at top, time of day (ZT) at bottom. In each
actogram, date of lesion surgery (NMA or SHAM) and changes in lighting
conditions (LD, DD) are indicated along the left side of each actogram ............ 118

Figure 5.5: Representative double-plotted actograms of three LSPVx animals, #
234 (A), # 241 (B) and # 254 (C). LD cycles are indicated at top, time of day (ZT)
at bottom. In each actogram, date of lesion surgery (NMA or SHAM) and
changes in lighting conditions (LD, DD) are indicated along the left side of each
actogram. .......................................................................................................... 1 19

Figure 5.6: Amplitude of activity rhythms of LSPVx (black), MISS (gray) and
SHAM (white) grass rats in each stage. * = significantly different from PREOP
within group. ** = significantly different from all other stages within group ....... 125

Figure 5.7: Number of days to re-entrain daily activity onset (left) and offset
(right) for LSPVx (black), MISS (gray) and SHAM (white) grass rats following
free-run. ............................................................................................................ 127

Images in this dissertation are presented in color.

xiv

INDEX OF ABBREVIATIONS

 

Abbreviation Definition

1' tau

20G 2-deoxyglucose

3v Third ventricle

ABC Avidin-biotin-peroxidase complex

ac Anterior commissure

ACB Nucleus accumbens

AHA Anterior hypothalamic area

AHC Anterior hypothalamic area, central part

AHP Anterior hypothalamic area, posterior part

ANOVA Analysis of variance

Arc Arcuate nucleus

AVPV Anteroventral periventricular nucleus

BDA Biotinylated dextran amine

BNST Bed nucleus of the stria terminalis

CalB Calbindin

CM Centromedial thalamic nucleus

Cry, 00/ Cryptochrome

CT Circadian time

D3v Dorsal third ventricle

DAB diaminobenzidine

DBB Diagonal band of Broca

DD Constant darkness

DMH Dorsomedial hypothalamic nucleus

ER Estrogen receptor

f Fornix

fr Fasciculus retroflexus

GnRH Gonadotropin-releasing hormone

hbc Habenular commissure

HDB Horizontal limb of the diagonal band of
Broca

ic Internal capsule

IGL lntergeniculate leaflet

-ir immunoreactive

LA Lateroanterior hypothalamic nucleus

LD Light: dark cycle

LGN Lateral geniculate nucleus

LH Leuteinizing hormone

LHA Lateral hypothalamic area

XV

LHb
LPO
LSi
LSPV
LSv
MeA
MHb
MMn
MnPO
MPO
mPOA
MS
mt
MUA
NMA
NDS
NGS
ocx
opt
OVLT
PBS
pc

Pe
PeF
Per, Per
PF
PH
PK2
PLP

PMD
PMV
POA
PrC
PVG
PVN
PVT
PT
RCh
Re
Rh
Rt
SCN
SCO
SHy
SM

Lateral habenula

Lateral preoptic area

Lateral septum, intermediate part
Lower subparaventricular zone
Lateral septum, ventral part

Medial amygdala

Medial habenula

Medial mammillary nucleus, median part
Median preoptic nucleus

Medial preoptic nucleus

Medial preoptic area

Medial septum

Mammillary tract

Multiple-unit activity
N-methyl-DL—aspartic acid

Normal donkey serum

Normal goat serum

Optic chiasm

Optic tract

Vascular organ of the lamina terminalis
Phosphate-buffered saline

Posterior commissure
Periventricular hypothalamus
Perifornical region

Penod

Parafascicular thalamic nucleus
Posterior hypothalamus

Prokineticin 2

4 "/0 paraformaldehyde with lysine and
sodium periodate

Premammillary nucleus, dorsal part
Premammillary nucleus, ventral part
Preoptic area

Pre-commissural nucleus
Periventricular gray

Hypothalamic paraventricular nucleus
Paraventricular thalamic nucleus
Parataenial thalamic nucleus
Retrochiasmatic area

Reuniens thalamic nucleus
Rhomboid thalamic nucleus
Reticular thalamic nucleus
Suprachiasmatic nucleus
Sub-commissural organ
Septohypothalamic nucleus
Nucleus of the stria medullaris

xvi

sm
SON
SPF
sPVZ
st
Sub
SuM
TMN
VDB
VIP
VLPO
VMH
VMPO
VP
VTU
Zl

Stria medullaris

Supraoptic nucleus
Subparafascicular nucleus
Subparaventricular zone

Stria terminalis

Submedius thalamic nucleus
Supramammillary nucleus
Tuberomammillary nuclei

Vertical limb of the diagonal band of Broca
Vasoactive intestinal polypeptide
Ventrolateral preoptic area
Ventromedial hypothalamic nucleus
Ventromedial preoptic area
Vasopressin

Ventral tuberal region

Zona incerta

Zeitgeber time

xvii

CHAPTER 1

Introduction

General introduction

Nearly all life forms studied to date exhibit circadian rhythms in behavior
and physiology (Dunlap et al., 2004). While neurobiological study into the
workings of the mammalian circadian clock proceeds at a lightning pace, most of
this research focuses on nocturnal rodent models such as the lab rat, mouse and
hamster. Diurnal animals (including humans) display dramatically different
circadian rhythms, and without a clear understanding of these differences, the
extent to which research on nocturnal rodents can be applied to diurnal species
is unclear. Yet, the mechanisms that underlie differences between diurnal and
nocturnal patterns of behavior and physiology are unknown (reviewed in Smale
et al., 2003). This introductory chapter will first review the neural mechanisms
underlying the fundamental circadian clock in mammals and present evidence
suggesting that it is the same in diurnal and nocturnal species. This will be
followed by a discussion of some of the ways in which responses to signals from
that clock could differ in diurnal and nocturnal rodents. Finally, summaries of the

research questions to be addressed in each chapter will be presented.

Cellular and molecular basis of mammalian circadian rhythms

Circadian rhythms, endogenous 24-hour cycles of behavioral,
physiological and molecular processes, are ubiquitous in nature. The first

indication that specific genes play central roles in the production of circadian

rhythms was the discovery of the period gene in the fruit fly Drosophila
melanogaster(Konopka and Benzer, 1971). In mammals, discovery of the tau
mutant hamster, which exhibited an abnormally short circadian period (Ralph
and Menaker, 1988), and the Clock mouse, which exhibited along I followed by
disintegration of rhythmicity (Vitaterna et al., 1994), was soon followed by
characterization of many more ‘clock genes’ and a model of how they generated
self-sustaining 24-hour oscillations in gene transcription and translation.
Although the details of these models are still changing, it is clear that in both
prokaryotic and eukaryotic organisms, circadian rhythmicity arises from the
interaction of positive and negative feedback loops in gene transcription and
translation (Dunlap, 1999; Bell-Pedersen et al., 2005). In mammals, the
transcription factors Clock and Bmal form the positive portion of the loop. Their
protein products dimerize, enter the nucleus and activate transcription of three
sets of genes. One of these, the mammalian ortholog of Period, contains three
variants, Per1, Per2, and Per3; the function of Per3 is unclear, but the first two
are intrinsic to the basic molecular clock. The other genes whose transcription is
activated by Clock and Bmal are the Cryptochrome genes (Cry1, 00/2), and Rev-
Erba. Protein products of Rev-Erba inhibit expression of Bmal (Reppert and
Weaver, 2002). The protein products of the Per and 00/ genes turn dimerize, re-
enter the nucleus and repress transcription of their own genes, forming the
negative loop, while the simultaneous repression of Rev-Erba leads to increases
in Bmal transcription (Reppert and Weaver, 2002). The resulting 24-hour

oscillations in Per1 and Per2 mRNA, and Per1 and Per2 protein are frequently

used as markers for the oscillation of the molecular clock and allow direct
observation of its phase.

In mammals, the hypothalamic suprachiasmatic nucleus (SCN) contains
the primary circadian pacemaker (reviewed in Weaver, 1998). Complete
electrolytic lesions of the SCN abolish circadian rhythmicity in a number of
species (Moore and Eichler, 1972; Stephan and Zucker, 1972), and the SCN
continues to oscillate after being isolated from other inputs in vivo (lnouye and
Kawamura, 1979) and in vitro (Green and Gillette, 1982; Yamazaki et al., 2000).
Furthermore, transplanting SCN tissue into the brain of an arrhythmic host
restores locomotor rhythmicity, and does so with the characteristics of the donor,
not the host (Ralph et al., 1990). These data and other findings have clearly
demonstrated that the SCN contains a self-sustaining circadian oscillator that is

necessary for the generation of circadian activity rhythms.

The SCN is similar in diurnal and nocturnal rodents

Theoretically, the SCN could represent a potential source of species
differences in the timing of circadian rhythms in behavior. However, this does not
appear to be the case, as in many respects the SCN is similar in diurnal and
nocturnal species. For example, SCN metabolic activity measured by 2-
deoxyglucose uptake (2DG) peaks during the subjective day in an array of
diurnal and nocturnal species (Schwartz, 1991), and rhythms in multiple-unit
electrical activity peak during the day in the SCN of the nocturnal lab rat and the
diurnal Siberian chipmunk (Kubota et al., 1981; Sato and Kawamura, 1984b;

Kurumiya and Kawamura, 1988). Furthermore, daily rhythms in expression of

the immediate early gene cFos are similar in the SCN of diurnal grass rats and
nocturnal lab rats (Katona et al., 1998; Nunez et al., 1999). These data suggest
that rhythmic activity in the SCN is similar at a general level in several diurnal
and nocturnal species. Nevertheless, species differences have been reported in
some features of SCN structure and function. For example, the SCN’s response
to stimulation by light is different in lab rats and diurnal Octodon degus (Krajnak
et al., 1997; Jiao et al., 1999), and responses to infusions of muscimol, a GABA-
A agonist, into the SCN are different in grass rats and hamsters (Novak and
Albers, 2004). It thus remains possible that mechanisms associated with
diurnality involve changes in specialized subpopulations of cells within the SCN.
However, in diurnal grass rats, rhythms in the SCN’s responsiveness to light are
similar to those of nocturnal species (Mahoney et al., 2001 ), and both light pulses
and muscimol injection have similar effects on Per expression in the SCN of
grass rats and hamsters (Novak et al., 2006). Taken together, the data suggest
that although some features of SCN function vary across species, most features
of SCN function and anatomy are similar in nocturnal and diurnal species.

Accumulating evidence suggests that one fundamental component of the
SCN that is the same across species is the core molecular clock that it contains.
The SCN of diurnal rodents such as grass rats, 13-lined ground squirrels and
Arvicanthis ansorgei express high levels of Per1 and Per2 mRNA in the
subjective day (Mrosovsky et al., 2001; Caldelas et al., 2003; Lambert et al.,
2005; Novak et al., 2006), as has been demonstrated in lab rats (Yan and

Okamura, 2002) mice (Hastings et al., 1999; Bae et al., 2001) and hamsters

(Hamada et al., 2004). These similarities are also seen in rhythms of Per1 and
Per2 proteins in the SCN of diurnal grass rats (Ramanathan et al., 2006) and
nocturnal lab rats (Amir et al., 2004), mice (Hastings et al., 1999; Field et al.,
2000) and hamsters (Nuesslein-Hildesheim et al., 2000). Although the molecular
clock includes many components whose function could conceivably differ, the
similarity in phase of Per mRNA and protein oscillations suggests that the
molecular clock is very likely to be similar in nocturnal and diurnal species.
The mechanisms coupling the molecular clock within the SCN to output
rhythms are poorly understood at the molecular level. However, several proteins
and peptides are emerging as candidate output molecules that communicate
clock information to systems regulating physiological and behavioral events.
Proteins such as prokinetlcin 2 (PK2; Cheng et al., 2002), transforming growth
factor-alpha (Kramer et al., 2001) and cardiotrophin-like cytokine (Kraves and
Weitz, 2006) may play roles in communicating circadian information to SCN
targets that regulate behavior. In addition, vasopressin (VP) and vasoactive
intestinal polypeptide (VIP) may function as SCN signals regulating
neuroendocrine rhythms. In all cases in which molecular outputs have been
examined in diurnal species they have been the same as those documented in
nocturnal ones (Reppert et al., 1983; Krajnak et al., 1998; Lambert et al., 2005;
Mahoney et al., 2006). Taken together, the data on the SCN of nocturnal and
diurnal species suggest that the primary mechanisms underlying diurnality are
likely to lie downstream of the SCN, in cells that receive direct or indirect input

from it.

Outputs of the SCN

The SCN appears to use a combination of signaling methods, paracrine
and synaptic, to communicate with its targets. Diffusible signals can induce 20G
and Per rhythms in vitro (Allen et al., 2001), and can restore locomotor rhythms
in arrhythmic hamsters in vivo (Silver et al., 1996a). However, SCN transplants
that reinstate locomotor rhythms in hamsters reinstate Per rhythms in some but
not other peripheral targets (Guo et al., 2006), and also fail to reinstate
neuroendocrine rhythms (Meyer-Bernstein et al., 1999). Thus, diffusible signals
are unable to sustain the full panoply of circadian rhythms seen in intact
hamsters, implying that neural efferents from the SCN are an essential aspect of
circadian output (de la lglesia et al., 2003; de la lglesia and Schwartz, 2006).
Virtually nothing is known about the roles of axonal projections of the SCN, or
diffusible signals that might be released by it, in the regulation of rhythms in
animals other than nocturnal rodents.

SCN efferents have been characterized using anterograde and retrograde
neural tracers in lab rats and hamsters (Watts and Swanson, 1987; Watts et al.,
1987; Morin et al., 1994; Leak and Moore, 2001; Kriegsfeld et al., 2004). In mice,
SCN projections have been examined indirectly via immunohistochemical
analysis of VP and VIP fibers originating in the SCN (Abrahamson and Moore,
2001). Although some species differences in efferent fiber distributions have
been reported (Morin et al., 1994), these studies have all found that the SCN of
lab rats, mice and hamsters projects widely within the hypothalamus. These
projections are especially concentrated in the medial hypothalamus, are present

but more moderate in the limbic forebrain and midline thalamus, and almost, if

not completely, absent in the midbrain, hindbrain and cerebral cortex. There are
no published data on the efferent projections of the SCN in diurnal rodents,
although at least some of them are likely to be similar (Mahoney and Smale,
2005b). A major goal of this thesis, therefore, is to evaluate whether these
projections differ in grass rats compared to lab rats or hamsters, as well as to
identify pathways by which circadian signals could reach target populations of

cells whose activity is inverted in diurnal and nocturnal species.

The lower subparaventricular zone

Mechanisms underlying diurnality could be located outside of the SCN
such that its output signals are the same in nocturnal and diurnal species but are
interpreted differently at, or downstream of, one or more SCN targets. The
subparaventricular zone (sPVZ) is one promising candidate for such a
downstream site. This is a relatively cell-sparse region adjacent to the
hypothalamic periventricular nucleus extending dorsally and caudally from the
dorsal border of the SCN to the ventral border of the hypothalamic
paraventricular nucleus (PVN; Watts et al., 1987; PVN; Morin et al., 1994; Smale
and Boverhof, 1999). The SCN of both nocturnal and diurnal species sends
massive AVP- and VIP-immunoreactive (-ir) projections into the sPVZ. In
nocturnal species, the sPVZ is known to project in turn to nearly all of the same
targets as the SCN (Watts et al., 1987; Morin et al., 1994; Kriegsfeld et al., 2004).
Based on these data, it has been suggested that circadian signals from the SCN

are amplified by the sPVZ en route to their targets (Watts and Swanson, 1987).

The lower part of the subparaventricular zone (LSPV) is located
immediately dorsal to the SCN and includes parts of the terminal field of VP- and
VlP-ir axons emanating from the SCN (Watts and Swanson, 1987; Watts et al.,
1987; Lu et al., 2001). However, while the latter includes the entire terminal field
extending dorsally and caudally to the PVN, the LSPV includes only the part of
the terminal field that is immediately dorsal to the SCN. Under a 12:12 light-dark
cycle (LD), cFos-ir in the SCN of diurnal grass rats, like that of the nocturnal lab
rats, peaks early in the light phase and declines steadily thereafter (Nunez et al.,
1999). However, cFos rhythms in the LSPV are very different in the two species.
In the grass rat LSPV, cFos rises sharply at Zeitgeber time (ZT) 17 1 and
remains high until several hours after lights-on. cFos-ir in the LSPV of lab rats,
by contrast, does not rise until 8 hours, later, at ZT 1. While the rising phase of
the LSPV and SCN rhythms are in phase in lab rats, this is not the case in grass
rats. In summary, the LSPV, a direct target of the SCN in both diurnal and
nocturnal species, exhibits a rhythm in cFos expression with a rising phase that
is advanced by approximately 9 hours in a diurnal species relative to a nocturnal
one and almost reversed from the rhythm seen in the SCN in a diurnal species,

but not a nocturnal one. It is not yet known whether these patterns reflect

 

‘ In circadian biology, time is denoted in either zeitgeber time (ZT) or circadian time (CT),
depending on the lighting conditions. Zeitgeber hours are equivalent to clock hours, and
referenced to the timing of the light-dark transition; in a 12:12 LD cycle, ZT 0 corresponds to
lights-on and activity onset for a diurnal animal, and ZT 12 corresponds to lights-off and activity
onset for a nocturnal animal. Under constant lighting conditions, circadian hours are referenced
to subjective daylength, where one circadian hour is equal to 1/24th of the organism’s free-running
circadian period (1:). CT 0 corresponds to the start of the subjective day and activity onset In a
diurnal animal, and CT 12 corresponds to the start of subjective night and activity onset for a
nocturnal animal. For a more detailed review see

Moore-Ede MC, Sulzman FM, Fuller CA (1982) The clocks that time us: physiology of the
circadian timing system. Cambridge. Mass.: Harvard University Press.

endogenous mechanisms as the data described above were all collected from
animals maintained in LB conditions. However, these data raise the possibility
that the LSPV might modulate circadian signals originating in the SCN, and do so

differently in diurnal and nocturnal species.

Overview of chapters

The experiments in this thesis, broadly speaking, will explore output
mechanisms used by the circadian system, whether these mechanisms might
differ in some way in diurnal and nocturnal rodents, and, if so, how those
differences might contribute to determining whether an animal exhibits a diurnal
or nocturnal pattern of rhythmicity. The latter could theoretically occur in several
ways. Circadian output signals sent by the SCN to other brain regions could
differ in diurnal and nocturnal rodents. Although differences could still exist that
have not yet been identified, the phases of rhythms that are intrinsic to the
molecular clock, and of the rhythms in SCN output signals that have been
Identified, are similar in diurnal and nocturnal species (see above). Existing data
thus suggest that the key to diurnality lies beyond the SCN. A very plausible
alternative is that SCN targets of diurnal and nocturnal species respond
differently to circadian signals sent to them by the SCN. This thesis will evaluate
a specific variant of this hypothesis that has emerged from consideration of the
results reviewed above. In this model, circadian signals originating within the
SCN are modulated by the LSPV in different ways in nocturnal and diurnal
species. This leads, in turn, to different patterns of coupling between the SCN

and its other downstream targets. This then leads to reversed circadian patterns

of intracellular rhythms in those target cells, and to reversed patterns of
behavioral rhythms in diurnal and nocturnal species. Several predictions emerge
from this hypothesis:

Predictiom: If the LSPV is necessary for maintaining diurnal patterns of
rhythmicity, then its rhythmic activity should be endogenous. In chapter two, this
prediction is tested by comparing rhythms in cFos expression in grass rats and
lab rats kept in constant darkness. Additional experiments evaluate the time
course of cFos expression in the LSPV when animals are kept in an LD cycle,
and further characterize the neurochamical anatomy of the LSPV.

Predictionzz If the LSPV provides a signal to areas of the brain important
for maintaining circadian rhythms, then it should have some mechanism of
communicating with those areas. This prediction is tested in studies described in
chapters three and four. In chapter three, injections of the anterograde tracer
biotinylated dextran amine (BDA) are used to map the efferent projections of both
the SCN and the LSPV of grass rats. Chapter four examines whether the SCN
and/or the LSPV project directly onto a population of cells whose rhythmic activity
is reversed in diurnal and nocturnal species by comparing appositions between
gonadotropin-releasing hormone neurons and fibers from the SCN and LSPV.

Predictionaz If the LSPV is a necessary component for the maintenance of
diurnal patterns of physiological and behavioral rhythms, its destruction should
significantly alter or abolish these rhythms. Chapter five evaluates this prediction
by examining general activity rhythms in grass rats with neurotoxic lesions of the

LSPV that spare both the SCN and its efferent fibers.

10

In lab rats, the region dorsal to the SCN appears to be an important target
of the SCN that plays a key role in regulation of a variety of rhythms (Watts and
Swanson, 1987; Watts et al., 1987; Saper et al., 2005). These include rhythms in
neuroendocrine processes associated with estrous cyclicity (Arendash and Gallo,
1979; Docke et al., 1982; Watts et al., 1989; Fernandez-Galaz et al., 1999),
sleep-wake cycles (Lu et al., 2001) and locomotor rhythms (Lu et al., 2001;
Moore and Danchenko, 2002; Abrahamson and Moore, 2006). The experiments
described in this thesis will help to identify features of this output system that may
generalize from lab rats to other species, including ones showing very different
patterns of rhythmicity. These studies will also evaluate ways in which the LSPV
differs across species and test the hypothesis that it contributes to emergence of
a diurnal phase preference. If the LSPV does not specifically mediate diurnality
in the grass rat, it may still be integrally involved in output of SCN timing signals,
in the production of rhythms, or in the interaction between the clock and rhythmic
behavior. These experiments will therefore make important contributions to our
understanding of diurnality specifically as well as of circadian output systems

more generally.

11

CHAPTER 2

Schwartz MD, Nunez AA, Smale L (2004) Differences in the suprachiasmatic
nucleus and lower subparaventricular zone of diurnal and nocturnal rodents.
Neuroscience 127:13-23.

12

CHAPTER 2
Differences in the suprachiasmatic nucleus and lower

subparaventricular zone of diurnal and nocturnal rodents

Introduction

The suprachiasmatic nucleus (SCN) of the hypothalamus is the site of the
principal circadian pacemaker in mammals (Moore and Leak, 2001), and in many
respects the SCN is similar in diurnal and nocturnal mammals. For example,
rhythms in multiple-unit activity (MUA) peak during the day in the SCN of both lab
rats and the diurnal Siberian chipmunk (Kubota et al., 1981; Sato and Kawamura,
1984b; Kurumiya and Kawamura, 1988), and glucose utilization measured by
uptake of 2-deoxyglucose (2DG) peaks during the subjective day in both
nocturnal and diurnal mammals (reviewed in Schwartz, 1991). Diurnal rodents
such as the 13-lined ground squirrel and Arvicanthis ansorgei express high levels
of mRNA for the clock genes per1 and per2 in the SCN in the subjective day
(Mrosovsky et al., 2001; Caldelas et al., 2003), as has been demonstrated in the
lab rat (Yan and Okamura, 2002) and mouse (Hastings et al., 1999; Bae et al.,
2001). Overall these data imply that, with respect to some very basic functions,
the SCN is more similar than different in diurnal and nocturnal species.

However, diurnal Octodon degus differ from lab rats with respect to the pattern of
responsiveness of SCN cells to light (Krajnak et al., 1997; Jiao et al., 1999).
Furthermore, light-unresponsive SCN cells exhibit circadian rhythms in
spontaneous firing rate in vivo in lab rats, but not in degus (Jiao et al., 1999).

The SCN is a functionally and anatomically heterogeneous nucleus, and

13

differences within some SCN neurons could lead to divergent patterns of
rhythmicity.

It is also possible that differences between diurnal and nocturnal patterns
of circadian rhythmicity emerge from processes downstream of the SCN. For
example, SCN targets may interpret output signals differently in one species
relative to another. This could occur in a decentralized fashion whereby re-
interpretation occurs at many sites receiving input from the SCN. Alternatively, a
single target could modulate signals coming from the SCN differently in diurnal
and nocturnal species. Different patterns of signals could then be broadcast from
this central location to multiple distal targets.

The hypothalamic subparaventricular zone (sPVZ) is one possible location
for a modulator of SCN efferent signals. Extending dorsally and caudally from
the dorsal border of the SCN to the paraventricular hypothalamic nucleus, the
sPVZ is one of the major targets of the SCN in lab rats (Watts and Swanson,
1987; Watts et al., 1987; Watts, 1991; Moore and Leak, 2001), mice
(Abrahamson and Moore, 2001) and hamsters (Morin et al., 1994). Cells in the
sPVZ project to many of the same regions as do cells in the SCN in lab rats
(Watts and Swanson, 1987; Watts et al., 1987; Watts, 1991) and in hamsters
(Morin et al., 1994; Orpen and Steiner, 1994; Kriegsfeld et al., 2004). Here we
focus on the ventral portion of the sPVZ, which extends outward from the dorsal
border of the SCN and overlaps partially with the perisuprachiasmatic region.
We refer to this region as the lower subparaventricular zone or LSPV. The daily

rhythm in expression of the immediate early gene product cFos in the LSPV is

14

dramatically different in nocturnal lab rats as compared to a diurnal rodent, the
unstriped Nile grass rat, Arvicanthis niloticus (referred to hereafter as the grass
rat; (Nunez et al., 1999). Specifically, in lab rats, the number of cFos-
immunoreactive (-ir) cells in the LSPV displayed a sharp peak at Zeitgeber time
(ZT) 1 and declined rapidly thereafter, whereas in grass rats, the rising phase of
the rhythm began at 2T 17, and cFos was still high at ZT 1, after which it began
to decline. These species differences raise the possibility that the LSPV Is a
central point at which a common efferent signal emanating from the SCN could
be modulated differently in one species relative to the other, leading ultimately to
differential patterns of coupling between SCN output and behavioral rhythms
(Nunez et al., 1999).

The present work sought to characterize further the grass rat LSPV and to
compare its rhythmic activity to that of lab rats. The first study had two primary
goals: to assess whether cFos in the grass rat LSPV remains high through the
night, and to determine when cFos begins to rise in the lab rat LSPV. In the
second study, we determined whether daily rhythms in this region are
endogenous by releasing grass rats and lab rats into constant darkness (DD) for
two to three weeks, and sacrificing them at different phases of their rhythms in
general activity. Because cells containing calbindin (CalB) had been previously
seen within and around the grass rat SCN (Mahoney et al., 2000), we processed
tissue from these animals for visualization of both cFos and CaIB-ir to assess the
degree to which cFos expression overlaps with that of CalB In the LSPV. Finally,

we mapped the distribution of cFos-positive cells in the grass rat LSPV in relation

15

to cells immunopositive for vasopressin (VP), vasoactive intestinal peptide (VIP)
and CalB to characterize the LSPV anatomically in relation to the SCN, and to

assess the utility of CaIB-ir as a peptidergic marker for the LSPV.

Experimental Procedures:

Animals:

Animals were adult male Sprague-Dawley lab rats obtained from Charles
River Laboratories (Wilmington, MA), and adult male grass rats obtained from a
breeding colony at Michigan State University (Katona and Smale, 1997).
Animals were singly housed in Plexiglas cages (lab rats: 26 x 48 x 20 cm; grass
rats 34 x 28 x 17 cm) with access to food (lab rats: Harlan Teklad 8640, Madison,
WI; grass rats, PMI Nutrition Prolab RMH 2000, Brentwood, MO) and water ad
Iibitum in a 12:12 light: dark (LD) cycle (lights on at 06:00). A red light (<5 lux)
remained on constantly for animal care purposes. All experiments were
performed in compliance with guidelines established by the Michigan State
University All-University Committee on Animals Use and Care, and the NIH
Guide for the Care and Use of Laboratory Animals. All efforts were made to

minimize the number of animals used and their discomfort.

General immunocytochemical procedure:

At the time of sacrifice, animals were given an overdose of sodium
pentobarbital and perfused transcardially with 0.01 M phosphate-buffered saline
(PBS), pH 7.2, followed by 4% paraformaldehyde (Sigma, St. Louis, MO) in 0.1 M
phosphate buffer, unless othenivise noted. Animals sacrificed during the dark

phase or in DD were fitted with a light-tight hood prior to perfusion to prevent

16

acute exposure to light. Brains were post-fixed for 1-2 hours, transferred to 20%
sucrose solution overnight, and then stored in cryoprotectant at -20°C until
sectioning. Brains were sectioned coronally at 30 pm on a freezing microtome,
and sections were either placed into PBS for immediate processing or into
cryoprotectant for further storage.

Unless noted othenivise, incubations were conducted at room temperature
for 1 hour on a shaker, and tissue was rinsed 3 x 10 minutes in PBS between
incubations. Free-floating sections were rinsed in PBS, and then incubated in (i)
5% normal serum in PBS with 0.3% Triton-X, followed by (ii) primary antibody on
a rotator for 48 hours at 4°C. Following primary incubation, tissue was incubated
in (iii) biotinylated secondary antibody, followed by avidin-biotin peroxidase
complex (ABC Vectastain Kit, Vector Laboratories, Burlingame, CA). Protein
was visualized by reacting with diaminobenzidine (DAB; 0.5mg/mL; Sigma) in
Trizma buffer (Sigma), catalyzed with hydrogen peroxide. For cFos/CaIB double-
labeling, cobalt chloride was added to the cFos reaction, yielding a blue-black
rather than brown reaction product. The reaction was quenched in PBS, then
rinsed and incubated with the second primary antibody. Reagents for each
reaction were as follows. cFos: (i) normal goat serum (NGS; Vector), (ii) rabbit
anti-cFos (Santa Cruz Biochemistry, Santa Cruz, CA, 1:25.000), (iii) biotinylated
goat anti-rabbit; (Vector, 1:200). VP: (i) NGS, (ii) guinea pig anti-VP (Peninsula
Laboratories, CA, 1:10,000), (iii) biotinylated goat anti-guinea pig (Vector, 1:200).
VIP: (i) 5% normal donkey serum (NDS; Jackson Laboratories, CA), (ii) guinea

pig anti-VIP (Peninsula, 122000), (iii) biotinylated donkey anti-guinea pig

17

(Jackson, 1:200). CaIB: (i) NDS, (ii) mouse anti-CalB (Sigma, 1:20,000), (iii)
biotinylated donkey anti-mouse (Jackson, 1:200). Specificity of CalB and F03
labeling was verified by blocking experiments in which the primary antibody was
pre-incubated with its antigen (CalB: Swant, Bellinzona, Switzerland, 3 ,ug/mL;
cFos: lmmunostar, CA, 5 pg/mL); in both cases pretreatment with blocking
peptide dramatically reduced or eliminated staining (data not shown). Specificity
of VP and VIP labeling in the grass rat SCN was previously documented in
(Smale and Boverhof, 1999). Primary deletion controls for each peptide also
resulted in absence of stain (data not shown). Tissue was mounted, dehydrated,

and coverslipped with Permount.

cFos rhythms in LD:

To evaluate cFos rhythms in the SCN and LSPV of entrained animals, 35
grass rats and 30 lab rats were kept in 12:12 LD and perfused at ZT 1, 5, 13, 17,
20, or 23 as described above except that 1.3% lysine and 4% sodium periodate
were added to the 4% paraformaldehyde. Every third section through the
rostrocaudal extent of the SCN was used for single-label cFos
immunocytochemistry.

Two sections through the mid- to caudal SCN were selected from each
animal for analysis. All cFos-positive neurons were counted within the SCN, and
within a region of the LSPV defined by a 215 um x 160 um rectangle placed
immediately dorsal to the SCN and lateral to the third ventricle (Fig. 2.1).

Bilateral counts for each area were averaged across the 2 sections.

18

M. 5;“ I.“ 3: 1:3,. 133:"
.g‘Dz33} fitP {.U‘ 0 :j‘\..
4." . 3 I

   
   

Figure 2.1: Photomicrograph of the SCN and LSPV of a grass rat
perfused at ZT 22 and stained for cFos-ir and Nissl. The SCN is
outlined on the left, and the sampling region used for LSPV
counts is drawn to scale and placed above the SCN. The section
depicted here corresponds to the map of cFos-ir in the mid-SCN
of the series depicted in Fig. 2.6. 3v: third ventricle; ocx: optic
chiasm. Scale bar = 200 um.

19

cFos Rhythms in DD:
To assess the rhythmicity of cFos-ir in the SCN and LSPV under constant

conditions, 20 grass rats and 23 lab rats were first housed in 12:12 LD with
infrared motion detectors placed directly over the cage to monitor general activity
rhythms. Activity data was collected and viewed on a PC running the Dataquest
3 program (Data Sciences International, St. Paul, MN). After an acclimation
period of 1-2 weeks, lights were turned off at the normal time (18:00) and
remained off from this point onwards.

Animals were allowed to free-run in DD for 21-22 days (grass rats) or 16-
17 days (lab rats). At the end of this time, actograms for each animal were
examined and activity onsets were eye-fitted independently by two investigators.
All animals were randomly assigned to specific circadian times (CT) for sacrifice,
where CT 0 = activity onset (grass rats), or CT 12 = activity onset (lab rats).
Grass rats were sacrificed at CT 5 (n=6), CT 13 (n=8), or CT 22 (n=6); lab rats
were sacrificed at CT 1 (n=5), CT 5 (n=6), CT 13 (n=5), or CT 22 (n=7). Every
third section through the SCN and LSPV was processed for double-label
cFos/CalB immunocytochemistry. Bilateral maps were made of 2 sections
through the mid- to caudal SCN and LSPV as described above using a camera
lucida, and counts were made of the numbers of cells positive for cFos only,

CalB only, and cells positive for both cFos and CalB.

2O

Statistical analyses:

Cell counts from each species were analyzed by single-factor analyses of
variance (ANOVA) using Statview (SAS Institute) followed by posthoc painivise
Tukey comparisons (LD experiment) or a Bonferroni correction (DD). Differences
for all tests were considered significant at the P<0.05 level. All counts were

made by an investigator unaware of the time of sacrifice of each animal.

Anatomical characterization of the LSPV region:

Two male grass rats kept in 12:12 LD were perfused at ZT 22 (when we
found cFos-ir in the LSPV to be maximal; see below). To identify the boundaries
between the SCN and the surrounding region in a more traditional manner, three
series through the SCN of one animal were processed for cFos-ir, VP-ir, and
VlP-ir, respectively. Maps were made of the VP, VIP and cFos cell distributions
in the SCN and LSPV through 5 sections extending from just rostral to the SCN
to just caudal to it. Following mapping, the coverslips were removed from the
cFos-ir series, tissue was Nissl-stained, and re-coverslipped in order to
determine the Nissl-defined boundary of the SCN. To evaluate the spatial
relationship between cFos and one potential marker for this region, we
processed one series from the second animal for cFos/CalB double-labeling.
Maps were made of the cFos and CalB cell distributions in the SCN and LSPV
through 5 sections extending from just rostral to the SCN to just caudal to it.
Following mapping, the coverslips were removed from the series, tissue was
Nissl-stained, and re-coverslipped in order to determine the Nissl-defined

boundary of the SCN.

21

 

 

250 ASCN i+ Lab rat‘s

— -O — Grass rats

 

200 ~

150 ~

100 .

50I

 

 

 

 

O

 

70 ~ B. LSPV
60 .
5O 4
4O 4 \
30 . .. \ /
\{x /
20 J , \ T
10 ~
0 . . 4 , . .
1 5 13 17 20 23
Time (ZT)
Figure 2.2: Numbers of cFos-positive cells in (A) the SCN, and (B)
the LSPV of grass rats (open circles) and lab rats (filled circles) kept
in a 12:12 LD cycle. Asterisks indicate time points where numbers

of cFos-positive cells are significantly elevated relative to other time
points in (*) grass rats, (**) lab rats.

# immunopositive cells

 

 

 

22

Resuns:

cFos rhythms in LD:
SCN: In both grass rats (F 5,29 = 44.696, p < .0001) and lab rats (F524 =

14.079, p < .0001), the numbers of cFos-positive cells in the SCN peaked early in
the day and decreased steadily thereafter. In grass rats, cFos was highest from
ZT 1-5 and lowest from ZT 17-23 (Fig. 2.2a, 2.3). In lab rats, cFos was also
highest at ZT 1 but declined to a trough by ZT 5, and remained low for the rest of
the day (Fig. 2.2a, 2.3). cFos-positive nuclei appeared throughout the SCN in
both species, with no clear regional differences emerging at any time point.
Thus, under a 12:12 LD cycle, both species displayed basically similar daily cFos
rhythms in the SCN, though in lab rats cFos-ir declined more rapidly from peak
values than in grass rats.

LSPV: In the grass rat LSPV, there was a significant effect of time (F529 =
19.548, p < 0.001) on the numbers of cFos-positive cells in the SCN. Posthoc
comparisons revealed that the number of cFos-positive cells at ZT 5 and ZT 13
were significantly lower than at all other time points (Fig. 2.2b, 2.3). The number
of cFos-positive nuclei rose sharply from 2T 13 to ZT 17 and remained high
through ZT 1. In lab rats, cFos was also rhythmic (F524 = 5.416, p = .0018), but
in contrast to the pattern seen in grass rats, the number of cFos-positive cells did
not rise until the interval between ZT 23 and ZT 1, and had declined back to
baseline levels by ZT 5 (Fig. 2.2b, 2.3). Low levels were maintained for the

remainder of the day and night. Thus, both species displayed

23

 

Figure 2.3: Photomicrograph of the SCN and LSPV of representative
grass rats (top row) and lab rats (bottom row) perfused at ZT 1 (left), ZT 13
(middle) or ZT 23 (right) and processed for cFos-ir. 3v: third ventricle; ocx:
optic chiasm. Scale bar = 100 um.

24

daily rhythms in cFos-ir in the LSPV under LD, but in grass rats cFos rose ~8

hours earlier and remained elevated ~8 hours longer.

cFos and CalB rhythms in DD:

SCN: In the grass rat SCN, cFos remained rhythmic after animals had
been in DD for 3 weeks (F217: 9.701, p = .0015; Fig. 2.4a). Posthoc
comparisons indicated that the number of single-labeled cFos-positive nuclei was
higher at CT 5 than at either CT 13 or CT 22, a pattern similar to that seen in LD.
The cFos rhythm in the SCN was only found in cells that did not contain CalB; no
effect of time was seen on double-labeled cells (F217 = 0.689, p = .5164). The
number of single-labeled CalB-positive cells also did not vary as a function of
time (F217 = 0.070, p =.9327). In grass rats cFos-ir appeared throughout the
SCN, with no clear regional differences. At all time points, a small number of
CaIB-positive cells were scattered throughout the SCN in rostral and middle
sections but in the caudal SCN these cells were concentrated in a roughly
circular region slightly lateral to the center of the nucleus. Double-labeled cells
were rare, and were not restricted to any subregion of the SCN.

In lab rats there were no effects of time on the number of single-labeled
cFos-positive nuclei (Fang = 2.070, p = .1382), single-labeled CaIB-positive cells
(F349 = 0.029, p = .9931 ), or in the number of double-labeled cells (F3,19 = 2.425,
p 2.0973) in the SCN after 2 weeks in DD (Fig. 2.4b). cFos-positive nuclei
tended to be concentrated in the dorsomedial SCN, whereas CalB-positive cells

were fewer in number and scattered throughout the SCN.

25

120

 

A. SCN: Grass Rats
100 ~

80~ i

60~

A
O

.T__
M I T
l

i l
LJJ!" l i ‘_ i

5 13

 

N
O

 

 

O

 

 

120

 

 

B. SCN: Lab Rats [m
l E gate...

100 -

# of immunopositive cells

80 -
j i-
401 l W

209

 

 

 

 

 

 

O
_. IA
H
|_
:m_-*__
H
(F?

Time (CT)

Figure 2.4: Counts of cells positive for cFos only (grey bars),
CaIB only (white bars), and cells positive for both cFos and
CalB (black bars) in the SCN of A) grass rats, B) lab rats, after
2-3 weeks in DD. Asterisks indicate time points where
numbers of cFos-positive cells (*) are significantly elevated
relative to other time points, p=.0015.

26

LSPV: In the grass rat LSPV, there were significant effects of time on the
number of cells singly labeled for cFos (F217 = 27.801, p < .0001; Fig. 2.5a), such
that cFos-ir was higher at CT 22 than at either CT 5 or CT 13. This pattern was
also observed in cells that contained both cFos and CalB (F217 = 19.068, p <
.0001; Fig. 2.5a) Thus, cFos increased at night both inside and outside of CalB-
positive cells. The total number of CaIB-positive cells was also affected by time
(F217 = 5.795, p = .0201; data not shown), and was higher at CT 22 than at CT 5.
However, there was no effect of time on numbers of CaIB-positive cells that did
not contain cFos (F217 = 1.460, p = .2600; Fig. 2.5a). Thus, CalB increased at
night, but only in cells also positive for cFos.

In the lab rat LSPV, there were no effects of time on the number of single-
Iabeled cFos-positive nuclei (F349 = 0.058, p =.9811), single-labeled CalB-
positive cells (an9 = 0.516, p = .6759), or the number of double-labeled cells

(F339: 0.087, p = .9663, Fig. 2H5b)

Distributions of cFos, VP, VIP and CalB in the SCN/LSPV:

SCN: In order to characterize in more detail the distribution of cFos-ir in
the LSPV in relation to the SCN, we compared labeling for VP and VIP, two
peptides commonly used to demarcate the SCN, with labeling for cFos at ZT 22,
when cFos-ir is highest in the LSPV. The distributions of VP and VIP within the
SCN of a grass rat perfused at ZT 22 were similar to those described previously
(Smale and Boverhof, 1999) (Fig. 2.6). VP-positive neurons were distributed in a
roughly crescent-like pattern within the lateral, dorsal and medial SCN. The

dorsal

27

# of immunopositive cells

 

 

 

 

 

 

 

 

 

 

 

70 “ A. LSPV: Grass Rats
60 '
50 - *
at... ,_T
*1 Ill
1 iil‘ii
i i iIWnI * *
y‘lliy‘
T i “ lnljwiiiii
WIT": [:“IIIIW:
it.” (3:333
iiil‘l“ ' Lilli-W '
13 22
70 B. LSPV: Lab Rats page;
60 — cFos/CalBj
50
4O '
l . . .
30 ’ . 7 F"
20 i I l .
j a l , , T , ,i l
10 j . . 2‘ . a
l l , .
O ..j__,,i L] i 1 Lil.
1 5 13 22
Time (CT)

Figure 2.5: Counts of cells positive for cFos only (grey bars),
CalB only (white bars), and cells positive for both cFos and
CalB (black bars) in the LSPV of A) grass rats, B) lab rats, after
2-3 weeks in DD. Asterisks indicate time points where
numbers of cFos-positive cells (*) or double-labeled cells (**)
are significantly elevated relative to other time points, p<.0001.

28

boundary of the region containing VP-positive cells coincided almost perfectly
with the dorsal boundary of the SCN as defined by Nissl in an alternate series
(one section is shown in Fig. 2.1 ). VIP-positive neurons were concentrated in the
ventral SCN, extending to the optic chiasm. The lateral edge of the VIP-positive
cell-containing region corresponded to the lateral border of the SCN as defined
by Nissl. Distributions of VP-ir and VlP-ir overlapped slightly along the medial
and lateral edges of the SCN, but not along its dorsal or ventral regions. Neither
VP- nor VIP-positive cells were found outside the boundaries of the SCN as
defined by Nissl. VP-positive fibers extended dorsally from the SCN, and
coursed dorsally and caudally to the PVN; VIP-positive fibers were present in
approximately the same regions, but to a somewhat lesser extent. At ZT 22 few
cFos-positive cells were observed within the SCN, whereas large numbers were
seen dorsal and dorsolateral to the nucleus (Fig. 2.1, 2.6). Similarly, few cFos-
positive cells were found either rostral to the SCN or caudal to it.

In tissue processed for cFos/CalB-ir from a grass rat perfused at ZT 22, a
few cFos-positive nuclei were observed in the SCN as indicated by a blue-black
reaction product (Fig. 2.7). CalB-positive cells were light brown, and both
perikarya and fibers were visible within the SCN. CalB-positive cells were
scattered evenly through the SCN from its rostral to its middle regions (Fig. 2.8).
However, in the caudal SCN they were concentrated in a round subnucleus near
the center of the SCN (Fig. 2.7b), as seen in the previous experiment. The
distribution of CalB within the caudal SCN appears to be similar to that of gastrin-

releasing peptide (GRP)-containing cells seen previously in the grass rat SCN

29

 

 

      

Fos VIP
Pre <
SCN
. 1". . 2
W; _ ocx W
' . ‘2 2
Rostral '
SCN '5'".;:.‘.~'°.- .

   

 

 

 

 

 

 

 

OCX

Figure 2.6: Distributions of AVP, VIP and cFos-ir through sections rostral
to the SCN, in the SCN, and caudal to the SCN of a grass rat perfused at
ZT 22. 3v: third ventricle; ocx: optic chiasm. Scale bar = 100 um.

30

(Katona et al., 1998; Smale and Boverhof, 1999). The CalB-positive region of
the caudal SCN was clearly separated from the LSPV by a more CalB-sparse
zone around it (Fig. 2.7b).

LSPV: In a grass rat perfused at ZT 22, many cFos-positive nuclei were
observed just dorsal and dorsolateral to the SCN (Fig. 2.6, 2.7a, 2.8). In the
rostral-most sections, cFos-positive cells were located in the area immediately
dorsal to the SCN. In more caudal sections, cFos-positive cells increased in
density and spread out both dorsally and to a lesser extent laterally, such that in
mid- and caudal sections cFos-positive cells were present from the
periventricular hypothalamic nucleus (Pe) to the anterior hypothalamic area
(AHA). cFos labeling was restricted to the area dorsal and dorsolateral to the
SCN; few cFos-positive cells were visible directly lateral to the SCN at any
rostrocaudal level (Fig. 2.6, 2.7a). The region containing the highest
concentration of cFos-positive cells at 2T 22 was clearly outside the SCN as
defined by Nissl stain, and did not overlap with either VP- or VIP-positive cell
bodies (Fig. 2.1, 2.6).

In tissue processed for cFos/CalB-ir, the distribution of cFos-positive cells
was very similar to that described above (Fig. 2.7a, 2.8). A large number of
CalB-positive cells were interspersed with cFos-positive cells at all rostrocaudal
levels of the LSPV. These CaIB-positive cells extended dorsally and laterally
from the outer boundaries of the SCN in a pattern similar to that of cFos-positive
cells (Fig. 2.8). However, there were more CalB-positive than cFos-positive cells

lateral to the SCN in caudal sections. CaIB-positive cells in the LSPV were not

31

 

‘ . I J I}
., ‘g l
A ~ ° - ~
0 . ‘I 1 .5 ' ‘ ‘g
‘ ‘ ' 10 ‘ ..Q h
. e. 3v~.'- . ' -
i‘ ' ' . \ O ‘ O
. . g‘ - .0 H ' .3 o d c
‘ , 0.3.. '1 «A I s' u n”; '
' ‘ . I ,‘ | '2'": O \r.‘:' e t..-
. e ”A... ‘. c .1, 'g. 'l. I. .2. ' ' ’
0“ . ...‘ntb’t .‘;. é .' .6-3
s \ _v . ”‘0. .0 . 5" o ‘I 9..» p. ,‘-‘,§ .-
'.'.-"-*"‘.' "w‘ a -
.. '1: “I$ "'§‘ ' .. . '. 2' ..I.I ’
,0 . (6 ' J‘l'."v \' I. z '3‘; ‘ 5'. ' ’
.. D" .13.. ‘ ‘ § .3 so ’I". M. (35‘: - .. . u...
.. 'I “ ' n0.- -' I... .
1 :. -. . ‘;'.'a‘- "‘5 ‘ . “"3“; '. ‘91 ~. ' ‘
s“ ‘ ' . . .I . '.- . : .- ' O .
b..- . -_.,;r,:- - .. . 1
K . 3 f! . . ‘ a". .-
I .\ ‘e . 34‘- f}! v:_ ‘ - Q :
‘-. -., 77"; :-
. ' Vi
- , 5" ,
/ ‘ .ch:'\
I ' . ". ' 0
-fl “ “ ‘J ~
3' .. 3 - 0'
r. ,w’ V‘C. /‘
'I..‘ ‘ ll‘nf. 1...; * ‘
.‘ ‘6‘. .1" ,‘ '
«~- i Q - C
' '3’. . of “0‘
o . ‘ .+.| -.
o ‘ J ‘
Q. ..~‘:‘,.:.‘:
5 a 32‘ .
.7 "' 'L‘l- '1 - c . .
:g.’ :52“, 'T *.
9: ' '12.". (t \
3" e “7' . v- n
I‘ 4‘ “a
\‘t. '- .1 V ‘2 / _
—‘ \s . ._ %

Fig. 2.7. (A) Photomicrograph of the SCN and LSPV of a grass rat perfused
at ZT 22 showing the distributions of cFos and CalB-ir; scale bar?= 200 ?
um. (B) Detail of the caudal SCN in the same animal. The section pictured
here is the one from which the map of cFos and CalB-ir in the caudal SCN
are drawn in Fig. 2.8. Scale bar?= 200 ?um. (C) Double-labeled cells in the
LSPV of the same animal shown in (A) and (B) taken at 100x, showing
cFos-ir (arrowheads), CaIB-ir (arrows), and double-labeled cells
(asterisks); scale bar?= 20 um. ocx, optic chiasm; 3v, third ventricle.

32

densely packed in a nucleus, but were clearly more numerous in the LSPV than
in the dorsal SCN, which contained little to no CalB-ir. While significant numbers
of CalB-positive cells were observed in a subnucleus near the center of the
caudal SCN (see above), this cell group was separated from the LSPV by a
CalB-sparse zone around it. Double-labeled cells were easily identifiable at high

power (Fig. 2.7c).

Discussion:

These experiments serve to expand on and clarify the species differences
in cFos-ir rhythms in the LSPV reported earlier (Nunez et al., 1999). Our results
establish that the daily rhythm in LSPV cFos-ir differs in grass rats and lab rats
with respect to both its phase and its waveform, that both the SCN and LSPV
cFos rhythms are endogenous in the diurnal grass rat, but not in the nocturnal
lab rat, and that in grass rats the LSPV cFos rhythm is paralleled by a rhythm in

CaIB-ir within cFos-positive cells.

Neuroanatomy of the grass rat LSPV:

At a time of peak cFos-ir in the LSPV (as demonstrated in experiments 1
and 2), the distribution of cFos-ir did not overlap with the distributions of two
peptides (VP and VIP) that clearly mark the boundaries of the SCN. Moreover,
the distribution of cFos-Ir was unambiguously outside the Nissl-defined border of
the SCN. Thus, the cFos-positive LSPV region of grass rats that we sampled
represents an area outside the conventional boundaries of the SCN. CalB-
positive cells were numerous outside of the SCN, where their distribution

overlapped with that of cFos-positive cells surrounding the SCN. While it is

33

Fo's ' CaI.B'.I_,-.'5:33.351.

Pre
SCN

 

 

 

Rostral . , ..z ‘_-=..' j . :: : -"
SCN '. __ ~.‘":":'_'.':-.=.'~ . .' ' .‘5: =.-

oc_x " ' ocx

Mid-
SCN

 

Caudal
SCN

 

Post
SCN

   

\"O '1 1' ' " 105s. ' .
ocx ' ocx' '
Figure 2.8: Distributions of cFos— and CalB-ir in the SCN and LSPV of

a grass rat perfused at ZT 22. 3v: third ventricle; ocx: optic chiasm.
Scale bar = 100 ?um.

34

difficult to assign a discrete outer boundary of the LSPV based on these data, it
may be thought of as the edge of a region dorsal and dorsolateral to the SCN
that contains both CalB-positive cells and cells that express cFos in the late
night. This region overlaps well with the distribution of VP and VlP-ir efferent
fibers from the SCN that characterize the sPVZ in lab rats (Watts and Swanson,
1987; Watts et al., 1987; Watts, 1991) and grass rats (Smale and Boverhof,
1999). However, it is distinct from the perisuprachiasmatic region in that the
overlap of cFos- and CalB-ir is markedly higher in the area dorsal to the SCN,
than in the areas lateral, rostral or caudal to the SCN. The inner boundary of the
LSPV is easier to identify than its outer boundary because it is delineated by the
borders of the SCN, as defined by the VP cells of the nucleus or by Nissl stain.
To summarize, the mapping data presented here clearly show that at ZT 22,
cFos-ir in the region that we call the LSPV is outside the SCN. This cFos-ir
occupies a region roughly dorsal to the SCN that corresponds to the ventral-most
aspect of the sPVZ as defined in the lab rat by Watts and Swanson (1987) and

Watts et al. (1987).

cFos rhythms in the SCN:

The daily cFos rhythm in the SCN was fundamentally similar in grass rats
and lab rats kept in a 12:12 LD cycle, with a peak early in the light phase, at ZT
1, followed by a decline to levels that remained low through the dark phase.
However, in grass rats cFos did not decline until after ZT 5, whereas in lab rats it
had already dropped to basal levels by that time. Although previous studies of

cFos in the SCN of grass rats (Katona et al., 1998; Nunez et al., 1999) and lab

35

rats (Nunez et al., 1999) have found varying results at ZT 5, all reveal elevated
cFos expression in the SCN of both species in the hour after light onset, followed
by decreases from the light to the dark phase of the cycle. A rapid decline in
cFos expression circa ZT 5 could be responsible for the differences between
experiments.

After 2-3 weeks in DD, grass rats, but not lab rats, displayed a robust
endogenous cFos rhythm in the SCN very similar to that seen in LD. In grass
rats, this cFos-ir was not restricted to regions of the SCN defined by a particular
neuronal phenotype (i.e. CalB, VP, VIP). The lack of cFos/CaIB double-labeling
in the grass rat SCN indicates that the cFos rhythm was specifically outside of
the CalB-positive subpopulation of cells. The high levels of cFos expression
during the subjective day (CT 5) seen here in the SCN of grass rats was not
detected in a previous study in which grass rats were sampled at CT 6 after only
24 hours in DD (Mahoney et al., 2001). The difference between the two studies
could reflect the emergence of a rhythm following extended periods in DD.
Alternatively, cFos may have been low at CT 6 in the study of Mahoney et al
(2001), and high at CT 5 in the current study, because cFos declines rapidly
around this time (see above).

In lab rats, we observed no effect of time on cFos-ir in the SCN after 2
weeks in DD. In contrast, several reports have described an endogenous rhythm
in cfos mRNA (Guido et al., 1999a; Schwartz et al., 2000) and protein (Sumova
et al., 1998; Guido et al., 1999a; Schwartz et al., 2000; Beaule et al., 2001) in the

dorsomedial SCN of lab rats as well as hamsters (Guido et al., 1999b).

36

Differences in the time points chosen for sampling (between 4-24 time points in a
24-hour period), and/or the duration of exposure to DD (2 days or less in earlier
studies, versus 2 weeks in the present study) may account for the divergent
findings. To our knowledge, this is the first report examining spontaneous cFos
expression in the SCN of either lab rats or a diurnal species after more than 5
days in DD. More work is needed to determine the functional implications of this

species difference in SCN cFos expression.

cFos rhythms in the LSPV:

In 12:12 LD, there were clear species differences in the daily cFos rhythm
in the LSPV. In grass rats the rising phase of the rhythm in the LSPV was out of
phase with that of the SCN rhythm (in grass rats), and with rhythms in both
regions in lab rats. In an earlier study, grass rats kept in LD exhibited increased
cFos-ir in the LSPV at ZT 16 relative to ZT 4 (Smale et al., 2001), but in another
this trend did not reach statistical significance (Mahoney et al., 2000). Thus, the
rising phase of Fos in the LSPV may occur rapidly between ZT 16-17 in the grass
rat. The cFos rhythm in the LSPV differed not only with respect to the phase of
the rise but also its waveform, in that cFos remained elevated considerably
longer in grass rats than in lab rats. The divergent patterns of cFos-ir in the
LSPV of these species may therefore reflect two distinct processes, one
promoting an increase in cFos expression and another promoting its
maintenance. The cFos rhythm in the grass rat LSPV also differs from that of lab
rats in that it is clearly endogenous, as it persisted for 3 weeks in DD without a

reduction in amplitude. These species differences in both the phase and the

37

dependence on light of the cFos rhythm suggest a functional role for the grass rat
LSPV that is distinct from that of the lab rat.

In addition to the rhythm In single-labeled cFos-positive cells, a parallel
rhythm in numbers of CalB-positive cells containing cFos was evident in the
LSPV of grass rats kept in DD. Interestingly, a sizable proportion of the total
CalB cells in the LSPV did not contain Fos, and the numbers of CalB cells
without cFos did not change over time. Thus, there seem to be two distinct
populations of CaIB-positive cells in the LSPV: one that does not change with
respect to either CalB or cFos expression, and another in which both CalB and
cFos expression increase from CT 5 to CT 22.

The functional significance of CalB in either the SCN or the LSPV of the
grass rat is unclear at present. In hamsters, a CaIB-positive SCN subnucleus is
essential for maintenance of circadian locomotor rhythms (LeSauter and Silver,
1999). In that species, cells in this region are both responsive to light (Silver et
al., 1996b) and extensively interconnected with cells in other subregions in the
SCN (LeSauter et al., 2002; Jobst et al., 2004). Although the CaIB cells
themselves do not exhibit a spontaneous daily rhythm in firing rate (Jobst and
Allen, 2002) or cFos expression (Silver et al., 1996b), the distribution of CalB in
the cell nucleus relative to the cytoplasm is rhythmic (Hamada et al., 2003).
Furthermore, treatment with antisense to CalB causes time-dependent changes
in light-dependent per1 induction in the SCN and light-induced behavioral phase
shifts (Hamada et al., 2003). Thus, in at least one nocturnal species, CalB-

positive cells in the SCN contribute in a crucial way to the regulation of circadian

38

rhythms in the SCN and in behavior. Whether CalB-positive cells inside or above
the SCN of the grass rat are necessary for circadian rhythms, or similarly
organized in terms of function, remains to be determined.

In lab rats, cFos-ir in the LSPV was not rhythmic after 2 weeks in DD,
revealing a second species difference in the pattern of cFos expression in this
region. Spontaneous rhythms in multiple-unit activity in the region surrounding
the SCN (including but not limited to the LSPV) persist in blinded rats kept in DD
(lnouye and Kawamura, 1979). A lack of rhythmicity in cFos-ir, then, may not
reflect a total loss of rhythmic neural activity in the region. For example, in lab
rats, MUA rhythms in the area around the SCN could be driven by the SCN and
thus persist in the absence of photic input, whereas cFos rhythms may be
dependent on light. In grass rats, however, cFos rhythms in the LSPV are
endogenous. The species differences in both the phase of the LSPV cFos
rhythm and its dependence on light suggest that the function of the cFos rhythm,
and possibly the LSPV itself, is very different in these two species.

The only other data available comparing daily rhythms within and adjacent
to the SCN of diurnal animals comes from MUA records. In the diurnal Siberian
chipmunk, MUA outside the SCN peaked during the day and was low at night
(Sato and Kawamura, 1984b), whereas in lab rats, MUA outside the SCN peaked
at night and was low during the day (lnouye and Kawamura, 1979). Thus, daily
rhythms in MUA outside the SCN were reversed in a diurnal species relative to a
nocturnal one, and in the nocturnal species the rhythms outside the SCN were

reversed relative to those in the SCN. In the present study of cFos, rhythms in

39

the LSPV also differed in a diurnal species relative to a nocturnal one, but in this
case it was the diurnal species in which these rhythms were out of phase with
those in the SCN. Thus, MUA rhythms in diurnal Siberian chipmunks and cFos
rhythms in diurnal grass rats were not in phase with each other. This may reflect
a basic species difference in the region, or perhaps cFos-ir does not reflect the
same type of activity as MUA. In either case, these species differences in the
area around the SCN, contrasted with the similarities seen within the SCN, are
consistent with the hypothesis that differences between nocturnal and diurnal
species may arise from direct targets of the SCN, or from SCN cells that project

to these targets.

Conclusions:

The precise function of either the LSPV, or the sPVZ, in either a diurnal or
nocturnal species has not been firmly established, but seems likely to be closely
tied in with the circadian system. In lab rats, anterograde and retrograde tract
tracing revealed that the sPVZ projects to almost all of the same targets as the
SCN, but to a greater degree (Watts and Swanson, 1987; Watts et al., 1987),
leading the authors to suggest that the sPVZ might modulate SCN output signals
(Watts and Swanson, 1987; Watts, 1991 ). Several lines of evidence suggest that
the sPVZ may be involved in the timing of estrus-related events in lab rats
(Nunez and Stephan, 1977; Arendash and Gallo, 1979; Docke et al., 1982; Watts
et al., 1989; van der Beek et al., 1997b; Fernandez-Galaz et al., 1999). The
sPVZ may also be involved in the regulation of circadian locomotor rhythms in

lab rats (Lu et al., 2001; Moore and Danchenko, 2002), although if this is the

40

case, fibers projecting from the SCN to the sPVZ are probably not necessary for
this (Brown and Nunez, 1986; Watts et al., 1989). It should be noted furthermore
that these studies considered the sPVZ as a whole, and that similar
manipulations targeting the LSPV more specifically could yield different results.
Although many questions remain, it appears that the sPVZ in lab rats is involved
with a subset of the SCN’s output, and may in addition help integrate this
information with other factors such as hormonal state to organize timing of
estrus-related neuroendocrine events and behaviors. If so, then a near-reversal
in the rhythmic activity in the region, as indicated by cFos-ir, could be part of a
mechanism mediating differences in the timing of expressed circadian rhythms in
nocturnal and diurnal species. In lab rats, in which a nocturnal pattern of
behavior is expressed, rhythmic cFos expression in the LSPV is either in phase
with the SCN (LD) or nonexistent (DD). In grass rats, however, the 8-hour
difference in the rising phase of the daily cFos rhythms in the LSPV as compared
to the SCN could reflect a mechanism promoting a diurnal pattern of behavioral
and physiological rhythms. For example, neurons in the grass rat LSPV could
receive rhythmic signals from the SCN, transform those signals, and then
transmit them to downstream targets. More work is needed to determine if and
how the LSPV may modulate the outputs of the SCN in diurnal and nocturnal
species.

In conclusion, the LSPV of the grass rat, a region that receives substantial
input from the SCN, displays a daily rhythm in cFos expression that differs from

that of lab rats with respect to 1) its rising phase, 2) the duration of the peak, 3)

41

its phase relationship relative to cFos rhythms in the SCN, and 4) its dependence
on a LD cycle. Furthermore, an endogenous rhythm in CalB expression within
cFos-positive cells parallels that of cFos in the LSPV of grass rats, but not lab
rats. Cellular processes mediated by cFos and CalB, such as gene expression
and calcium regulation, are likely to be timed very differently in the LSPV of these
two species. These data may reflect a species difference in the mechanisms
whereby efferent signals from the SCN are coupled to their downstream targets.
It is tempting to speculate that differences in this coupling could lead to the
dramatic differences in daily rhythms in physiology and behavior seen between

diurnal and nocturnal rodents.

42

CHAPTER 3
Efferent projections of the suprachiasmatic nucleus and lower
subparaventricular zone in the Nile grass rat (Arvicanthis

niloticus)

Introduction

Diurnal and nocturnal species display marked differences in a host of
behavioral and physiological rhythms. The hypothalamic suprachiasmatic
nucleus (SCN) houses the central circadian pacemaker in mammals (Weaver,
1998), but the phase of rhythms that have been examined in the SCN appears
similar in diurnal and nocturnal species. This is the case for daily rhythms of
multiple—unit activity (Kubota et al., 1981; Sato and Kawamura, 1984b; Kurumiya
and Kawamura, 1988), 2-deoxyglucose uptake (Schwartz, 1991) and clock gene
expression (Hastings et al., 1999; Mrosovsky et al., 2001; Yan and Okamura,
2002; Caldelas et al., 2003; Ramanathan et al., 2006), as well as for rhythms in
mRNA for the clock-controlled genes prokinetlcin 2 (Cheng et al., 2002; Lambert
et al., 2005) and vasopressin (Dardente et al., 2004). While differences between
day- and night-active species could result from unidentified processes or cells
within the SCN, they are more likely to arise from processes occurring
downstream of the SCN.

One region that could play an important role in such a process is the
hypothalamic subparaventricular zone (sPVZ). The sPVZ extends dorsally and

caudally from the SCN to the ventral edge of the hypothalamic paraventricular

43

nucleus (Watts, 1991; Saper et al., 2005), and is a major target of the SCN in the
nocturnal laboratory rat (Watts and Swanson, 1987; Watts et al., 1987; Moore
and Leak, 2001), mouse (Abrahamson and Moore, 2001) and hamster (Morin et
al., 1994; Orpen and Steiner, 1994; Kriegsfeld et al.). The sPVZ in turn projects
to many of the same efferent targets as the SCN in lab rats (Watts and Swanson,
1987; Watts et al., 1987) and hamsters (Morin et al., 1994; Orpen and Steiner,
1994). Daily rhythms in cFos expression in the lower part of the
subparaventricular zone (LSPV) are dramatically different in diurnal Nile grass
rats and nocturnal lab rats (Nunez et al., 1999; Schwartz et al., 2004), and daily
rhythms in multiple-unit activity adjacent to the SCN are reversed in diurnal
Siberian chipmunks compared to nocturnal lab rats (lnouye and Kawamura,
1979; Sato and Kawamura, 1984b). The grass rat LSPV also displays daily
rhythms in immunoreactivity for calbindin and for the clock gene products Perl
and Per2 (Schwartz et al., 2004; Ramanathan et al., 2006). Thus, this region
may have functional characteristics that would enable it to modulate circadian
timing signals originating in the SCN, and to do that differently in diurnal and
nocturnal animals (Watts, 1991; Schwartz et al., 2004; Saper et al., 2005). If this
is the case, then signals from the LSPV should interact with those from the SCN
either at, or downstream of, SCN targets. The projections of the SCN and LSPV,
however, have not yet been described in a diurnal species.

In the present study, we have mapped the efferent projections of the grass
rat SCN and LSPV using the anterograde tract tracer biotinylated dextran amine

(BDA), and compared the distributions of fibers from these two areas. We report

44

that the efferent projections of the SCN and LSPV are quite similar to those
reported for nocturnal species, with projections extending through the
diencephalon, and the densest innervation in a relatively small subset of midline
hypothalamic and thalamic nuclei. Taken together with reports from nocturnal
Species, these data suggest that the efferent neuroanatomy of the SCN and
LSPV is highly conserved in rodents, and that the LSPV has projections that may

enable it to modulate the responsiveness of target cells to signals coming from

the SCN.

Experimental Procedures

Animals:

Adult female grass rats obtained from a breeding colony at Michigan State
University (Katona and Smale, 1997) were singly housed in Plexiglas cages (34 x
28 x 17 cm) with access to food (PMI Nutrition Prolab RMH 2000, Brentwood,
MO) and water ad Iibitum. Animals were kept in a 12:12 light: dark cycle; a red
light (<5 qu) remained on constantly for animal care purposes. All experiments
were performed in compliance with guidelines established by the Michigan State

University All-University Committee on Animals Use and Care, and the NIH

Guide for the Care and Use of Laboratory Animals.

Surgical procedures and tissue processing:

All animals received a unilateral iontophoretic injection of 10,000 MW BDA
(Molecular Probes; Eugene, OR) directed at the SCN or LSPV. Animals were
deeply anesthetized with sodium pentobarbital (50mg/kg i.p.) prior to surgery,

and then the head was shaved, wiped with betadine, and injected with lidocaine

45

(0.03cc s.c.). The animal was then placed into a stereotaxic apparatus (Stoelting
00.; Wood Dale, IL) with the incisor bar set to -6.0 mm and the micropipette set
at an angle of 10 degrees, and a 1-2 cm incision was made in the scalp to
expose the skull. Final injection coordinates were: 1.4 mm anterior to bregma,
1 .2 mm lateral to bregma, and 6.3 mm ventral to dura. A small hole was drilled in
the skull, and dura was gently broken to allow uninterrupted passage of the
micropipette. A glass micropipette (WPI; Sarasota, FL) with inner tip diameter of
1 0-15 pm was then filled with 10% BDA in 0.01 M phosphate-buffered saline
(PBS; pH 7.4) and lowered slowly to injection depth. Injections were made under
5 mA positive alternating current at a rate of 7 seconds on/ 7 seconds off, for 7
minutes, followed by a 7-minute rest period. Micropipettes were then withdrawn
slowly under negative current. The incision was closed with autoclips, and
animals were given 0.06 cc buprenorphine i.m. and 1.0 cc sterile saline s.c.
Seven days later, animals were given an overdose of sodium pentobarbital and
perfused transcardially with 0.01M phosphate-buffered saline (PBS), pH 7.2,
followed by 4% paraformaldehyde (Sigma, St. Louis, MO) in 0.1M phosphate
buffer. Brains were post-fixed for 4 hours, then transferred to 20% sucrose
solution overnight. Brains were sectioned coronally at 30 pm on a freezing
microtome into 3 series of sections, which were then stored in cryoprotectant at -

20°C until processing began.

One series of free-floating sections were rinsed 3 x 10 minutes in PBS at
room temperature, and then incubated in avidin-biotin peroxidase complex (ABC

Vectastain Kit, Vector Laboratories, Burlingame, CA). Protein was visualized by

46

reacting with diaminobenzidine (0.5mg/mL; Sigma) in Trizma buffer (Sigma)
enhanced with nickel sulfate. Tissue was mounted, dehydrated, and
coverslipped with Permount. An alternate series from each animal with a
successful injection was reacted as above, mounted and counterstained with
neutral red to assist in mapping brain regions. Injections were considered
successful if they were isolated to either the SCN or the LSPV, with no spread to
the other area. Data presented come from 2 injections localized to the SCN and

4 injections localized to the LSPV (Fig. 3.1).

Microscopy:

Sections were analyzed on a Zeiss Axioskop microscope (Zeiss,
Germany) under brightfield illumination, and all photomicrographs were taken
using a Zeiss Axiocam MR (Zeiss) and processed in Adobe Photoshop 7.0
(Adobe Systems; Mountain View, CA). Labeled fibers were mapped from two
representative cases (SCN: # 74, LSPV: # 34), using selected sections through
the hypothalamus. For each single-labeled (i.e. non-counterstained) section, a
series of photomicrographs was taken at 10x and combined in Photoshop to form
a composite photomicrograph of the section. Images were adjusted for
sharpness, brightness and contrast. BDA-ir fibers were traced in Photoshop
using a drawing pad (Wacom; Vancouver, WA). Although BDA can be carried in
a retrograde direction (Fieiner et al., 2000), we saw very few BDA-ir cell bodies
that were not associated with the injection sites in each of the cases analyzed.
All anatomical landmarks and nuclear boundaries were drawn based on

comparison between neutral-red counterstained and single-labeled sections in

47

 

 

 

 

 

 

 

 

 

 

Figure 3.1: Line drawings depicting the location of biotinylated dextran
amine (BDA) injections in all cases described in the present study. See
page xv-xvi for abbreviations.

48

the representative cases. The map of the injection sites was prepared in
Photoshop using # 74 as a template, with outlines of injection sites for other

cases traced in.

Results

Injection sites:

Both of the injections centered in the SCN filled the entire SCN (Fig. 3.1 ).
However, injection spread in # 74 was limited to the boundaries of the SCN,
whereas # 81 included a small amount of spread into the retrochiasmatic area.
The distribution of BDA-labeled fibers in these two cases was overall very similar,
with slightly increased fiber density in # 81. Of the 4 injections centered in the
LSPV (Fig. 3.1), cases 34 and 61 were the most similar with respect to
placement, extending up to 400 pm from the dorsal and dorsolateral edges of the
mid- to caudal SCN. They were also similar with respect to the distribution of
labeled fibers. Case 40 was centered slightly more rostrally than # 34 and # 61,
but overlapped with both of these injections at their centers. Case 75 was
centered farther away dorsally from the SCN than the other 3 cases, and was
similar to # 40 in its rostrocaudal placement. Fiber distributions of these two
cases were very similar to each other, and while generally similar to those of # 34
and # 61, exhibited some important differences. Of these 4, # 34 was the
smallest injection and # 75 the largest. None of the 4 injections centered in the
LSPV labeled cells in the SCN as revealed by neutral red counterstaining in

alternate sections. Cases 74 and 34 will be described in detail as representative

49

 

   

" .. f2, /
r: ..J .:-r ”i-A—fréNST
I a], ”'17, \Q

l‘ .1, ,- ..
v -.

i ‘ 111W“
Ni ‘

 

 

 

 

 

 

 

 

 

 

 

Figure 3.2: Line drawings depicting the distribution of BDA- labeled
fibers following an injection centered in the suprachiasmatic nucleus
(SCN; case 74). Panels are arranged rostral-to-caudal, and were traced
from sections as described in text. See page xv-xvi for abbreviations.

50

 

 

 

 

(MW

iiv‘séfil‘l‘” "

.lllll'f‘l“ ‘

OCX

 

 

 

WW a 1 1 ‘ , All"?
1 MUAW‘W ...IHM‘,H .. “ml“,

minim

 

 

 

 

 

 

/,»\)§\PVT
\\ CM ‘/
\‘-<-—.../’

 

 

 

Figure 3.2 (continued).

51

 

 

_, NII‘I'M 1| \( :3

 

 

 

     

IIIIIM.

 

 

:‘H‘IIII

 

 

 

 

     
         

  

     
   
  
   

 

 

   

  

I
g 1“ 10‘ 1
“1“ -II\“ ‘11I1““‘“ IIII "M11" h

I‘II
",iIII‘m

    

 

{1 “I“ IIH
“II ,I'I‘ IN” 1iI I

 

 

W11 \“I'l I
M“ ,IIIII“1III‘\~11‘1 IIIIIIII IIIIII
III‘ I1111II“ IIII‘
I QsII N1‘111I‘ 11111' ,/ 1,IIIII ,IWH
II 11“1 1II1, "“I“ I”l “
5“,» ,1 1 111 1151,0111“ 11‘”
\ ‘ Sm ‘ \\:11IIII‘“ 1111‘“ v
.\_‘\ . /_-.
\ I1 ”\LHb [MM
\ '
, ., \'1 11;":

 

Fig ure 3.2 (continued).

52

 

 

 
  
   
 
   

 

 

 

hits in the SCN and LSPV respectively, with descriptions of the other cases

included when they differ.

Injections centered in the SCN:
Following injections centered in the SCN, BDA-labeled fibers were most

dense in periventricular and medial aspects of the hypothalamus, parts of the
basal and limbic forebrain, and a few thalamic targets (Fig. 3.2). Fibers were of
variable thickness, usually varicose, and in most cases exhibited moderate to
heavy branching. This pattern was nearly identical in cases 74 and 81, with

slightly increased fiber density overall in # 81 compared to # 74.

Rostral Paths:
Rostrally-directed fibers from the SCN were organized in two pathways.
The first coursed ventrally along the ocx through the medial preoptic area
(mPOA), terminating in the diagonal band of Broca and the vascular organ of the
lamina terminalis (Fig. 3.2a-d). This pathway included a dense group of thick
fibers that traversed the ventromedial preoptic area, anteroventral periventricular
nucleus and medial preoptic nucleus, and more diffuse, mixed-caliber fibers that
innervated more dorsal parts of the mPOA (Fig. 3.2c-d). The ventrolateral and
later 3' preoptic areas were lightly innervated. Rostral to the mPOA, thin fibers
bra”Ched laterally and dorsally to innervate the vertical and horizontal diagonal
band» With many fibers terminating in and around the vascular organ of the
Iamina terminalis (Fig. 3.3a). Few fibers were observed rostral to the vascular

Organ .

53

SCN _ ‘ LSPV

 

Figure 3.3: Photomicrographs of BDA- labeled
fibers in the vascular organ of the lamina terminalis
(OVLT; A, B) and anteroventral periventricular nu-
cleus (AVPV; C, D) of cases 74 (A, C) and 61 (B, D).
' = 3rd ventricle. Scale bar = 100 pm.

SCN LSPV
B

BNST

Figure 3.4: Photomicrographs of BDA— labeled fibers in
the lateral septum (LS; AB) and bed nuclei of the stria
terminalis (BNST; CD) of cases 74 (A,C) and 61 (B, D).
* = lateral ventricle. Scale bar = 100 pm.

55

The second rostrally-directed pathway ran more dorsally through the
hypothalamic periventricular nucleus and medial preoptic nucleus, then
continued rostrally and laterally into the septal region (Fig. 3.2a-e). These
rostrally-directed fibers coursed under and around the edge of the anterior
commissure to innervate more dorsal and lateral aspects of the mPOA and the
medial septum before terminating in the ventral and, to a lesser extent, the
intermediate lateral septum (Fig. 3.4a). These fibers were thin, and once past
the anterior commissure became very straight with few branches or varicosities.

Once in the lateral septum, however, abundant branching, varicosities and

terminals were observed.

Midline thalamic paths:
Dorsally-projecting fibers also continued caudal to the anterior
COfT‘Imissure to densely innervate the anterior paraventricular thalamus (PVT),
the Darataenial nucleus, and, to a lesser extent, medial aspects of the bed nuclei
of the stria terminalis (Fig. 3.2e-g). More fibers were present in medial divisions
of the bed nuclei in # 81 than in # 74, and neither case exhibited fibers in the oval
nucleus of the bed nuclei (Fig. 3.40). A moderate number of fibers branched
dorsal to the third ventricle to innervate the contralateral PVT. Some of these
fibers also continued caudally to innervate the mid- and posterior PVT; however,
fibers Were far more dense in the anterior PVT than more caudal aspects of the

nucleus. At all levels of the PVT, fibers were visible along the ependymal wall

I‘ .
Imng the dorsal third ventricle (FIQ- 3.5a).

56

SCN

 

Figure 3.5: Photomicrographs of BDA- labeled fibers in
the paraventricular thalamic nucleus (PVT; A, B) and the
centers of the injection sites (C, D) of cases 74 (A, C) and
61 (B, D). * = 3rd ventricle. Scale bar = 100 um.

57

Lateral Path:

A moderate to dense plexus of fibers exited the SCN and coursed laterally
along the optic chiasm to the supraoptic nucleus. The majority of these fibers
either terminated just outside the supraoptic nucleus or continued rostrally,
although some fibers did enter the supraoptic nucleus. A few fibers followed the
optic tract from the SCN to the ventral lateral geniculate nucleus and
intergeniculate leaflet (IGL). These fibers were thin, largely without varicosities
or branches, and appeared sparse as they coursed along the optic tract and
through the ventral lateral geniculate. However, the terminal field in the IGL was
quite robust (Fig. 3.6a). Fibers within the IGL were thin and spindly in
appearance, with clear terminal morphology on the ipsilateral and contralateral
IGL. This path was the most bilateral of all the paths; fiber density in the IGL was
nearly equal on the two sides of the brain. A few sparse fibers were seen in the
amygdala close to the optic tract in # 81 but not in # 74; these were localized to

the medial amygdalar nucleus and its immediate vicinity.

Caudal paths:

Caudally-directed fibers from the SCN were organized into two pathways.
The first and most dense of these was in the sPVZ, particularly the area
immediately dorsal to the SCN which we have previously described as the LSPV
(Fig. 3.29-k; Schwartz et al., 2004). Branching and terminals were evident
immediately as these fibers exited the SCN such that the entire region

surrounding the SCN was moderately innervated with thin and thick fibers. The

58

SCN LSPV
8

Figure 3.6: Photomicrographs of BDA- labeled fibers
in the intergeniculate leaflet of cases 74 (A) and 34
(B). Scale bar = 100 pm.

59

distribution of these fibers overlapped almost completely with that of cFos and
calbindin-positive cells previously reported in the grass rat LSPV (Schwartz et al.,
2004), and fiber density in the LSPV (dorsal to the SCN) was noticeably greater
than in the area lateral to the SCN (Fig. 3.50). This plexus of fibers continued
dorsocaudally to innervate the dorsal sPVZ, the PVN (Fig. 3.2i-k), and the
dorsomedial hypothalamic nucleus (DMH; Fig. 3.2m-p). While many fibers
terminated in the sPVZ just ventral to the PVN, many were also evident within
periventricular and dorsal parvocellular compartments of the PVN, with fewer in
the medial parvocellular compartments and only sparse fibers in the
magnocellular subnuclei (Fig. 3.8a, 3.8c). A few fibers continued past the PVN
dorsally as far as the Reuniens nucleus and medial zona incerta (Fig. 3.2j-k).
Caudal to the sPVZ and PVN, labeled fibers surrounded the area dorsal and
medial to the ventromedial hypothalamic nucleus, with some fibers penetrating
the nucleus itself (Fig. 3.2l-n). These fibers were thin and twisted, with many
terminals and moderate branching (Fig. 3.9a, 3.90). All parts of the DMH were
innervated, though fibers were particularly concentrated in the pars compacta of
the DMH, with fewer fibers in other subregions (Fig. 3.2n-o; Fig 3.8e, 3.8g).
Dorsal to the DMH, thin spindle-like fibers ran dorsocaudally through the
periventricular gray to the habenula and precommissural nucleus (Fig. 3.2n-p).
Caudal to the DMH, a plexus of fibers that was sparse to moderately dense was
observed in the dorsal tuberomammillary nucleus, with some of these fibers

continuing caudally to innervate the supramammillary nucleus.

60

 

  
 

      
 
 
 

   

I1I1" 1
I 1
,‘I‘III

    

 

 

ll’Ilh‘ I
1‘ ,11I‘II‘ , 1‘ ,I 1 '1
IIIIII I111“ I ‘ ‘ 1 ‘1 , 1\ “ 1 II ‘ 1‘1‘
1II' 1 “I ‘ I ‘ ‘1 l I
. 1 II‘1‘ , 1 III .‘I‘ 1I1I‘1 III I ,"11' ‘1‘
1,1,11II'“ 1 1111‘I‘I‘ \II'HH‘ 1“ 11 I H‘1 I MIII'I |..II‘I ,1,le ”III“
1I ‘11 1 1,II1 ,‘ ,1 , ,1I .1III 1 III 1 II ‘1 ,III, 1I
'“ I, 1“" ,11'” 1‘ "‘ ‘ ,1111I1I‘II-‘ 11 , III .,11I1I‘ .111‘11 1.,1‘“ 1II1I”‘ «MW “1
11 ' 111 1111‘ ‘ 11“" ,i ,I1I1 .“ 1 1‘ 1i
‘ 1“" 1. ‘I ‘ fl 11‘I111‘III‘IH 11II I‘I‘H‘ ' 1 I1II‘1IHW 11‘ I” ‘ 1.11I‘.‘1\‘ 1“” ,II III "N 1‘1‘III‘I‘H I1I’IIIIIIU‘ 11I1IIIIN

 

 

 

 

 

 

 

 

 

 

Figure 3.7: Line drawings depicting the distribution of BDA- labeled fibers
following an injection centered in the lower subparaventricular zone
(LSPV; case 34). Panels are arranged rostral-to-caudal, and were traced
from sections as described in text. See page x-xii for abbreviations.

61

 

IIIIL. 1‘ 111 '1 D3v ‘II

‘\1IIII1I"II \ I

 

           

 

    
  

1‘ 1‘ I,

1lI1 I‘ y.111 II‘I' I:I:1IIIII::

111“ ”III“ ”WI WI! WI II
,1

IIIII\ 1 l' 1
I‘HIII I M\ IIII ,
311 II . l
IIIII I 1I1‘1Is III‘I IIII'II‘M. ,VIIII‘II‘IHI h‘lI‘N I, I 1“III
I III‘WIH III .,1\I‘ ”IIII” IIIIII‘I: VI-H , I '11II‘1III I \ V“I‘m!

\‘I'I I
I1
In 'IIIIII1‘1‘III‘,

 

 

 

 

 

 

III 1 .

111II‘

 

 

 
 

   
 
 
   
    

     
 

III IIII

 

       

 

 

,II 11' 3,22%“ . III“ III III I.
. '1 1 III IIIIII .‘
1II‘II IIIII' I‘I11IIII1 -.1 11IIIII‘,II1IIIII 1 :
IIII III“ III"? I III\ 4.4. _ L g , 1II IIIIII IIII:
IIII‘ iIII \IIIIIIIII‘IIIIII‘. IIII“\IIIIII\‘.\IIIIII‘\I\I II‘I“ \ I 1 1 I 1 II n. 1 ' IIIIIIIIIII“ IIIIIIIy H ~|1IIIIIIIIIIII‘II‘II‘II‘1‘\LNIIIIIIIIIIIIMW

 

 

 

Figure 3.7 (continued).

62

 

  

II I

IIIII I. II I ~
IIIIIII1 I1I111IIII IIIIIIIII

1,111IIIII IIIIIIIII ..1IIIIII

 
 
   
  
 
   
   

I1 II” III
1I ,_II1I11I,III1I111IIII I1II

 

 

 

 

 

 

.11 . IIIIIIII»I.1III1I11 IIIIIIIIIIII'IIIIIIII‘ IIIIIII'II‘ . 111 IIIIII- IIIII 111‘" 1.11I II“||I' IIIIIIIIIIIIM IIIIIIIII I' IIIIIIIIIIH IIIIIIIIII11IIIIIIII
T- II‘ 111" IIIIIIIIII IIII 1II ., 1 ‘HMHhuIIIII ,1 II IIIIIII III NBC I1 IIIIIIIIIIIIIIIIIIIIIII I
1| '1 ‘I'I‘IIIII .' 1

   

 

 

 

II” '.

\
l' I
I1III1II "4
III‘I' III.1III

 

 

 

Figure 3.7 (continued).

63

The final fiber pathway projected caudally through the retrochiasmatic
area to innervate more ventral and ventrolateral aspects of the hypothalamus.
These fibers spread throughout the ventral tuberal region of the hypothalamus,
providing moderate input to the ventromedial nucleus, low-to-moderate input to
the anterior hypothalamic and perifornical areas, and few fibers in the lateral
hypothalamus (Fig. 3.2l-o). Fibers also extended towards the dorsal

tubermammillary nucleus where they merged with fibers from the sPVZ pathway.

Injections centered in the LSPV:

The distribution of BDA-ir fibers following injections into the LSPV was
quite similar to that described following injections into the SCN. However, LSPV
efferents tended to be more evenly distributed along the medial-lateral axis of the
hypothalamus than were SCN efferents, which were more concentrated in the
medial and periventricular zones. LSPV fibers, like SCN fibers, were of variable
caliber, often varicose, and displayed frequent branching. Proximal to the
injection center, many thin, varicose fibers radiated dorsally and dorsolaterally
from the LSPV through the nearby periventricular nucleus and anterior
hypothalamic area. Very few fibers penetrated the SCN, and these were
restricted to the area closest to the injection site (Fig. 3.5d). Sparse fibers were
also visible in the ependymal layer near the SCN, as well as in the contralateral

LSPV.

Rostral paths:

Anterior to the SCN, fibers extended rostrally from the LSPV along the ocx

to innervate the ventromedial preoptic area and the anteroventral periventricular

64

nucleus (Fig. 3.7a-d), and to a lesser extent the ventrolateral preoptic area and
the vascular organ of the lamina terminalis. Fibers also lined and entered the
ependymal wall of the third ventricle around the anteroventral periventricular
nucleus (Fig. 3.3b, 3.4d). While this was still a substantial projection, the fibers
were thinner and fewer in overall number than those seen after SCN injections.

A larger fiber plexus radiated dorsolaterally from the LSPV through the
mPOA. These fibers spread out from the midline at more rostral levels to provide
a diffuse input to an area extending from the basal forebrain and diagonal band
of Broca to the lateral septum (Fig. 3.7b-c). The medial septum was most heavily
innervated close to the anterior commissure, and in # 34 more fibers were seen
at the ventricular edges of the lateral septum than in all other cases. Fibers were
denser in the horizontal than the vertical limb of the diagonal band in # 34 and #
40, whereas in # 61 and # 75 the inputs to the two limbs were similar. While the
septum was also heavily innervated by the SCN, in general the LSPV projection

along this path was more dense than that of the SCN (Fig. 3.4a-b).

Midline thalamic paths:

A third major fiber plexus ran dorsally from the LSPV to the PVT and, to a
lesser extent, the anterior aspects of the parataenial and Reuniens nuclei (Fig.
3.7f-g). These fibers were longer and thinner than those emanating from the
SCN, branched significantly and had many boutons and terminals. As they did
following injections into the SCN, labeled fibers in the PVT entered the
ependymal layer, sometimes contacting the dorsal third ventricle (Fig. 3.5b).

Fibers also extended into the bed nuclei of the stria terminalis just anterior to the

65

PVT, innervating medial aspects and avoiding the oval nucleus (Fig.7e).
Whereas fibers in the PVT were somewhat less dense than those seen following
injections of the tracer into the SCN, labeled fibers in the bed nuclei were slightly

more dense following injections into the LSPV (Fig. 3.4c-d).

Lateral paths:

A small group of fibers coursed along the optic chiasm laterally from the
SCN to lightly innervate the supraoptic nucleus (Fig. 3.79). Some of these fibers
continued to follow the optic tract, and entered the amygdala. In cases 34 and
61, where injections were centered squarely in the LSPV, fiber density in the
amygdala was sparse, whereas more rostral injections (# 40 and # 75) resulted
in greater numbers of labeled fibers. The medial amygdalar nucleus was the
only part of the amygdala with BDA-ir fibers in all 4 cases of injections into the
LSPV. Because the increased innervation of the amygdala was associated with
more rostrally-centered injections, it seems likely that some of these fibers
originated in the caudal preoptic area or anterior hypothalamus rather than the
LSPV.

In cases 34 and 75, a few sparsely-distributed fibers were visible in the
IGL (Fig. 3.6b), but none were observed in # 40 and # 61. Given the similarity in
location of injections 34 and 61, and the fact that the larger injection of the two (#
61) did not label fibers in the IGL, it is not possible to confirm a projection from
the LSPV to the IGL. However, the injection site for # 34 and # 75 extended

slightly further laterally than those of # 61 and # 40, so the fibers seen in # 34

66

and # 75 could have arisen from this region. Further studies will be necessary to

verify the location of neurons in and around the SCN that project to the IGL.

Caudal paths:

Caudal to the SCN, fibers from the LSPV coursed along two pathways.
The majority of these efferents extended dorsocaudally through the sPVZ, with
some fibers terminating ventral to the PVN and others continuing into the nucleus
(Fig. 3.7h-j). Although some fibers were present in the periventricular subregion,
the most substantial input from the LSPV was in the caudal parvocellular
subregion (Fig. 3.8b), where fibers spread out laterally, generally avoiding
magnocellular cell bodies. Innervation of the medial zona incerta was more
dense, and the Reuniens nucleus less dense, than that seen following SCN
injections (Fig. 3.7j-l). These fibers were far more extensive in cases 61 and 75
than in # 34, whereas in # 40 no fibers were seen in the zona incerta or dorsal to
the PVN, and magnocellular paraventricular neurons were heavily interweaved
by BDA-ir fibers, suggesting that projections from the area just rostral to the
LSPV are very different from those of the LSPV itself. The perifornical region
was sparsely— to moderately innervated by fibers that extended caudally and
laterally from the sPVZ, PVN and zona incerta in # 34 and # 61 (Fig. 3.7l-n),
while in # 40 and # 75, fibers were dense in the perifornical region. In the DMH,
the overall density of labeled fibers was lower after injections into the LSPV than
the SCN, but fiber density increased in the pars compacta region as was noted

for SCN efferents (Fig. 3.7l-n, Fig. 3.8e-h). From this point, fibers continued

67

 

Figure 3.8: Photomicrographs of BDA— labeled fibers in the paraven-
tricular hypothalamic nucleus (PVN; A, B), dorsal subparaventricular
zone (SPVZ; C, D), and dorsomedial hypothalamic nucleus (DMH;
E-H) of cases 74 (A, C, E, G) and 61 (B, D, F, H). Adetail of the DMH
pars compacta (DMHpc) is pictured in panels G and H. * = 3rd ventri-
cle. Scale bar = 200 pm (A-F), 100 pm (G. H).

68

SCN .. -. LSPV
B

Figure 3.9: Photomicrographs of BDA- labeled fibers in
the ventromedial hypothalamic nucleus(VMH; AB) and
adjacent ventral tuberal region (VT U; CD) of cases 74
(A,C) and 34 (3,0). " = 3rd ventn‘cle. Scale bar = 300
pm (A, B), 100 pm (C, D).

69

dorsally and caudally past the DMH to innervate the posterior hypothalamic area,
subparafascicular nucleus and periventricular gray (Fig. 3.7n-o), eventually
reaching the habenula and precommissural nucleus. Fibers were thin and
straight until the precommissural nucleus where they branched and terminated.

A ventral group of fibers proceeded caudally from the LSPV along the
basal edge of the hypothalamus to innervate the retrochiasmatic area, as well as
ventral and lateral aspects of the anterior hypothalamic area (Fig. 3.7i-l'); labeled
fibers were relatively diffuse in these regions. These fibers eventually
surrounded and entered the ventromedial hypothalamic nucleus and the ventral
tuberal region lateral to it (Fig. 3.7I-n). Fibers were thin, often short, with
numerous branches and boutons. These fibers also provided input to the
perifornical region and the DMH. In contrast to the SCN efferents, LSPV fibers
were more dense within the ventromedial nucleus than in the DMH, particularly in
the central and ventrolateral subregion as well as the perinuclear region (Fig.
3.9b, 3.9d). As they extended caudally past the ventromedial nucleus, these
fibers projected to the dorsal and to a lesser extent, the ventral tuberomammillary

nuclei, and gradually decreased in density until they were no longer seen.

Discussion

The observed distributions of SCN and LSPV efferents in grass rats are
Very similar to those previously reported for nocturnal species, consisting of
Overlapping projections to the hypothalamus, thalamus and limbic forebrain. This
Suggests, first, that the basic efferent circuitry of the circadian system is highly

c0nserved in diurnal and nocturnal species. Second, while some differences in

70

the relative density of projections from the SCN and LSPV were apparent, nearly
all areas receiving fibers from the SCN also received at least some input from the
LSPV, and vice versa. These findings are consistent with the hypothesis that
convergence of LSPV terminals on SCN targets could play a role in the
promotion of differences in circadian rhythms seen in diurnal and nocturnal

species.

Common output pathways of the SCN and LSPV:

The major projections from the SCN are very similar to those of the LSPV
and can be organized into 6 pathways according to their respective termination
points as follows: 1) an anterior ventral pathway to the mPOA and diagonal band,
2) an anterior dorsal path extending into the septal region, 3) a dorsal pathway to
the PVT, 4) a lateral pathway to the supraoptic nucleus and IGL, 5) a
dorsocaudal pathway through the sPVZ to the DMH, and 6) a caudal pathway
innervating the ventral aspect of the caudal hypothalamus. A very similar
distribution for SCN and LSPV efferents exists in lab rats (Watts and Swanson,
1987; Watts et al., 1987), and in hamsters (Morin et al., 1994; Orpen and Steiner,
1994; Kriegsfeld et al.). Thus, the gross organization of SCN/LSPV outputs is
essentially the same in diurnal grass rats and at least two nocturnal species.

The majority of projections from the SCN and LSPV were restricted to the
hypothalamus, particularly in the medial and periventricular zones. However,
direct projections to hypothalamic areas involved with sleep/wake behavior such
as the ventrolateral preoptic area and median preoptic nucleus were

comparatively sparse. Circadian signals may influence sleep/wake regulation

71

indirectly through projections to the mPOA, sPVZ and DMH (Deurveilher and
Semba, 2004; Saper et al., 2005). These sites, in turn, project to the
ventrolateral preoptic area (Deurveilher et al., 2002), the median preoptic nucleus
(Deurveilher and Semba, 2003) and arousal-promoting groups such as the
tuberomammillary nuclei and the orexin neurons in the perifornical area
(Deurveilher and Semba, 2004). The preoptic area, sPVZ, and DMH are also
extensively interconnected in lab rats (Simerly and Swanson, 1986; ter Horst and
Luiten, 1986; Watts and Swanson, 1987; Watts et al., 1987; Simerly and
Swanson, 1988; Thompson et al., 1996; Gu and Simerly, 1997; Thompson and
Swanson, 1998; Deurveilher et al., 2002; Deurveilher and Semba, 2003, 2004),
and may comprise part of an integrative network regulating multiple behaviors
(Swanson, 2000), including locomotor activity and the sleep/wake cycle. Thus,
complex rhythms such as those in sleep-wake behavior could be coordinated
both through converging SCN/LSPV projections onto this medial hypothalamic
circuit, and by interactions between the components of that circuit.

In contrast, direct projections from the SCN/LSPV to more specialized
populations of cells may play a greater role in circadian regulation of
neuroendocrine functions, particularly those associated with reproduction (de la
lglesia and Schwartz, 2006). Circadian rhythms in mating behavior (Mahoney
and Smale, 2005a), the ovulatory surge in luteinizing hormone (McElhinny et al.,
1999), and cFos expression in gonadotropin-releasing hormone (GnRH) neurons
(Mahoney et al., 2004) are reversed in nocturnal and diurnal rodents. Direct

projections from the SCN onto GnRH neurons in the preoptic area, and estrogen

72

receptor- containing neurons in the anteroventral periventricular nucleus have
been described in lab rats (van der Beek et al., 1993; de la lglesia et al., 1995;
Van der Beek et al., 1997a), and both VP- and VlP-ir fibers, some of which are
presumed to originate in the SCN, form appositions with GnRH- and estrogen
receptor-positive neurons in grass rats (Mahoney and Smale, 2005b). In the
present study, both SCN and LSPV projections were seen in parts of the septum,
diagonal band, vascular organ of the lamina terminalis and mPOA that contain
GnRH and estrogen receptor-positive cells (Mahoney and Smale, 2005b). While
it is not known if the LSPV is necessary for normal reproductive rhythms in grass
rats, neural connections from both the SCN (Meyer-Bernstein et al., 1999; de la
lglesia et al., 2003) and the sPVZ (Watts et al., 1989) appear to play important
roles in the circadian control of neuroendocrine rhythms in nocturnal rodents, and
at least some of these rhythms are reversed in diurnal rodents (Mahoney et al.,
2004). The current data are thus consistent with the hypothesis that
convergence of SCN and LSPV projections on neuroendocrine cells might
promote the differences in their regulation by the circadian system.

The most robust extrahypothalamic projection from the SCN and LSPV
was to the PVT, which projects in turn to several other SCN/LSPV targets within
the hypothalamus (Moga et al., 1995). Unlike the SCN and LSPV, however, the
PVT also sends major ascending fibers to regions in the telencephalon, including
the amygdala and cerebral cortex (Moga et al., 1995); in some cases this may be
the most direct route through which SCN signals could reach these areas

(Sylvester et al., 2002). The PVT is a primary target of the SCN signaling

73

molecule prokineticin 2 in both lab rats and grass rats (Cheng et al., 2002;
Lambert et al., 2005). In the present study, the PVT was densely innervated by
both the SCN and the LSPV throughout its entire rostrocaudal extent, a pattern
consistent with a role for the LSPV in modulating circadian signals originating in
the SCN. However, daily cFos rhythms in the PVT are somewhat similar in grass
rats and lab rats kept in a light-dark cycle (Novak and Nunez, 1998; Novak et al.,
2000). This suggests that rhythmic cells within the PVT may respond differently
to SCN and/or LSPV input compared to cells in other hypothalamic regions. The
PVT also sends a robust projection to the SCN (Moga et al., 1995), and PVT
stimulation can phase-shift rhythms in drinking (Salazar-Juarez et al., 2002). It is
thus possible that the LSPV could indirectly influence the activity of SCN neurons

through the PVT.

Differences in efferents of the SCN and LSPV:

Almost all areas innervated by the SCN were also innervated by the
LSPV, and vice versa. The only possible exception to this was the IGL, in which
labeled fibers were abundant following all SCN injections, but sparse following
some LSPV injections, and absent in other cases. In hamsters (Morin et al.,
1992; Vrang et al., 2003; Vidal et al., 2005) and lab rats (Card and Moore, 1989;
Moore and Card, 1994), retrograde tracer injections into the IGL labeled small
numbers of cell bodies in and around the SCN and large numbers of cells in the
retrochiasmatic area, and anterograde tracer injections into the SCN and LSPV
each labeled fibers in the IGL (Watts et al., 1987; Morin et al., 1994; Kriegsfeld et

al., 2004). In the present study, IGL labeling was consistently associated with

74

injections centered in the SCN, but not the LSPV. However, the two LSPV
injections that did label fibers in the IGL were also centered further laterally than
the two that did not. While this difference in location is slight, the SCN is both
functionally and anatomically complex (Morin et al., 2006), and such
heterogeneity may also exist in the tissue surrounding the SCN (Ramanathan et
al., 2006). Retrograde tracing will be necessary to determine exactly which cells

within and around the SCN project to the IGL

Functional Implications:

While the SCN and the LSPV of both diurnal and nocturnal species
appear to project to essentially the same target areas, several lines of evidence
suggest that LSPV outputs may differ functionally from those of the SCN in at
least two respects. First and foremost, the SCN is likely to use different
neurotransmitters than the LSPV. While is it estimated that the majority of SCN
neurons are GABAergic (Moore and Leak, 2001), the SCN of both diurnal and
nocturnal species is both functionally and anatomically heterogeneous (Antle and
Silver, 2005). At least some SCN peptides are involved in signaling to SCN
target cells, as both VP- and VlP-ir SCN neurons send major projections out of
the SCN (Van der Beek et al., 1997a; Abrahamson and Moore, 2001; Mahoney
and Smale, 2005b), and experimental manipulations of these peptides have
revealed a role for them in regulation of neuroendocrine rhythms (Kriegsfeld and
Silver, 2006). While the identity of the transmitter(s) used by projection neurons
of the LSPV is still unknown, these are likely to differ from those used by SCN

neurons, as cells in the LSPV do not express many of the peptides and proteins

75

seen in the SCN (e.g. vasopressin, vasoactive intestinal polypeptide, gastrin-
releasing peptide, corticotropin-releasing factor; Watts and Swanson, 1987;
Moore and Silver, 1998; Smale and Boverhof, 1999; Abrahamson and Moore,
2001). This also raises the possibility that some targets may respond more, or
differently, to signals from the SCN and LSPV, depending on the type and timing
of the signal. A second difference between the grass rat SCN and LSPV is that
they exhibit different patterns of rhythmicity at the cellular level. For example,
when grass rats are kept on a light-dark cycle cFos in the LSPV rises within 5
hours of lights-off, but does not go up in the SCN until lights-on (Nunez et al.,
1999; Schwartz et al., 2004). These differences in the pattern of rhythmic
activity, and in the chemical signals used, suggest that in grass rats, the LSPV is
more than a relay for SCN signals. Rather, the SCN and LSPV may send
distinctly different signals to their various targets.

Although the grass rat LSPV and SCN are different with respect to cell
phenotypes and their patterns of rhythmicity, they are similar in other respects. In
both regions, cFos rhythms are endogenous (Schwartz et al., 2004) and are
buffered against dramatic changes in activity state (Schwartz and Smale, 2005).
These data indicate that rhythmic activity in the grass rat LSPV, like that within
the SCN, is not dependent on a light-dark cycle, nor is it heavily influenced by
sleep/wake state or behavioral arousal. Signals coming from the LSPV are thus
more tightly coupled to the pacemaker than to the external environment or
internal state. Both the SCN and LSPV also express daily rhythms in expression

of the clock gene products Per1 and Per2 (Ramanathan et al, 2006). The

76

observation that these signals appear to reach nearly all of the same target
regions as those from the SCN makes it even more likely that the output of the

LSPV is related to timekeeping in grass rats.

Conclusions:

The function of the LSPV in either diurnal or nocturnal species is not fully
understood. In lab rats, the lower part of the sPVZ may be involved in regulating
the timing of reproduction-related events (Arendash and Gallo, 1979; Docke et
al., 1982; Watts et al., 1989; Fernandez-Galaz et al., 1999) and locomotor activity
(Lu et al., 2001; Moore and Danchenko, 2002). However, circadian body
temperature rhythms persist with reduced amplitudes in animals with ventral
sPVZ lesions (Lu et al.), suggesting that, although the ventral sPVZ may be
necessary for expression of some rhythmic processes, it is not essential for all.
In grass rats, the LSPV displays an endogenous rhythm in cFos expression that
is 8 hours advanced relative to that in the SCN, whereas in lab rats, this rhythm
is in phase with the SCN rhythm under a light-dark cycle and is not rhythmic
under constant dark conditions (Schwartz et al.). Thus, rhythmic activity in the
grass rat LSPV is different from both the activity of its own SCN, and from the
activity of the LSPV in nocturnal lab rats. The clear parallelism in the efferents of
the SCN and LSPV in grass rats demonstrated here is consistent with the
hypothesis that projections from the grass rat LSPV may converge with those
from the SCN on their target cells. If this is the case, then signals from the LSPV

could alter the responses of these cells to signals originating in the SCN, and

77

perhaps do this differently in diurnal grass rats compared to nocturnal species

such as lab rats.

78

CHAPTER 4
The suprachiasmatic nucleus and the lower subparaventricular

zone project onto gonadotropin-releasing hormone neurons in

the Nile grass rat (Arvicanthis niloticus)

Introduction:
In female rodents, reproductive behavior involves the precise temporal

coordination of multiple physiological and behavioral events. The activation of
gonadotropin-releasing hormone (GnRH) neurons underlying the ovulatory surge
in leuteinizing hormone (LH) requires both a hormonal signal, mediated by
circulating levels of estrogen that peak on the day of proestrus, and a daily
temporal signal. For example, nocturnal female laboratory rats (Raltus
norvegicus, hereafter referred to as ‘lab rats’) given barbiturates on the afternoon
of proestrus delay ovulation by 24 hours (Everett and Sawyer, 1950), and
ovariectomized female rats with constantly elevated levels of circulating estradiol
exhibit daily LH surges (Legan and Karsch, 1975). Rhythms in the timing of the
LH surge also free-run in female lab rats housed in constant lighting conditions
(Cheung and McCormack, 1983; Takeo, 1984), indicating that the daily signals
governing GnRH activation and the LH surge are endogenous. In diurnal rodents
Such as Nile grass rats (Arvicanthis niloticus), mating behavior occurs in the early
morning, 180 degrees out of phase with the time of day at which it occurs in lab
rats (Mahoney and Smale, 2005a). This difference is accompanied by reversals

in rhythms in the LH surge (McElhinny et al., 1999) and cFos expression within

79

GnRH neurons, and these phase differences persist in constant darkness
(Mahoney et al., 2004). However, the neural basis of these species differences
is not known.

Circadian influences on reproductive events in rodents are governed by
the suprachiasmatic nucleus (SCN), which contains the primary mammalian
circadian pacemaker (Moore and Eichler, 1972; Stephan and Zucker, 1972).
However, the SCN appears to be similar in day- and night-active species. For
example, daily rhythms in expression of the clock genes part and per2 peak at
the same times of day in the SCN of diurnal and nocturnal species (Hastings et
al., 1999; Mrosovsky et al., 2001; Yan and Okamura, 2002; Caldelas et al., 2003;
Ramanathan et al., 2006). lndices of cellular activity in SCN neurons such as 2-
deoxyglucose uptake (Schwartz, 1991) and multiple-unit activity (Sato and
Kawamura, 1984b; Kurumiya and Kawamura, 1988) are also similar in diurnal
and nocturnal species (but see Jiao et al., 1999). These similarities suggest that
the mechanisms underlying differences between diurnal and nocturnal patterns
Of rhythmicity are located downstream of the pacemaker in the SCN.

The subparaventricular zone (sPVZ) dorsal to the SCN may be part of
Such a mechanism. The sPVZ extends from the dorsal border of the SCN along
the midline to the hypothalamic paraventricular nucleus, and is a major target of
the SCN in nocturnal and diurnal species (Watts and Swanson, 1987; Watts et
al. . 1 987; Morin et al., 1994; Abrahamson and Moore, 2001; Moore and Leak,
2001). The sPVZ in turn projects to many of the same target regions as the SCN

in bOth lab rats (Watts and Swanson, 1987; Watts et al., 1987) and in diurnal Nile

80

grass rats (chapter 3). Daily rhythms in multiple-unit electrical activity
immediately outside the SCN are reversed in diurnal Siberian chipmunks
compared to lab rats (Kubota et al., 1981; Sato and Kawamura, 1984b), and
circadian rhythms in cFos expression in the lower part of the subparaventricular
zone (LSPV) are dramatically different in diurnal Nile grass rats and nocturnal lab
rats (Nunez et al., 1999; Schwartz et al., 2004). This region thus displays the
characteristics necessary to communicate circadian signals indirectly to targets
that are also innervated by the SCN, and to do that differently in diurnal and
nocturnal animals (Schwartz et al., 2004).

Direct projections from the SCN are likely to play a major role in circadian
regulation of neuroendocrine functions, including those associated with
reproduction (de la lglesia and Schwartz, 2006). Direct projections from the SCN
onto GnRH neurons in the preoptic area (POA), and estrogen receptor- (ER-)
containing neurons in the anteroventral periventricular nucleus (AVPV) have
been described in lab rats and hamsters (de la lglesia et al., 1995; Watson et al.,
1 995; Van der Beek et al., 1997a). In lab rats, these projections may use

Vasoactive intestinal polypeptide (VIP; van der Beek et al., 1993) and/or
VaSOpressin (VP; Kalamatianos et al., 2004) as transmitters. In grass rats, both
VP and VIP-positive fibers, some of which are thought to originate in the SCN,
for m appositions with GnRH- and ER-positive neurons (Mahoney and Smale,
2OOSb). However, rhythms in VP and VIP mRNA in grass rats and lab rats are
Similar (Krajnak et al., 1998; Mahoney et al., 2006), suggesting that differences in

the timing of rhythmic activity in GnRH neurons may be regulated by differences

81

in responsiveness to SCN signals and/or other sources of inputs to these cells
(Mahoney and Smale, 2005a).

In grass rats, both the SCN and the LSPV project to areas containing
GnRH- and ER- positive cells, including the medial septum (MS), the diagonal
band of Broca (DBB), the vascular organ of the lamina terminalis (OVLT), and
the POA (chapter 3). The general distributions of these efferents are consistent
with the idea that signals from cells in both the LSPV and the SCN influence
rhythms in target cell populations regulating reproduction. However, it is not
known whether LSPV neurons project directly onto GnRH neurons in any
species. If the LSPV is involved in maintaining a diurnal pattern of circadian
rhythmicity, then cells receiving input from it should show different patterns of
rhythmicity in diurnal and nocturnal species. Here, we used injections of the
anterograde tracer biotinylated dextran amine (BDA), in conjunction with light-
level and confocal approaches, to determine if SCN and/or LSPV cells project

directly to GnRH neurons, a population of cells that is rhythmically inverted in

diurnal and nocturnal species.

Materials and Methods:

Arlimals
Animals were adult female grass rats (Arvicanthis niloticus) obtained from

a breeding colony at Michigan State University (Katona and Smale, 1997).
Ahimals were singly housed in Plexiglas cages (34 x 28 x 17 cm) with access to
food (PMI Nutrition Prolab RMH 2000, Brentwood, MO) and water ad Iibitum.

Aninnals were kept in a 12:12 light: dark (LD) cycle; a red light (<5 lux) remained

82

on constantly for animal care purposes. All experiments were performed in
compliance with guidelines established by the Michigan State University All-

University Committee on Animals Use and Care, and the NIH Guide for the Care

and Use of Laboratory Animals.

Surgical Procedures:
All animals received a unilateral iontophoretic injection of 10,000 MW BDA

(Molecular Probes; Eugene, OR) directed at the SCN and LSPV. Animals were
deeply anesthetized with sodium pentobarbital (50mg/kg i.p.) prior to surgery.
The animal’s head was shaved, sterilized with betadine, and injected with 0.03 cc
lidocaine s.c. The animal was then placed into a stereotaxic apparatus (Stoelting
00.; Wood Dale, IL) with the incisor bar set to -6.0 mm and the micropipette
angled 10 degrees, and a 1-2 cm incision was made in the scalp to expose the
skull. Injection coordinates were: 1.4 mm anterior to bregma, 1.2 mm lateral to
bregma, and 6.3 mm ventral to the dura. A small hole was drilled in the skull,
and the dura was gently broken to allow uninterrupted passage of the
micropipette. A glass micropipette (WPI; Sarasota, FL) with inner tip diameter of
1 0-15 pm was then filled with 10% BDA in 0.01 M phosphate-buffered saline
(PBS; pH 7.4) and lowered slowly to the final position. Injections were made
UDder 5 mA positive alternating current at a rate of 7 seconds on/ 7 seconds off,
for 7 minutes, and were followed by a 7-minute rest period. Micropipettes were
then withdrawn slowly under negative current. The incision was closed with
autOclips, and animals were given 0.06 cc buprenorphine i.m. and 1.0 cc sterile

Sa'ine s.c. Seven days later, animals were given an overdose of sodium

83

pentobarbital and perfused transcardially with 0.01M phosphate-buffered saline
(PBS), pH 7.2, followed by 4% paraformaldehyde (Sigma-Aldrich Co., St. Louis,
MO) in 0.1M phosphate buffer. Brains were post-fixed for 4 hours and then
transferred to 20% sucrose solution overnight. Brains were sectioned into 3

series coronally at 30 pm on a freezing microtome, and sections were stored in
cryoprotectant at -20°C until processing. One series of sections was used here;

the other two series were used in a separate study (chapter 3).

Light-level analyses:
For analysis of tissue at the light level, we used DAB with and without

nickel as the chromagen. Unless noted otherwise, incubations were conducted

at room temperature for 1 h on a shaker, and tissue was rinsed 3 x 10 min in
PBS between incubations. Free-floating sections were rinsed in PBS, and then
incubated in avidin-biotin peroxidase complex (ABC; ABC Vectastain Kit, Vector
Laboratories, Burlingame, CA). Tissue was rinsed in sodium acetate buffer and
BDA was visualized by reacting with diaminobenzidine (DAB; 0.5mg/mL; Sigma)
enhanced with nickel sulfate in acetate buffer. Sections were rinsed in PBS, then
incubated in i) 5% normal horse serum (Vector) in PBS with 0.3% Triton X-100,
ii) mouse anti-GnRH antibody (1 :5000; HU4H antibody characterized in Urbanski,
1 991) on a rotator for 48 h at 4 °C, iii) biotinylated horse anti-mouse antibody
(1 3200; Vector), iv) ABC. Protein was visualized by reacting with
diar‘ninobenzidine (0.5mg/mL; Sigma) in Trizma buffer (Sigma). Sections were

then mounted, dehydrated, and coverslipped with Permount.

84

GnRH-ir neurons with and without BDA-ir appositions (also referred to as
contacts) were counted in 3 rostrocaudal zones encompassing the MS and DBB
(6 sections), the OVLT (2 sections), and the POA and SCN (6 sections).
Sections were examined for appositions on a Zeiss Axioskop microscope (Zeiss,
Gottingen, Germany) under brightfield illumination using a 100x oil immersion
lens. A GnRH-ir neuron was considered to be contacted by a BDA-ir fiber if, in
the same plane of focus, a swelling on a BDA-ir axon (blue-black fiber) was
observed abutting a GnRH-ir cell body (brown cell), or a labeled process
emanating from a GnRH-ir cell body. All photomicrographs were taken using a
Zeiss Axiocam MR (Zeiss) and processed in Adobe Photoshop 7.0 (Adobe

Systems; Mountain View, CA). The ‘sharpen’ filter was applied once to each

image, and then images were adjusted for brightness and contrast.

Confocal analysis:
Free-floating sections were rinsed in PBS, and then incubated in i) ABC, ii)

biotinylated tyramide from a Renaissance TSA-indirect Kit (Perkin-Elmer; Boston,
MA), iii) FlTC-conjugated streptavidin (1:200; Molecular Probes). From this point
onwards, tissue was isolated from ambient light during all incubations and rinses.
Sections were then rinsed in PBS and incubated in i) 5% normal donkey serum
(Jackson Laboratories, Bar Harbor, ME) in PBS with 0.3% Triton X-100, ii) a
Combination of mouse anti-GnRH (1 :5000; HU4H) and rabbit anti-VIP (1 :1000;
'mmunostar, St. Paul, MN) antisera for 48 hours at 4 degrees C, iii) a
ColTibination of Texas Red-conjugated donkey anti-mouse (1:200; Jackson) and

c33’5~conjugated donkey anti-rabbit (1:200; Jackson) antisera. Sections were

85

mounted onto gelatin-coated slides and coverslipped with ProLong Gold antifade
reagent (Sigma). Upon examination, VlP-ir fibers, while clearly visualized in and
around the SCN as well as in the bed nuclei of the stria terminalis, were not
reliably visualized in other areas previously shown to contain them (Mahoney and
Smale, 2005b). For this reason, tissue was analyzed only for BDA-ir appositions.
GnRH neurons were examined for appositions with BDA-ir fibers under
mercury vapor epifluorescence using a 40x objective. Those GnRH-ir neurons
with apparent BDA-ir contacts were then examined on a Zeiss LSM 5 Pascal
confocal microscope (Zeiss). Using a 63x oil immersion objective, GnRH
neurons were optically sectioned through the cell body at 0.42 pm. Each z-stack
was generated using 2 tracks. The first track visualized FITC (BDA) using a 488
nm argon-ion laser and a 123 um pinhole diameter; the second track visualized
Texas Red (GnRH) using a 543 nm helium-neon laser and a 108 um pinhole
diameter. All settings were kept constant for each z-stack. Contacts between
BDA and GnRH neurons were verified using the LSM Image Browser (Zeiss) that
allows reconstruction, rotation and viewing of z-stacks in 3 dimensions. 2-
dimensional single frames and projections were converted to TIF format and

adjusted in Photoshop as described above.

Resufls:

The distribution of GnRH-ir cells:
The appearance and distribution of GnRH neurons were similar to those

described previously (McElhinny et al., 1999; Mahoney and Smale, 2005b).

GnRH neurons had a fusiform appearance, with processes extending from both

86

ends of the cell, often for considerable distances. Rostrally, small numbers of
GnRH neurons were observed in the MS and vertical DBB. These neurons were
oriented dorsoventrally, and were most common on or close to the midline.
Numbers of GnRH neurons increased dramatically around the OVLT and the
adjacent DBB and rostral POA. In these regions GnRH neurons were oriented
along a dorsomedial-ventrolateral axis close to the OVLT, and into the vertical
DBB and septal region. Caudal to the OVLT, GnRH neurons became clustered
in two areas. One group of cells was concentrated in the medial POA, whereas
the other extended laterally and caudally parallel to the optic chiasm where they
loosely surrounded the rostral and medial supraoptic nucleus. There were fewer

GnRH neurons in the MS/DBB than in the OVLT or the POA/SCN.

BDA injection sites:

BDA injections were centered in the SCN in cases 74 and 81. Both
injections filled the entire SCN, and in # 74 the injection did not extend beyond
the borders of the SCN, whereas the injection in # 81 bled slightly into the RCh
(Fig. 1). In both cases numerous BDA-ir fibers were apparent throughout the
region in which GnRH neurons were distributed; the density of labeled fibers in
this area was slightly higher in case 81 than 74. Overall, BDA-ir fibers were far
more dense ipsilaterally than contralaterally, although certain targets exhibited
highly bilateral innervation; these targets included the M8, the vertical limb of the
DBB, the area surrounding the OVLT, and the POA.

BDA injections were centered around the SCN in 4 cases; of these, 2

were limited to the LSPV (34 and 61) and 2 extended beyond the LSPV laterally

87

and dorsally (40 and 75). In none of these 4 cases did the tracer bleed into the
SCN. The distributions of BDA-ir fibers in 34 and 61 were very similar to those
seen following SCN injections, with a slightly higher density of labeled fibers in
case 61 than 34. Fiber densities through the region containing GnRH neurons
were lower than those seen following SCN injections in all cases except for # 75.
In cases 40 and 75, the distributions of BDA-ir fibers were generally similar to
those of # 34 and # 61, although fibers extended further laterally past the DBB,
and covered more of the mPOA, particularly in #75. Fiber densities through the
areas containing GnRH neurons were also higher in both of these cases than in
either # 34 or # 61. Overall, labeled fibers were more dense ipsilateral to the
injection site, although contralateral fibers were evident, as was the case

following injections into the SCN.

Appositions between GnRH-ir neurons and BDA-ir fibers:
Following injections into both the SCN and the LSPV, BDA-ir fibers were

interspersed with GnRH-ir neurons at all rostrocaudal levels, but were most
dense in and around the OVLT and the adjacent POA. Varicose BDA-ir fibers
were often found in close association with GnRH-ir cell bodies and proximal
processes (Fig. 2, 3), and in some cases these labeled fibers appeared to end in
a large terminal bouton-like structure (Fig. 2a). In at least one case, a single
BDA-ir fiber was observed to contact 2 GnRH-ir neurons in the same section
(Fig. 3a, b). Varicose BDA-ir fibers were also frequently observed in apposition

with GnRH-ir processes at some distance from the cell body (which was not

88

 

 

 

 

 

 

 

 

 

 

Figure 4.1: Schematic depiction of BDA injection sites in and around the
SCN. See page x-xii for abbreviations.

89

always visible; Fig. 2e, 3d). These fiber-fiber appositions were observed
throughout the region of the OVLT and adjacent POA.

Following BDA injections into the SCN, appositions were concentrated in
the mid-to-caudal parts of the distribution of GnRH cells and were found on both
sides of the brain. In the MS/DBB 17.2 % — 31.6 % of GnRH neurons were
contacted by BDA-ir fibers (Table 4.1 ); these proportions were similar on the two
sides of the brain. In the OVLT and POA/SCN, 53.1 % — 65.6 % and 44.2 % —
84.8 % of GnRH neurons, respectively, were contacted by BDA-ir fibers. The
difference between ipsilateral and contralateral sides was low in the MS/DBB and
OVLT. In the POA/SCN, there were more contacts on the side ipsilateral to the
injection site; this was especially true in case 81, in which the tracer extended
somewhat caudal to the SCN.

Following injections centered in the LSPV, appositions were concentrated
in the mid-to-caudal parts of the distribution of GnRH neurons, where they were
most numerous ipsilateral to the injection site. Rostrally, however, the
projections were more evenly distributed. Specifically, in the MS/DBB 4.2 % -
26.8 % of GnRH neurons ipsilateral to the injection site were contacted by BDA
fibers, while 16.7 % — 25.0 % of contralateral GnRH neurons apposed BDA fibers
(Table 4.1). In the OVLT, 26.7 % — 73.5 % of GnRH neurons formed appositions
with BDA-ir fibers from the LSPV, as did 17.3 % — 79.5 % of GnRH neurons in
the POA/SCN area. In both regions, ipsilateral appositions increased with
injection size. Numbers of appositions in the OVLT and POA/SCN were lower on

the side contralateral to the injection site than ipsilateral to it in all 4 cases, and

90

 

Figure 4.2: Photomicrographs of appositions between BDA-ir
fibers (black) and GnRH neurons (brown; A — D) and processes
(E) following injections centered in the SCN. BDA-ir fiber in (E)
is not the same fiber pictured in (D). All images were taken using
a 100x oil objective. Scale bar = 20 pm.

91

varied less from case to case (OVLT, 26.7 % - 37.5 %; POA/SCN, 17.3 % - 34.8
%). The distribution of GnRH neurons forming appositions with fibers from the
LSPV was thus similar to that of neurons contacted by SCN fibers, in that
appositions were more frequent in the OVLT and POA/SCN. However, LSPV
efferents formed more appositions ipsilaterally than contralaterally in these areas,

whereas SCN efferents projected more equally to the two sides.

Confocal imaging:

In two additional animals with BDA injections centered in the LSPV,
confocal microscopy was used to examine the relationship between BDA-ir fibers
and GnRH cell bodies in greater detail. In cases 87 and 90, injections were
comparable in size to those of # 61 (Fig. 1). Case 87 was centered in the caudal
LSPV, dorsolateral to the caudal SCN. In case 90, the injection was centered
slightly more rostral than # 87, with significant coverage in the LSPV. In both
cases, the distributions of BDA-ir fibers were consistent with those from the
previously-described LSPV injections. BDA-ir fibers from the LSPV directly
contacted GnRH-ir cell bodies and processes, forming appositions very similar in
appearance to those observed using light-level microscopy (Fig. 4). This was
confirmed by rotation of the image in 3 dimensions, as well as by examination of
appositions in single optical sections (0.42 pm thickness; Fig. 4d). Varicose
fibers coursed alongside GnRH neurons, forming multiple contacts along the
entire length of the cell body (Fig. 4a). Often, BDA-ir fibers apposed GnRH

neurons at the base of the processes (Fig. 4b). As was observed at the light

92

 

 

Subject MS/DBB OVLT POA/SCN
ipsi contra ipsi contra ipsi contra
O O O O O O
SCN 74 18.9 /o 31.6 /o 65.6 /o 58.1 A: 51.3 /o 44.2 /o
81 17.2% 17.9% 54.5% 53.1% 84.8% 57.9%
34 26.8% 24.4% 27.3% 26.7% 36.4% 1 7.3%
LSPV 40 4.2% 16.7% 50.0% 32.0% 69.0% 21.4%
61 26.7% 25.0% 43.8% 37.5% 48.4% 34.8%
75 22.0% 20.7% 73.5% 27.9% 79.5% 24.5%

 

Table 4.1: Percentages of GnRH-ir neurons that formed appositions with
BDA-labeled fibers. Values represent the percentage of the total number of
GnRH neurons ipsilateral or contralateral to the BDA injection site with
appositions in each of three rostrocaudal regions.

93

 

Figure 4.3: Photomicrographs of appositions between BDA-ir
fibers (black) and GnRH-ir neurons (brown; A- C) and processes
(D) following injections centered in the LSPV. GnRH neuron in
(B) is the same neuron depicted in bottom center of (A). All
images were taken using a 100x oil objective. Scale bar = 20 um.

level, the density of appositions was low in the MS/DBB, highest in and around

the OVLT, and decreased in frequency in the POA/SCN.

Discussion

Rhythms in activation of GnRH neurons associated with the LH surge
require axonal outputs from the SCN (Meyer-Bernstein et al., 1999; de la Iglesia
et al., 2003) and are reversed in diurnal and nocturnal species (McElhinny et al.,
1999; Mahoney et al.; Mahoney and Smale, 2005a). However, the output
pathways through which the SCN regulates these functions are not well-
understood in diurnal species. In the present study, we asked whether the SCN
and the LSPV might project to GnRH neurons. BDA-labeled fibers originating in
the SCN appeared to contact GnRH neurons, as they do in nocturnal rodents.
The appositions were most concentrated in and around the OVLT and the
adjacent preoptic area; fewer appositions were present in the MS, rostral DBB
and dorsal to the SON. BDA-labeled fibers coming from the LSPV also formed
numerous appositions with GnRH neurons in the same areas, particularly in the
OVLT and the rostral POA, where in several cases both SCN and LSPV fibers
apposed more than 50% of the GnRH neurons. It is thus likely that some, and
perhaps many, GnRH neurons receive input directly from both the SCN and the
LSPV.

The appositions seen here with a light level microscope are highly
suggestive of synaptic interaction between the SCN/LSPV efferents and GnRH
neurons, but do not establish the presence of functional synapses. However, in

nocturnal rodents, light-level techniques similar to those used here have revealed

95

contacts between SCN fibers and GnRH neurons and ER-positive neurons that
were subsequently verified using electron microscopy (Watson et al., 1995; van
der Beek et al., 1997b). Additionally, confocal analysis in the current study
revealed appositions between varicose LSPV fibers and GnRH neurons in optical
sections of less than 0.5 pm thickness, indicating that the labeled fibers were
very likely to interact functionally with these cells. Taken together, these data
provide strong evidence that the contacts observed here represent synapses
between BDA-labeled fibers and GnRH neurons, as they do in lab rats.

The distribution of efferent fibers from the grass rat SCN is very similar to
that seen in nocturnal lab rats and hamsters (Watts et al., 1987; Kriegsfeld et al.,
2004) and includes prominent projections to the septum, diagonal band, OVLT
and preoptic area. In the present study, labeled fibers from both the SCN and
the LSPV formed appositions with GnRH neurons in all of these areas. However,
the appositions with SCN fibers were bilaterally distributed in grass rats, whereas
those with LSPV fibers were most heavily concentrated on the side ipsilateral to
the BDA injection site. Although SCN input to contralateral GnRH neurons has
been documented in lab rats, the degree of laterality has not yet been quantified
in that species (Van der Beek et al., 1997a). At a more general level, SCN
projections to the contralateral side of the brain have been reported in both lab
rats and hamsters, though they are less pronounced than ipsilateral projections
(Watts et al., 1987; Kriegsfeld et al., 2004). It is possible that differences in
tracers used (Phaseolus vs. BDA in the present study), and/or the sizes and

locations of injections, account for differences in the extent of

96

Figure 4.4: Confocal laser fluorescent images of appositions
between BDA-ir fibers (green) and GnRH-ir cell bodies (red)
following injections centered in the LSPV. A — C: projections of
serial optical sections. D: single optical section from the cell

depicted in (C). All images were taken using a 63x oil objective.
Scale bar = 20 pm.

 

contralateral GnRH innervation between grass rats and lab rats. Alternatively,
the relative degree to which the SCN projects to the contralateral population of
GnRH neurons may differ in these species.

In nocturnal species, several lines of evidence suggest that at least some
of the projection from the SCN to the GnRH neurons is composed of VIP fibers.
Specifically, in lab rats VlP-ir fibers appose GnRH neurons (van der Beek et al.,
1994; Kriegsfeld et al., 2002), GnRH neurons in the OVLT and POA express the
VIP2 receptor (Smith et al., 2000) and VIP appositions decrease substantially
following unilateral SCN lesions (van der Beek et al., 1993). In grass rats, VP-
and VIP-ir fibers appear to contact GnRH neurons, and Mahoney and Smale
(2005) suggested that at least some of these contacts represent direct input from
the SCN. The present work provides further evidence that this may be the case,
by showing that BDA-labeled fibers from the SCN form close appositions with
GnRH neurons.

The SCN may also indirectly influence GnRH cell activity and the surge
via its projections to the AVPV, a nucleus essential for the generation of an LH
surge (Wiegand and Terasawa, 1982). In nocturnal species, SCN cells project to
ER-positive neurons in the AVPV (de la lglesia et al., 1995; Watson et al., 1995),
and ER-ir cells in that nucleus project to regions containing GnRH neurons
(Simonian et al., 1999). In female grass rats, appositions between VP- and VIP-
ir fibers and ER-positive cells have also been seen (Mahoney and Smale,
2005b). The current data support the suggestion that these VIP- and VP-ir fibers

could originate in the SCN. It thus appears likely that in diurnal rodents, as in

98

nocturnal ones, the SCN projects to GnRH neurons both directly and indirectly
via the AVPV, and that these pathways are likely to include VIP- and VP-ir fibers.

The temporal patterns of VIP and VP signals to GnRH neurons from the
SCN are likely to be the same in lab rats and grass rats, as rhythms in VP and
VIP mRNA are similar (Krajnak et al., 1998; Mahoney et al., 2006). It is thus
likely that GnRH neurons respond differently to these SCN signals in diurnal and
nocturnal rodents. This may also be the case for rhythms in behavioral
sensitivity to steroid hormones, which are also inverted in grass rats relative to
lab rats (Mahoney et al., 2004). These time-dependent differences in sensitivity
could reflect intrinsic properties of the GnRH neurons, or they could arise from
rhythms in extra-SCN inputs to them. The present data suggest that the LSPV
could represent one such input.

Appositions between GnRH cells and fibers emanating from the LSPV, as
with those coming from the SCN, were most concentrated around the OVLT and
adjacent POA. In lab rats, GnRH neurons in this area are preferentially activated
during the LH surge (Lee et al., 1990; Lee et al., 1992; van der Beek et al.,
1994), and the numbers of neurons expressing GnRH mRNA in this region, but
not in more rostral or caudal areas, are increased during proestrus (Porkka-
Heiskanen et al., 1994). The OVLT and rostral POA may therefore contain a
high concentration of GnRH cells that are specialized for regulation of events
occurring at estrus (Hiatt et al., 1992). The current observation of a relatively

high concentration of appositions between LSPV efferent fibers and GnRH

99

neurons in this region suggest that the LSPV may also act on these cells to
influence estrus-related events such as the LH surge and sex behavior.

In summary, the present data show that in grass rats, fibers emanating
from the SCN and from the LSPV, a direct target of the SCN, form appositions
with a group of GnRH cells concentrated in a region critical for the generation of
the LH surge. The hypothesis that LSPV inputs influence the same GnRH
neurons that receive input from the SCN is suggested by the extensive overlap in
the distributions of GnRH neurons forming appositions with SCN and LSPV
efferent fibers, and by the fact that in some areas over 50% of GnRH neurons
receive input from the LSPV and SCN, respectively. Inputs from these areas
differ with respect to their neurochemical signals, as LSPV neurons express
neither VIP nor VP (Watts and Swanson, 1987; Smale and Boverhof, 1999).
They are also unlikely to be the same in terms of their temporal pattern of
signaling, as the grass rat SCN and LSPV exhibit different endogenous rhythms
in cFos expression (Schwartz et al., 2004). The LSPV is thus in an ideal position
to influence the circadian regulation of GnRH neurons, and to do so differently in
diurnal and nocturnal species. The data also provide some support for the more
general hypothesis that convergence of LSPV projections on SCN target neurons

contributes to differences in circadian function of nocturnal and diurnal species.

100

CHAPTER 5
The effects of chemical lesions of the lower subparaventricular

zone on general activity rhythms in the Nile grass rat

Introduction

Diurnal and nocturnal rodents exhibit dramatic differences in circadian
rest-activity rhythms. However, the hypothalamic suprachiasmatic nucleus
(SCN), which contains the primary circadian pacemaker in mammals (Moore and
Eichler, 1972; Stephan and Zucker, 1972; Sato and Kawamura, 1984a; Ralph et
al., 1990), is very similar in diurnal and nocturnal species. Rhythms in expression
of the clock genes per1 and per2 peak during the subjective day in diurnal 13-
lined ground squirrels (Mrosovsky et al., 2001), grass rats (Lambert et al., 2005);
Arvicanthis ansorgei: (Caldelas et al., 2003) and in nocturnal laboratory rats
(Rattus norvegicus, hereafter referred to as ‘lab rats’; Yan et al., 1999), hamsters
(Hamada et al., 2004) and mice (Hastings et al., 1999). Furthermore, rhythms in
mRNA for the SCN output signals vasopressin (VP), vasoactive intestinal
polypeptide (VIP) and prokineticin 2, do not differ in diurnal and nocturnal species
(Krajnak et al., 1998; Lambert et al., 2005; Mahoney et al., 2006), suggesting that
the phase of both the molecular clock in the SCN and several of its output signals
are similar in diurnal and nocturnal rodents. Although it is still possible that SCN
function differs in some important ways in diurnal and nocturnal species (Krajnak
et al., 1997; Jiao et al., 1999), it is most likely that the fundamental difference in
their activity rhythms is determined by mechanisms somewhere downstream of

the SCN.

101

In nocturnal species, the SCN sends a major projection to the
hypothalamic subparaventricular zone (sPVZ), which extends dorsally and
caudally from the border of the SCN to the paraventricular nucleus of the
hypothalamus (Watts and Swanson, 1987; Abrahamson and Moore, 2001). The
sPVZ projects, in turn, to many of the same targets as the SCN (Watts et al.,
1987; Morin et al., 1994), leading to speculation that it relays or amplifies signals
emitted by the SCN (Watts, 1991; Saper et al., 2005). This hypothesis is
supported by data showing that free-running circadian wheel-running rhythms in
lab rats are attenuated by axon-sparing chemical lesions in the ventral sPVZ (Lu
et al., 2001) and are abolished by electrolytic lesions of the sPVZ (Abrahamson
and Moore, 2006). These data suggest that both cells in the sPVZ as well as
fibers passing through it are important for maintaining locomotor rhythms.
However, direct axonal projections from the SCN to the sPVZ are not necessary
for maintenance of wheel running rhythms in lab rats, as these persist even after
horizontal knife cuts have severed axonal connections between the SCN and the
sPVZ (Brown and Nunez, 1986; Watts et al., 1989). Furthermore, in nocturnal
hamsters wheel running rhythms may be maintained by a diffusible signal
secreted by the SCN, as they persist following isolation of the SCN (Hakim et al.,
1991). The sPVZ has been suggested as one possible target for such a signal
(Abrahamson and Moore, 2006). Normal circadian regulation of locomotor
activity in these species is thus likely to rely on multiple SCN outputs, axonal
and/or diffusible (Silver et al., 1996a), at least one of which is likely to target the

sPVZ.

102

In diurnal Nile grass rats, the lower part of the subparaventricular zone
(LSPV) projects to nearly all of the same targets as the SCN (chapter 3),
suggesting that circadian signals may be broadcast to other brain regions both
directly from the SCN, and indirectly via the LSPV. The LSPV also displays a
rhythm in cFos expression whose phase differs substantially from that of rhythms
in the LSPV of lab rats, and which persists in constant darkness in grass rats but
not in lab rats (Schwartz et al., 2004). These data suggest first, that in diurnal
rodents, circadian signals emitted by the LSPV and SCN could differ
substantially, and second, that LSPV signals could contribute to differences in
the phases of behavioral and physiological rhythms of diurnal and nocturnal
rodents. If this is the case, then a selective lesion of the LSPV, sparing both the
SCN and the fibers that project from it, should significantly affect rhythms that are
reversed in grass rats nocturnal lab rats.

The study described below examined the effects of bilateral destruction of
the LSPV on general activity rhythms in grass rats. We used microinjections of
the axon-sparing excitotoxin N-methyl-DL-aspartic acid (NMA; Stewart et al.,
1986; NMA; Sisk et al., 1988) to destroy cell bodies in the LSPV. Because the
SCN is resistant to this neurotoxin (Peterson and Moore, 1980; Hastings et al.,

1985) its cell bodies and efferent fibers remained intact.

Experimental Procedures

Animals:

Animals were 29 adult male grass rats obtained from a breeding colony at

Michigan State University (Katona and Smale, 1997). All animals were singly

103

housed in Plexiglas cages (34 x 28 x 17 cm) and were provided access to food
(PMI Nutrition Prolab RMH 2000, Brentwood, MO) and water ad Iibitum. Animals
were kept in a 12:12 light: dark cycle (LD; ~250 lux during the light phase; lights
on at 06:00 h) with a red light (<5 qu) on constantly for animal care purposes. All
experiments were performed in compliance with guidelines established by the
Michigan State University All-University Committee on Animals Use and Care,
and the NIH Guide for the Care and Use of Laboratory Animals. One preliminary
group of animals (n = 5) was used to fully characterize the effects of lesions on
the SCN and LSPV. A second, primary, group of animals (n = 21) was used to

examine effects of the lesions on behavioral rhythms.

Surgical procedures:

All animals received a bilateral injection of 0.2 M NMA (Sigma, St. Louis
MO) or control vehicle directed at the LSPV. Animals were deeply anesthetized
with sodium pentobarbital (50mg/kg i.p.) prior to surgery, and then the head was
shaved, wiped with betadine, and injected with lidocaine (0.03cc s.c.). The
animal was then placed into a stereotaxic apparatus (Stoelting Co.; Wood Dale,
IL) with the incisor bar set to -5.5 mm and the manipulator set vertically, and a 1-
2 cm incision was made in the scalp to expose the skull. A small hole was drilled
in the skull, and dura was gently broken to allow uninterrupted passage of the
micropipette. All injections were made using a 10 ,uL Hamilton syringe (Hamilton;
Reno, NV) in which the needle was cut to 2 cm in length. The syringe was filled
with either NMA in 0.2 M saline or saline alone and lowered slowly to injection

depth. Injection coordinates were: 1.4 mm anterior to bregma, 1 0.3 mm lateral

104

to bregma, and 6.3 mm ventral to dura. Animals were injected with 90 pL (n =
13) or 135 pL (n = 11) per side using a manual stereotaxic microinjector
(Stoelting); this allowed the NMA to be delivered in small, precisely-measured
volumes over the course of the injection. The incision was closed with autoclips,

and animals were given 0.1 cc buprenorphine i.m. and 2.0 cc sterile saline s.c.

Assessment of lesions:

In order to characterize the full effects of the lesions we first examined
three series of tissue taken from brains of five male grass rats that were lesioned,
kept for at least three weeks in DD, and then perfused at CT 22, when cFos
expression is high in the LSPV and low in the SCN (Schwartz et al., 2004). One
series of tissue from these animals was thionin-stained, while the second and
third were processed for cFos and VP-immunoreactivity, respectively. To
characterize the location and spread of the lesion, thionin-stained tissue was
examined for cell loss and gliosis. cFos expression was used to determine
whether lesions destroyed the cells that are endogenously rhythmic in the grass
rat LSPV. VP expression was used to verify that both VP-immunoreactive (-ir)
dorsal SCN cells and their efferent fibers in the LSPV were spared by the lesions.
Animals in the primary, behavioral, experiment were perfused at the end of the
study between ZT 1 and ZT 2, when cFos expression is high in both the SCN
and the LSPV of intact animals (Schwartz et al., 2004). This was used to assess
both how completely the cFos-producing cells in the LSPV were destroyed and

whether activity in the SCN was spared. One series of tissue from these

105

experimental animals was therefore processed for cFos expression while another
was thionin-stained.

All animals were given an overdose of sodium pentobarbital and perfused
transcardially with 0.01M phosphate-buffered saline (PBS), pH 7.2, followed by
4% paraformaldehyde (Sigma, St. Louis, MO) with lysine and sodium periodate
in 0.1M phosphate buffer (PLP). Brains were post-fixed for 4 hours in PLP and
then transferred to 20% sucrose solution overnight. Brains were sectioned
coronally at 30 pm on a freezing microtome into 3 series of sections, which were
then stored in cryoprotectant at -20°C until processing began. Unless noted
othenivise, incubations were conducted at room temperature for 1 hour on a
shaker, and tissue was rinsed 3 x 10 minutes in PBS between incubations. Free-
floating sections were rinsed in PBS, and then incubated in (i) 5% normal serum
in PBS with 0.3% Triton X-100, followed by (ii) primary antibody on a rotator for
48 hours at 4°C. Following primary incubation, tissue was incubated in (iii)
biotinylated secondary antibody, followed by avidin-biotin peroxidase complex
(ABC Vectastain Kit, Vector Laboratories, Burlingame, CA). Protein was
visualized by reacting with diaminobenzidine (DAB; 0.5mg/mL; Sigma) in Trizma
buffer (Sigma), catalyzed with hydrogen peroxide. For cFos immunostaining,
nickel sulfate was added to the cFos reaction, yielding a blue-black rather than
brown reaction product. Reagents for each reaction were as follows: cFos: (i)
normal goat serum (Vector), (ii) rabbit anti-cFos (Santa Cruz Biochemistry, Santa
Cruz, CA; 125,000), (iii) biotinylated goat anti-rabbit (Vector; 1:200). VP: (i)

normal goat serum (Vector), (ii) guinea pig anti-VP (Peninsula Laboratories, San

106

Carlos, CA; 1:40.000); (iii) biotinylated goat anti-guinea pig (Vector; 1:200).
Previous blocking experiments in which the primary antibodies were pre-
incubated with their antigens eliminated all staining for both VP (Smale and
Boverhof, 1999) and cFos (Schwartz et al., 2004). Tissue was mounted,
dehydrated, and coverslipped with Permount.

To quantify the extent to which lesions impacted the LSPV region of
special interest, two sections through the mid- to caudal SCN were selected from
each animal for counts of cFos-ir cells. The boundaries of the SCN were first
demarcated in thionin-stained sections, then superimposed onto adjacent cFos-
stained sections. All cFos-positive neurons were counted within the SCN, and
within a region of the LSPV defined by a 215 ,um x 160 ,um rectangle placed
immediately dorsal to the SCN and lateral to the third ventricle. This method
reliably samples a portion of the distribution of cFos-ir cells that is both clearly
outside the SCN, and in which cFos expression is rhythmic in grass rats in both
L0 and DD (Schwartz et al., 2004). Bilateral counts were summed and

expressed as the total number of cFos-positive cells in the SCN and LSPV.

Experimental Design:

All animals were housed in a 12:12 LD cycle (lights-on = 06:00 h) with
infrared motion detectors placed directly over the cage to monitor general
activity. Animals were stably entrained to the LD cycle for at least 10 days prior
to the start of the experiment. General activity data were collected on a PC
running the Vitalview program (Minimitter; Bend, OR). To ensure that activity

data used for analysis represented stable, long-term patterns of rhythmicity,

107

 

‘ .. _,._,_,.r_'_~ . ': ..‘.. . .
{alum-M-firéfii‘ar-fixrm: .

"..o

 

Figure 5.1: Photomicrographs of the SCN of a grass rat with a com-
plete LSPV lesion that was perfused at CT 22. Alternate sections
are stained for (A) Nissl, (B) cFos and (C) AVP. Scale bar = 200
pm.

108

animals were monitored for at least three weeks in each phase and a block
representing approximately the final two weeks of each stage was selected for
analysis (see below). In the preoperative phase (PREOP), baseline data were
collected in L0 for at least 21 days, of which the final 14 days prior to surgery
were selected for analysis. Following PREOP, animals received intracranial
injections of either NMA (n = 19) or vehicle (n = 5). Animals were kept in the
same LD cycle following surgery for at least four weeks. General activity was
recorded from them for at least the last three weeks of this period (LD1) and data
from the final 12 days were selected for analysis. Animals were then released
into constant darkness (DD) for four weeks, of which the final 14 days were
selected for analysis. At the end of DD, 12:12 LD was resumed with the same
phase as at the start of the study (lights-on at 06:00), and activity was recorded
for another three weeks before animals were perfused (LD2). The sampling
period analyzed for each individual in L02 began on the day that rhythms in both
the onset and the offset of activity were re-entrained, and ended on the day
before sacrifice (mean number of days in sampling period for all animals = 14.67

i 1.03 days).

Data analysis:

To assess rhythms in each stage, period (1:) and amplitude were
calculated for each animal via chi-squared periodogram analysis in the Clocklab
program (Actimetrics; Evanston, IL). Rhythmic patterns within the circadian
range were considered significant at the p = 0.05 significance level. Total activity

counts were summed for each stage and the percentage of activity in each day

109

occurring during the light phase was averaged for PREOP, LD1 and L02.
Additionally, daily onsets and offsets for the major daily activity bout were plotted
in Clocklab. From these, the number of days to re-establish a stable phase angle
with the light-dark cycle following the start of LD2 was determined. The day of
entrainment was defined as the day on which the times of activity onset and
offset no longer shifted more than 30 minutes for at least 3 consecutive days.
Finally, we calculated the variance in the time of daily activity onset and offset for
each animal at each stage of the experiment.

Six animals presented either unilateral lesions or no evidence of damage,
and were not included in the ANOVA analyses. The remaining animals were
classified as SHAM, LSPV-Iesioned (LSPVx) or (MISS) based on 1) lesion
placement determined from thionin-stained sections, and 2) the extent of LSPV
damage in the lesion; these groups are described in detail in the results section.
Separate 3 x 4 repeated-measures ANOVAs compared the influences of
treatment group (LSPVx, MISS, SHAM) and stage (PREOP, LD1, DD, LD2) on
the amplitude of activity rhythms, total activity counts, and variance in daily onset
and offset. A 3 x 3 repeated-measures ANOVA compared the influences of
treatment group (LSPVx, MISS, SHAM) and stage (PREOP, LD1, LDZ) on the
percentage of activity that occurred during the light phase of the 12:12 LD cycles.
Lastly, one-way ANOVAs compared the effects of treatment on r in DD, the
number of days to re-entrainment of activity onset and offset following DD, and
the number of cFos-ir cells in the SCN and the LSPV. All effects were

considered significant at the p = 0.05 level. Where significant effects were

110

.6: 8n u .3 28m 8% 6 .9 ecu .32 a .s 3.. 8%....
2m 2288 5522 .N E -F K a 83.8 a. $5 a .9 x>n_m._ Eu 6 .s
3.5 25:898. a e >me .25 zom 2.. co 285558;“. nu.» 2:2".

 

111

detected, Fisher’s posthoc tests were conducted to evaluate painNise
comparisons.

To assess relationships between the extent of cell loss in the LSPV and
general activity rhythms, data from all animals (n = 21) were pooled. The number
of cFos-ir cells in the SCN and LSPV were then correlated with the total activity
counts and the amplitude of activity rhythms for each postoperative stage (LD1,
DD, LDZ). In addition, the number of cFos-ir cells in the SCN and LSPV was
correlated with (1) the percent of activity that occurred during the light phase of
the 12:12 LD cycles (LD1 and LD2), (2) “C in DD, and (3) the number of days to
re-entrain activity onset and offset following DD. For analyses of these

correlations, Pearson’s rvalues were considered significant at the p = .05 level.

Resufls:

Histology:

In the five LSPV-lesioned animals perfused at CT 22 to characterize the
NMA-induced damage, thionin-stained tissue revealed substantial gliosis and a
near complete absence of neurons dorsal and, to a lesser extent, dorsolateral to
the SCN (Fig. 5.5.1a). The area outside the Nissl-defined SCN border was
almost completely devoid of cFos-positive cells (Fig. 5.5.1 b). Very little cFos
expression was observed within the SCN of these animals, as would be expected
for grass rats perfused at CT 22. VP-ir cells formed a semicircular pattern with
the most cells in the dorsal SCN, and some along the medial and lateral border

of the SCN, as has been previously described in intact grass rats (Fig. 5.5.10;

112

Smale and Boverhof, 1999; Fig. 5.5.10; Schwartz et al., 2004). VP-ir fibers-
streamed dorsally from the SCN through the LSPV and sPVZ.

In SHAM animals in the primary behavioral study, the SCN and LSPV
appeared as they have been described previously (Smale and Boverhof, 1999;
Schwartz et al., 2004). In thionin-stained sections, cells in the SCN were small
and tightly-packed, whereas cells in the area immediately dorsal to it were
substantially larger and more dispersed (Fig. 5.5.2a). While a glial scar
identifying the needle track was visible in several animals, there was no evidence
of cellular trauma beyond the track mark, and neurons surrounding the glial scar
were plentiful and healthy in appearance. The distributions of cFos expression in
the SCN and LSPV of control animals perfused at ZT 1 and ZT 2 were also
consistent with those observed previously in intact grass rats at ZT 1 (Fig. 5.5.20;
Nunez et al., 1999; Schwartz et al., 2004). cFos-ir cells filled the dorsal and
ventral SCN and were relatively scarce in the central SCN. Dorsal and
dorsolateral to the mid- to caudal portions of the SCN, cFos-ir cells extended
from the periventricular nucleus to the anterior hypothalamic area. These cFos-ir
cells were clearly outside the Nissl-defined border of the SCN, and were not
observed directly lateral to the SCN.

The five lesioned animals with the fewest cFos-ir cells in the LSPV were
classified as LSPVx. In these animals, lesions were centered in the LSPV and
sPVZ (Fig. 5.3a) and left the SCN and the PVN intact (Fig. 5.2b). cFos
immunoreactivity reflected the Nissl findings in LSPVx animals. That is, cFos-ir

cells were plentiful within but not outside of the SCN (Fig. 5.2d). The numbers of

113

A

211., .-

 

     

T7777

 

 

 

 

 

 

 

Figure 5.3: Schematic diagrams illustrating lesion placement in
animals classified as LSPVx (A) and MISS (B). Atlas adapted from
Paxinos & Watson (2005).

114

 

 

 

 

 

 

 

 

Figure 5.3 (continued).

115

ESE—5.0 u + .0005 .6 “out“... EeE 2.35:0? u . 0.95:0 5.3.0 new 00.5. 0900.. .6.— 0020 some 5 “were 2.20 ,

.0320 he 80:0 EeE

c. mocmtg new .095 2.00 :. monetg .oEEEu 05 c. 33:00 .6 «02:022. .3328 “_o 0.05. .39.. 3.0 030...

 

 

 

A25 35...; .300

 

 

A35 005...? .300

 

530% «Egg 2.0200

 

35.8. 3.3.8 .22.

 

.00 v *0 00 .0 3.0 00. F .00 A000 o\o0.0 o\o0.0 06000 F.0000 v. 00v mm H
00.0 000 0.00.0 00.0 v0.0 0 ...0 000.00 nx0000 o\o0.00 044000? 0.0009 0.0 0: 00..
00.0 00.»V 0V0 0rd 0fi0 000 0E Q: 0E #0000 0.020 0.000 ow H
0.00 .0 .P “N .0 0.00.0“ 00.0 00.0 00.0 m? 0E 0E 060000 0.0003 0.0000 F on
vvd v04 00:0 0.0 00. F 3.”...N o\00.0 o00.0 o\o0.v 0.000v 0.0000 0. 00F on H
V0.0 «0.0 0.00.05 0V0 00.0 v0.0 $0. ..0 $0.00 o\o v.00 0.0003. 0.0000 _. 0.0 .03 PG.—
00.0 00.0 00.0 00.0 00.0 00.0 o\ov.e o\00.v n$0.0 N .0000 0.0000 010000. on H
0.0 00.0 000.0 00.0 00.0 0.0 o\o0. _.0 000.00 o\o0.00 .000NON .0.»0VON .000va mmn.
52:0 00:2 >00.— E<I0 00.5. >m0._ 2<I0 00.5. >00.— s_<_._0 00:2 >n_0.._

 

 

116

cFos-ir cells in the LSPV of these animals ranged from 15.9 % to 56.8 % of those
obtained from controls.

Among the 6 animals classified as MISS, most lesions were centered in
the mid- to dorsal sPVZ, and either missed or only slightly damaged the LSPV
(Fig. 5.3b). In one of these animals, the lesion destroyed a large portion of the
mPOA, and in another case the lesion was centered in the rostral and mid-PVN.
The smallest bilateral lesion in this group covered approximately half of the
sPVZ, whereas the largest encompassed most of the sPVZ, as well as the
anterior and mid-PVN. In some MISS individuals, as with some LSPVx animals,
the shape of the SCN was slightly distorted and the 3v was distended. cFos-ir
cells in the LSPV of MISS animals were between 70.2 % and 121.8 % of average
control values. The number of cFos-positive cells in the LSPV was significantly
lower in LSPVx animals than either MISS or SHAM animals, which were similar
to each other (F212 = 16.501, p < 0.001). The surgeries did not affect the number

of cFos-positive cells in the SCN (F212 = 1.560, p = 0.250).

Activity rhythms
All animals had robust daily activity rhythms prior to surgery (Fig. 5.4).

Several individuals exhibited crepuscular tendencies, and one of these was
slightly more active in the dark phase than in the light because of a single large
activity peak centered around lights-off; (in this individual 46.9 % of total daily
activity occurred in the light phase; Fig. 5.50). In all other animals activity was

concentrated in the light phase (Table 1).

117

Fm fl J’s: n'.‘n. ‘1» -'@K$3£5$P
.39,“.,. . 01% .j.

.5

- Q“ . . .~‘ .' .
“$309.11 .43“ - -. -" __.'_+ .EEEEL--,-
12 12 18

 

Figure 5.4: Representative double-plotted actograms of a SHAM (A)
and MISS (B) grass rats. LD cycles are indicated at top, time of day
(ZT) at bottom. In each actogram, date of lesion surgery (NMA or
SHAM) and changes in lighting conditions (LD, DD) are indicated
along the left side of each actogram.

118

Figure 5.5: Representative double-plotted actograms of three LSPVx animals, #
234 (A), # 241 (B) and # 254 (C). LD cycles are indicated at top, time of day (ZT)
at bottom. In each actogram, date of lesion surgery (NMA or SHAM) and
changes in lighting conditions (LD, DD) are indicated along the left side of each

actogram.

119

 

 

. . ...r.- .12; ' . ' .- ..
.1 ‘9 ‘ f
a. . . .
iii-$5: .9159}?! ‘ ‘

' fi.0§é:.-N.1.

i
. J

m":-

 

 

B .2 . it:

3:
145.? ~

..a ..r- ,.
h.“ .— J

 

Figure 5.5.

120

Injections of both NMA and of vehicle were associated with short-term and
long-term nonspecific behavioral effects. The short-term effect was a brief period
of arrhythmicity in the first several days after surgery. Animals were hyperactive
and in some cases displayed an ultradian component to their activity pattern that
persisted for several days during this phase (Fig. 5.5). In all cases, these effects
subsided within seven days; we did not include behavior from this first post-
surgical week in our analyses.

There was also a long-term nonspecific effect of surgery on total activity.
Specifically, total activity counts were reduced in all three postoperative stages
compared to the PREOP period (F335 = 6.938, p < 0.001; Table 1), but were not
affected by treatment (F312 = 0.353, p = 0.709), and there was no interaction
between stage and group (F335 = 0.593, p = 0.734). There was also no
correlation between the total amount of activity and cFos expression in the LSPV
or the SCN (Table 2). Thus, total activity counts decreased similarly following
surgery in the three treatment groups at all stages of the experiment, and were
unrelated to cell loss in the LSPV or SCN. Effects of treatment on activity
rhythms are therefore not likely to have been caused by a decrease in overall
activity levels.

Rhythms of all animals were entrained in LD and free-ran in DD. Free-
running periods (I) were less than 24 hours in all but one animal (1 = 24.16 h),
and did not differ among groups (F212 = 1.000, p = 0.454; Table 1). However,
activity rhythms became noticeably less precise in several individuals after they

had been lesioned (Fig. 5.5). Amplitude was not affected by group (F212 = 1.778,

121

p = 0.211), but there was an interaction between group and stage (F636 = 2.900,
p = 0.021). Specifically, activity rhythms in LSPVx animals in LD1, DD and LD2
were of lower amplitude than they were during the PREOP stage, whereas in
MISS animals, only the amplitude in DD was lower than in the PREOP period
(Fig. 5.6). In SHAM animals, by contrast, amplitude was higher in LD2 than in
any other condition. Amplitude was positively correlated with cFos expression in
the LSPV in DD, but not at any other stage, and amplitude was not related to
cFos expression in the SCN (Table 2). Thus, although all animals were rhythmic
following surgery, LSPVx animals had lower-amplitude activity rhythms at all
postoperative stages, and cFos-ir cell loss in the LSPV was more closely related
to the amplitude of free-running activity rhythms than entrained ones.

Several lesioned animals exhibited disorganized activity following surgery,
particularly during and following the period in DD. Some of these were nearly
arrhythmic (Fig. 5.5b), whereas others had rhythms that were disrupted by
arrhythmic episodes that lasted up to five days (Fig. 5.5a, c). The variability of
daily activity offsets was affected by group (Fan = 7.43, p = 0.008) and stage
(F335 = 2.894, p = 0.048), with no interaction between these variables (F636 =
1.579, p = 0.182). Specifically, offset variance was higher in LSPVx animals than
in MISS and SHAM animals, and was higher in DD compared to PREOP
conditions (Table 1). Although the variance in daily activity onset appeared
slightly elevated in LSPVx animals (Table 1), it was not significantly affected by
treatment group (F212 = 3.121, p = 0.081 ), stage (F336 = 1.311, p = 0.286) nor the

interaction between these variables (F635 = 1.17, p = 0.340). Offset variance was

122

 

total activity counts

amplitude

Percent daytime activity

 

 

 

 

 

 

SCN LSPV SCN LSPV SCN LSPV
PRE 0.0131 -0.034 -0.1502 -0.0306 -0.0652 0.0546
LDt 0.2091 0.0153 -0.046 0.1218 0.1192 0.3179
DD 011% 0.076 0.1657 0.5185” n/a n/a
LD2 -0.3116 -0.017 -0.0737 0.3683 0.1217 0.4676'

 

Table 5.2: Correlations between the number of cFos-immunoreactive cells
in the SCN and LSPV and total levels of activity, amplitude of activity

rhythms, and percentage of activity in the daytime in each stage.

significant correlation.

123

*—

 

thus higher in LSPVx animals compared to MISS and SHAM animals, and onset
variance exhibited a nonsignificant trend in the same direction.

Following the transition from DD to LD all control animals re-entrained
within 5-9 days. Activity rhythms also re-entrained in all NMA-injected animals,
but the rates at which this occurred were highly variable, ranging from 5 days to
21 days, and the animal that took the longest to re-entrain was the one
sustaining the most complete lesion (Fig. 5.5a). Animals that took the longest
time to re-entrain often displayed sudden phase jumps, or prolonged arrhythmic
bouts as seen in DD (Fig. 5.5a, c). LSPVx animals exhibited somewhat slower
re-entrainment rates than either MISS or SHAM animals, but these differences
did not reach statistical significance (onset, F212 = 3.427, p = 0.066; offset, F312 =
2.846, p = 0.097; Fig. 5.7). However, the number of days to re-entrain onsets as
well as offsets was negatively correlated with the number of cFos-positive cells in
the LSPV (onset, r: -0.457, p = 0.037; offset, r = -0.557, p = 0.009), but not in
the SCN (onset, r: -0.215, p = 0.350; offset, r = -0.227, p = 0.321). The
distribution of activity following surgery was also highly variable between, and in
some cases within, animals. Postoperative changes in activity distributions
included increased crepuscular and nocturnal tendencies; this can be seen most
strikingly in # 254, in which daytime activity represented only 30.1 % of total daily
activity in LD2 (Fig. 5.50). However, there were no effects of treatment group
(F212 = 2.649, p = 0.111) or stage (F224 = 1.112, p = 0.345) on the percentage of
activity that occurred during the daytime, and there was no interaction between

these variables (F42 = 1.271, p = 0.309; Table 1). The percentage of activity

124

2000 ,
3W ‘
E 1800 'i
.C
y‘S,1600 ",
i'E
‘>.1400

' 'i Léevf

x.
3(-

V

—i
N
O
O

_l
030
CO
CO

Amplitude of acti it
N A O)
O O O
0 0 0

 

O

 

 

Fig.ure 5.6: Amplitude of activity rhythms of LSPVx (black), MISS
(gray) and SHAM (white) grass rats in each stage. * = significantly
different from PREOP within group. *" = significantly different from
all other stages within group.

125

occurring in the daytime was, however, positively correlated with cFos
expression in the LSPV in LD2, but not LD1, and was not related to cFos
expression in the SCN (Table 2). In other words, animals with fewer Fos-ir cells
in the LSPV were proportionally more active at night after they had been in DD

for several weeks, but not prior to that.

Discussion:

The LSPV is a major target of the SCN, and in turn projects to many of the
same areas as the SCN, raising the possibility that signals emanating from these
two regions interact to regulate circadian rhythms. The current data provide
support for that hypothesis by showing that excitotoxic lesions that destroyed
cells in the LSPV but left the SCN intact decreased the amplitude of activity
rhythms in LD and DD, impaired re-entrainment to an LD cycle, and shifted the
distribution of daily activity towards the nighttime after rhythms had re-entrained.
These changes were unrelated to cFos expression in the SCN, and were absent
in animals with lesions that spared the LSPV. Cells in the grass rat LSPV thus
appear to be an important part of the output pathways used by the pacemaker

within the SCN to communicate circadian information to other brain regions.

Histology:

The SCN of NMA-injected animals appeared normal in thionin-stained
sections, suggesting that it is resistant to chemical lesion in grass rats as is the
case in lab rats (Peterson and Moore, 1980; Hastings et al., 1985). VP-ir
neurons in the dorsal SCN and fibers passing through the LSPV were spared,

further indicating that the SCN was intact in animals receiving NMA injections.

126

 

 

I LSPVx
III MISS
El SHAM

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Onset Offset

 

 

 

Figure 5.7: Number of days to re-entrain daily activity onset (left) and
offset (right) for LSPVx (black), MISS (gray) and SHAM (white) grass
rats following free-run.

127

Furthermore, cFos-ir staining in the SCN of lesioned animals perfused at ZT 1-2
was similar to that of SHAM animals and to that seen previously in intact grass
rats (Nunez et al., 1999; Schwartz et al., 2004), suggesting that SCN neurons
were functioning normally. cFos expression in the SCN was also not correlated
with any of the observed changes in activity seen in LSPV-lesioned animals.
Thus, the SCN was both intact and functional following NMA lesions, as indicated
by basic morphology and assays of cellular activity.

All lesions that damaged the LSPV also damaged parts of the upper
sPVZ. However, it is unlikely that changes in behavior in LSPV-lesioned animals
were due to collateral damage sustained by the upper sPVZ. Many MlSSes had
large lesions centered in the upper sPVZ, so if the patterns seen in LSPVx
animals were caused by damage to the upper sPVZ, they should have been
seen in both groups. While MlSSes did exhibit reduced-amplitude rhythms in DD
compared the PREOP phase of the study, amplitudes in LD1 and LD2 did not
differ from those seen during the PREOP period, and MlSSes were similar to
SHAMs on all other behavioral measures. Thus, the behavioral effects of lesions
that included the LSPV were different and far more severe than those that did
not, suggesting that these effects were due specifically to damage within the

lower part of the sPVZ (i.e. the LSPV) rather than to other portions of it.

Free-running rhythms in LSPV-lesioned animals:

Activity rhythms were profoundly disrupted in LSPV-damaged animals.
LSPVx animals had lower-amplitude activity rhythms in all postoperative

conditions compared to the PREOP condition, and were more variable than

128

MISS or SHAM animals with respect to daily offset times. These deficits were
most apparent in DD, when amplitude was lowest in animals with the most
damage in the LSPV. However, activity in lesioned animals was not completely
arrhythmic in constant conditions. In some animals, activity was disorganized,
but still weakly rhythmic in DD, and in others free-running activity rhythms were
disrupted by arrhythmic episodes lasting several days. However, these animals
eventually resumed activity rhythms that free-ran with a phase that could be
predicted from that of the preceding rhythm. Thus, although LSPV-damaged
grass rats were unable to maintain a consistently stable endogenous activity
rhythm, the clock normally governing that activity appears to have been
undisturbed. In lab rats, free-running locomotor rhythms were also attenuated
following excitotoxic lesions that included the LSPV (Lu et al., 2001), and were
abolished following electrolytic lesions that destroyed both cells and fibers in the
sPVZ (Abrahamson and Moore, 2006). This raises the possibility that cells in the
LSPV act to reinforce the normal patterns of coupling between the clock and
general activity, and is consistent with the hypothesis that differences in this
coupling seen in diurnal and nocturnal rodents could be mediated in part by the
LSPV.

It is unclear what triggered arrhythmic episodes in LSPV-damaged
animals. They appeared to increase in frequency and last longer over time,
suggesting that prolonged exposure to constant conditions may be necessary for

them to emerge. We are not aware of similar phenomena in other species.

129

Further research should investigate whether these arrhythmias are evident in

functions other than activity, and whether they are unique to grass rats.

Entrainment in LSPV-lesioned animals:

LSPV lesions did not initially disrupt rhythms of animals maintained in an
LD cycle, but did slow re-entrainment to an LD cycle when grass rats were
returned to LD from DD. Although all animals re-entrained following re-
instatement of the LD cycle, animals with reduced cFos expression in the LSPV
took longer to do so than those whose LSPV was intact. The effect of the lesions
on re-entrainment was not secondary to an effect on the phase of the clock
rhythm in DD, as lesions did not systematically affect 1?. Rather, lesions could
have affected the recovery of entrainment via one or both of two mechanisms.
First, signals from the LSPV to the SCN may influence basic entrainment
mechanisms that couple the clock to the LD cycle. Alternatively, coupling of the
pacemaker in the SCN to activity could have been significantly weakened in
LSPV-damaged animals such that re-entrainment of activity rhythms lagged
behind that of the clock. Bouts of arrhythmia in lesioned animals, in LD2 (as in
DD), suggest that the LSPV may ordinarily facilitate this coupling by aiding in the
suppression of activity during the night (or subjective night), enhancing the ability
of the pacemaker to drive or entrain rhythms in downstream targets. We view this
as the most parsimonious explanation.

When animals did re-establish entrainment, the resulting activity rhythms
were less diurnal in animals with the most damage to the LSPV. Reduced

diurnality in LD2 did not simply reflect a decrease in the amplitude of the rhythm

130

in lesioned animals. cFos expression in the LSPV was not correlated with either
amplitude or total activity levels at this stage, and neither the amplitude nor total
activity levels of LSPVx animals differed in LD1 and LD2. However, there was no
relationship between cFos in the LSPV and diurnality in LD1, suggesting that
exposure to DD and/or difficulties related to re-entrainment may have induced a
shift in activity distribution. Lesions may have led to a more general vulnerability
of the circadian system to perturbation that became apparent when it was
challenged by transitions from LD to DD and back, consistent with the hypothesis
that the LSPV reinforces the coupling between the cock and activity. We do not
know if such vulnerability is unique to diurnal species as there are presently no
data on re-entrainment in LSPV- or sPVZ-lesioned nocturnal ones.

LSPV destruction does, however, appear to have a greater effect on
entrained activity rhythms in grass rats than lab rats in which chemical lesions
that included the LSPV did not affect locomotor activity rhythms in LD
immediately following surgery in lab rats (Lu et al., 2001). However, in that study
animals were only monitored for five days in LD following surgery, leaving open
the possibility that rhythms could have degraded over a longer period of time (as
they did in the present study). In another study, electrolytic sPVZ lesions
abolished locomotor rhythms in lab rats kept in LD as well as DD (Abrahamson
and Moore, 2006), and partial sPVZ ablation via wire knife lesions (Moore and
Danchenko, 2002) attenuated rhythms in DD and LD in proportion to the extent
of sPVZ damage. However, the methods used in these studies are likely to have

destroyed both cell bodies in the sPVZ and SCN efferents passing though it. The

131

extent to which the effects of sPVZ lesions in those studies might have been due
to damage to its lower (ventral) region, versus destruction of SCN projections
beyond the LSPV, is thus not clear. In the present study, LSPV-damaged grass
rats took longer to entrain to an LD cycle than did animals whose LSPV was
intact, and once re-entrained were also less diurnal. Because SCN efferents
were intact in these animals, the behavioral effects must be a result of damage to

cells in the LSPV.

Conclusions:

In grass rats, the LSPV is likely to transmit circadian signals to other brain
regions via projections that parallel those of the SCN, extending to the same set
of hypothalamic and limbic forebrain targets (Watts et al., 1987; Morin et al.,
1994). They may even project to the same cells, as neurons in the grass rat
LSPV and SCN both form appositions with gonadotropin-releasing hormone
neurons (chapter 4) and orexin neurons (Schwartz and Smale, 2006), two cell
populations whose rhythmic activity is reversed in diurnal and nocturnal species
(McElhinny et al., 1999; Martinez et al., 2002; Mahoney et al., 2004). The LSPV
thus exhibits the anatomical connectivity necessary to communicate circadian
signals to targets that regulate functions that are rhythmically inverted in
nocturnal and diurnal species.

The LSPV of grass rats is similar in some ways and very different in others
from that of lab rats. In both species the LSPV is devoid of signaling molecules
that are present in the SCN, such as VP, vasoactive intestinal polypeptide, or

prokineticin 2, (Smale and Boverhof, 1999; Schwartz et al., 2004; Lambert et al.,

132

2005). Thus, although these two regions project to the same areas, the signals
they communicate are different, in both species. However, grass rats differ from
lab rats with respect to the phase of the rhythm in the LSPV relative to that of the
SCN (and to the LD cycle), and with respect to the dependence of that rhythm on
an LD cycle (Schwartz et al., 2004). The present study provides experimental
evidence for the first time that the grass rat LSPV is involved with the
organization of stable activity rhythms in both LD and DD conditions, that it
affects re-entrainment rates and that its role may be different from that of the
LSPV of nocturnal species. Taken together, these data suggest that the LSPV,
like the SCN, communicates circadian signals necessary for the expression of
normal activity rhythms, and that species differences in signals coming from the

LSPV may contribute to the differences between diurnal and nocturnal rodents.

133

CHAPTER 6

Conclusions

Circadian rhythms in physiology and behavior are dramatically different in
diurnal and nocturnal rodents. However, the SCN, that houses the circadian
pacemaker, exhibits similar rhythms in clock gene expression (Hastings et al.,
1999; Bae et al., 2001; Mrosovsky et al., 2001; Yan and Okamura, 2002;
Caldelas et al., 2003; Lambert and Weaver, 2006; Novak et al., 2006), metabolic
activity (Schwartz, 1991), and firing rate (Kubota et al., 1981; Sato and
Kawamura, 1984b; Kurumiya and Kawamura, 1988). Furthermore, rhythms in
synthesis of three SCN output signals that have been examined are similar in
diurnal and nocturnal rodents (Krajnak et al., 1998; Lambert and Weaver, 2006;
Mahoney et al., 2006). Together, these data suggest that the primary neural
mechanisms determining whether an animal is diurnal or nocturnal are unlikely to
reside within the SCN. Rather, circadian signals from the SCN are probably
modulated by downstream mechanisms in ways that contribute to determination
of whether an animal has a diurnal or nocturnal pattern of rhythmicity. The
experiments in this thesis evaluated this hypothesis by exploring the function and
anatomy of the LSPV, a major target of the SCN (Watts et al., 1987) that
functions differently in diurnal and nocturnal rodents (Nunez et al., 1999).

In Chapter 2, cFos rhythms in the LSPV were assessed in lab rats and
grass rats housed in LD and in DD. The rising phase of the rhythm in cFos
expression in the LSPV was advanced by roughly 8 hours in grass rats

compared to lab rats, and this rhythm was endogenous in the former but not the

134

latter species. cFos expression in the LSPV was clearly outside the SCN’s
borders as defined by Nissl staining and AVP-ir. The grass rat LSPV thus
represents a region receiving major inputs from the SCN (Watts et al., 1987;
Smale and Boverhof, 1999) that exhibits dramatically different patterns of
rhythmicity in grass rats and lab rats.

In Chapter 3, the efferent projections of the SCN and LSPV were mapped
using injections of the anterograde tracer BDA. This study revealed that the SCN
and the LSPV project in parallel to a nearly identical set of targets within the
hypothalamus, midline thalamus, septal region and basal forebrain.
Furthermore, the distributions of efferent fibers from the grass rat SCN and LSPV
appear to be very similar to those of nocturnal lab rats and hamsters (Watts and
Swanson, 1987; Watts et al., 1987; Morin et al., 1994; Kriegsfeld et al., 2004).
These parallelisms suggest that in diurnal and nocturnal species the SCN
communicates with targets both directly via its own axonal projections, and
indirectly via the LSPV. Signals carried along these direct and indirect pathways
likely affect their targets differently, as their temporal patterns of function and
their neuropeptide content are not the same. These signals are also likely to
differ in lab rats and grass rats, as rhythms in the LSPV are quite different in
these two species.

Chapter 4 followed up on this idea by examining SCN and LSPV
projections to GnRH neurons, a specific population of cells whose rhythmic
activity is reversed in grass rats relative to lab rats (McElhinny et al., 1999;

Mahoney et al., 2004). Fibers from both the SCN and the LSPV formed

135

numerous appositions with GnRH neurons, including more than half of those
counted in the OVLT and rostral POA. Thus, the pattern of parallel innervation
by the SCN and LSPV is maintained in specific cell populations known to exhibit
reversed patterns of rhythmic activity in diurnal and nocturnal rodents. The
pattern also raises the possibility that individual GnRH neurons receive input
from both the SCN and the LSPV.

In Chapter 5, LSPV function was assessed via chemical lesions that
spared both the SCN and its efferents. Free-running general activity rhythms
and re-entrainment of these rhythms to an LD cycle were profoundly disturbed in
grass rats with LSPV lesions. Thus, the coordination of circadian behavioral
rhythms is influenced by the LSPV. However, lesioned animals were still capable
of maintaining and entraining low-amplitude activity rhythms, indicating that the
core oscillatory mechanisms underlying circadian activity rhythms were
preserved. The LSPV is therefore not essential for maintaining circadian activity
rhythms, or for a diurnal phase preference, but it is clearly essential for the
normal expression of both.

Together, these experiments suggest that the grass rat LSPV is involved
in transmitting circadian signals to the same suite of targets as does the SCN.
The LSPV expressed endogenous cFos rhythms that were nearly 8 hours
advanced relative to cFos rhythms in the SCN. The LSPV of grass rats also
expresses daily rhythms in Per1 and Per2 whose phase differs from that of Per
rhythms in the SCN (Ramanathan et al., 2006). These data are consistent with a

role for the LSPV in communication of circadian signals to other brain regions

136

and that these signals differ from those transmitted by the SCN. When signaling
from the LSPV was eliminated via neuron-specific lesions, general activity
frequently appeared to become decoupled from the circadian clock, but circadian
control was always re-established within several days. This decoupling was
especially common in situations where external time cues were absent (as in DD)
or in conflict with internal signals (as in re-entrainment to a new LD cycle).
However, LSPV lesions, while causing disruptions in activity rhythms, neither
abolished them nor caused stable long-term alterations in their phase. The LSPV
is thus not essential for the basic diurnal pattern of activity rhythms in these
animals. Rather, it may strengthen or reinforce the coupling between the SCN
and its targets such that a stable, diurnal pattern is maintained.

Diurnality in grass rats may ordinarily be determined primarily by the
nature of the direct coupling between the SCN and targets regulating locomotor
activity. For example, the SCN could emit patterns of signals that determine
whether an animal has a diurnal or nocturnal pattern of activity, although the
available evidence suggests that this is unlikely to be the case. Alternatively,
phase preference could be determined by the coupling between the SCN and a
network of key intermediates, of which the LSPV is only one. The highly
interconnected nature of major SCN and LSPV targets (POA, PVT, PVN, DMH,
VMH) is consistent with the hypothesis that phase preference emerges from a
networked organization. What the data in this thesis have established is that the
LSPV contributes to the regulation of diurnality, and that it is not likely to do this

on its own. More broadly speaking, phase preference, be it diurnal or nocturnal,

137

is likely to arise from the interactions of multiple direct and indirect output
pathways from the SCN as well as from other sources of temporal information.
For example, the presence or absence of light may also influence the behavior of
targets downstream of the SCN via masking. Pathways from the SCN to the
LSPV, and from the LSPV to targets that also receive direct input from the SCN,
appear to represent an important part of this system.

The mechanisms by which the LSPV regulates rhythmic activity in
conjunction with the SCN must be related to both its inputs and its outputs.
Inputs to neurons in the grass rat LSPV lead to expression of cFos in response to
phase-shifting light pulses (Mahoney et al., 2001), and, as described in Chapter
5, circadian rhythms were most strongly perturbed after prolonged exposure to
DD and during re-entrainment to an LD cycle, suggesting that LSPV cells are
involved in the entrainment processes. Further research should thus assess the
extent to which activity in the LSPV is sensitive to entraining stimuli, such as
light. Rhythmic activity in LSPV neurons, as measured by cFos expression and
Per expression, should be examined under free-running, phase-shifted and
entrained conditions in both diurnal and nocturnal rodents, to compare the time
course of entrainment of rhythms in the LSPV to that of SCN rhythms. In this
way, it may be possible to determine whether cells in the LSPV respond directly
to entraining signals, are synchronized by signals from the SCN, or both. It will
also be important to evaluate the extent to which the LSPV of nocturnal species

is involved in entrainment, as this is presently unknown.

138

Further research should also evaluate whether cells in the LSPV integrate
circadian signals with information pertaining to internal variables such as arousal
state, as has been suggested in nocturnal species (Saper et al., 2005). Initial
studies suggest that cFos rhythms in the grass rat LSPV are unaffected by
variation in sleep-wake schedule (Schwartz and Smale, 2005), but it is not known
whether the same is true in nocturnal species. To this end, description of the
major inputs to the grass rat LSPV via injection of retrograde tracers may identify
areas that contribute to the rhythmic activity of LSPV cells, and possibly to
species differences in LSPV activity. For example, cholinergic innervation of the
SCN and LSPV is more dense in lab rats than grass rats (Castillo-Ruiz,
unpublished data), which is consistent with the idea that LSPV activity is less
sensitive to variations in sleep-wake state in grass rats than in lab rats.

The LSPV’s influence on rhythms of multiple behavioral and physiological
parameters should also be examined more carefully. In nocturnal rodents,
different types of lesions around the SCN have different effects on activity, core
body temperature, drinking and neuroendocrine rhythms (Nunez and Stephan,
1977; Stephan and Nunez, 1977; Nunez and Casati, 1979; Dark, 1980; Brown
and Nunez, 1986; Watts et al., 1989; Lu et al., 2001; Moore and Danchenko,
2002; Abrahamson and Moore, 2006). Furthermore, certain rhythms, such as
that of pineal melatonin secretion, are not reversed in diurnal and nocturnal
species (Reiter et al., 1982; Maywood et al., 1993; Klante et al., 1999; Garidou et
al., 2002; Perreau-Lenz et al., 2003). These data raise the possibility that LSPV

neurons could regulate some but not all circadian rhythms. For example, the

139

area above the SCN may be unrelated to pineal function but very much involved
in rhythms associated with the hypothalamic-pituitary-gonadal (HPG) axis.
Lesions (Docke et al., 1982), knife cuts (Nunez and Stephan, 1977) and electrical
stimulation (Arendash and Gallo, 1979) centered in the area above the SCN
affect the ability of rats to generate an LH surge, and daily rhythms in glial activity
above the SCN are related to the ability of lab rats to generate an LH surge
(Fernandez-Galaz et al., 1999). In chapter four, both SCN and LSPV neurons
were found to project directly onto GnRH neurons in grass rats, suggesting that
circadian signals from both areas may be important for maintaining a diurnal
phase preference in the HPG axis. Future research should evaluate this
hypothesis by determining whether axon-sparing LSPV lesions affect the GnRH
neurons and their ability to coordinate the LH surge in grass rats.

In summary, the LSPV, a principal target of the SCN in diurnal and
nocturnal rodents, exhibits the basic functional and anatomical characteristics
necessary to communicate circadian signals originating in the SCN to other brain
regions. Differences in the timing and the persistence of rhythmic activity in the
grass rat LSPV suggest that this region could modulate those signals differently
in diurnal and nocturnal rodents. Finally, signals from the LSPV contribute to
rhythmic precision and stability in grass rats, as destruction of the LSPV causes
profound disturbances in free-running activity rhythms and in entrainment of
those rhythms to a light: dark cycle. Together, these results suggest that a

normally functioning circadian system is likely to depend on contributions from

140

multiple oscillators that are ultimately coordinated by the master pacemaker in

the SCN, and that the LSPV is one of those oscillators.

141

References

Abrahamson EE, Moore RY (2001) Suprachiasmatic nucleus in the mouse:
retinal innervation, intrinsic organization and efferent projections. Brain

Res 916:172-191.

Abrahamson EE, Moore RY (2006) Lesions of suprachiasmatic nucleus efferents

selectively affect rest-activity rhythm. Mol Cell Endocrinol 252:46-56.

Allen G, Rappe J, Earnest DJ, Cassone VM (2001) Oscillating on borrowed time:
diffusible signals from immortalized suprachiasmatic nucleus cells regulate

circadian rhythmicity in cultured fibroblasts. J Neurosci 21 :7937-7943.

Amir S, Lamont EW, Robinson B, Stewart J (2004) A circadian rhythm in the
expression of PERIOD2 protein reveals a novel SCN-controlled oscillator
in the oval nucleus of the bed nucleus of the stria terminalis. J Neurosci

24:781-790.

Antle MC, Silver R (2005) Orchestrating time: arrangements of the brain

circadian clock. Trends Neurosci 28:145-151.

Arendash GW, Gallo RV (1979) Regional differences in response to electrical
stimulation within the medial preoptic-suprachiasmatic region on blood
luteinizing hormone levels in ovariectomized and ovariectomized,

estrogen-primed rats. Endocrinology 104:333-343.

142

Bae K, Jin X, Maywood ES, Hastings MH, Reppert SM, Weaver DR (2001)
Differential functions of mPer1, mPer2, and mPer3 in the SCN circadian

clock. Neuron 30:525-536.

Beaule C, Arvanitogiannis A, Amir S (2001) Light suppresses Fos expression in
the shell region of the suprachiasmatic nucleus at dusk and dawn:
implications for photic entrainment of circadian rhythms. Neuroscience

106:249-254.

Bell-Pedersen D, Cassone VM, Earnest DJ, Golden 88, Hardin PE, Thomas TL,
Zoran MJ (2005) Circadian rhythms from multiple oscillators: lessons from

diverse organisms. Nat Rev Genet 6:544-556.

Brown MH, Nunez AA (1986) Hypothalamic circuits and circadian rhythms:
effects of knife cuts vary with their placement within the suprachiasmatic

area. Brain Res Bull 16:705-711.

Caldelas l, Poirel VJ, Sicard B, Pevet P, Challet E (2003) Circadian profile and
photic regulation of clock genes in the suprachiasmatic nucleus of a

diurnal mammal arvicanthis ansorgei. Neuroscience 116:583-591.

Card JP, Moore RY (1989) Organization of lateral geniculate-hypothalamic

connections in the rat. J Comp Neurol 284:135-147.

143

Cheng MY, Bullock CM, Li C, Lee AG, Bermak JC, Belluzzi J, Weaver DR, Leslie
FM, Zhou QY (2002) Prokineticin 2 transmits the behavioural circadian

rhythm of the suprachiasmatic nucleus. Nature 417:405-410.

Cheung PW, McCormack CE (1983) Splitting of the locomotor activity rhythm in

rats by exposure to continuous light. Am J Physiol 244:R573-576.

Dardente H, Menet JS, Challet E, Tournier BB, Pevet P, Masson-Pevet M (2004)
Daily and circadian expression of neuropeptides in the suprachiasmatic
nuclei of nocturnal and diurnal rodents. Brain Res Mol Brain Res 124:143-

151.

Dark J (1980) Partial isolation of the suprachiasmatic nuclei: effects on circadian

rhythms of rat drinking behavior. Physiol Behav 25:863-873.

de la lglesia HO, Schwartz WJ (2006) Minireview: timely ovulation: circadian
regulation of the female hypothalamo-pituitary-gonadal axis.

Endocrinology 147:1 148-1 153.

de la lglesia HO, Blaustein JD, Bittman EL (1995) The suprachiasmatic area in
the female hamster projects to neurons containing estrogen receptors and

GnRH. Neuroreport 621715-1722.

144

de la lglesia HO, Meyer J, Schwartz WJ (2003) Lateralization of circadian
pacemaker output: Activation of left- and right-sided luteinizing hormone-
releasing hormone neurons involves a neural rather than a humoral

pathway. J Neurosci 23:7412-7414.

Deurveilher S, Semba K (2003) Indirect projections from the suprachiasmatic

nucleus to the median preoptic nucleus in rat. Brain Res 987:100-106.

Deurveilher S, Semba K (2004) Indirect projections from the suprachiasmatic
nucleus to major arousal-promoting cell groups in rat: Implications for the

circadian control of behavioural state. Neuroscience 130:165-183.

Deurveilher S, Burns J, Semba K (2002) Indirect projections from the
suprachiasmatic nucleus to the ventrolateral preoptic nucleus: a dual tract-

tracing study in rat. Eur J Neurosci 16:1195-1213.

Docke F, Lung DN, Rohde W, Dorner G (1982) Perisuprachiasmatic lesions
enhance the increase of gonadotropin secretion in acutely ovariectomized

rats. Endokrinologie 79:311-314.

Dunlap JC (1999) Molecular bases for circadian clocks. Cell 96:271-290.

Dunlap JC, Loros JJ, DeCoursey PJ (2004) Chronobiology: biological

timekeeping. Sunderland, Mass.: Sinauer Associates.

145

Everett JW, Sawyer CH (1950) A 24-Hour Periodicity in the Lh-Release
Apparatus of Female Rats, Disclosed by Barbiturate Sedation.

Endocrinology 47:198-218.

Fernandez-Galaz MC, Martinez Munoz R, Villanua MA, Garcia-Segura LM
(1999) Diurnal oscillation in glial fibrillary acidic protein in a
perisuprachiasmatic area and its relationship to the luteinizing hormone

surge in the female rat. Neuroendocrinology 70:368-376.

Field MD, Maywood ES, O'Brien JA, Weaver DR, Reppert SM, Hastings MH
(2000) Analysis of clock proteins in mouse SCN demonstrates
phylogenetic divergence of the circadian clockwork and resetting

mechanisms. Neuron 25:437-447.

Garidou ML, Gauer F, Vivien-Roels B, Sicard B, Pevet P, Simonneaux V (2002)
Pineal arylalkylamine N-acetyltransferase gene expression is highly
stimulated at night in the diurnal rodent, Arvicanthis ansorgei. Eur J

Neurosci 15:1632-1640.

Green DJ, Gillette R (1982) Circadian rhythm of firing rate recorded from single

cells in the rat suprachiasmatic brain slice. Brain Res 245:198-200.

Gu GB, Simerly RB (1997) Projections of the sexually dimorphic anteroventral

periventricular nucleus in the female rat. J Comp Neurol 384:142-164.

146

Guido ME, de Guido LB, Goguen D, Robertson HA, Rusak B (1999a) Daily
rhythm of spontaneous immediate-early gene expression in the rat

suprachiasmatic nucleus. J Biol Rhythms 14:275-280.

Guido ME, Goguen D, De Guido L, Robertson HA, Rusak B (1999b) Circadian
and photic regulation of immediate-early gene expression in the hamster

suprachiasmatic nucleus. Neuroscience 90:555-571.

Guo H, Brewer JM, Lehman MN, Bittman EL (2006) Suprachiasmatic regulation
of circadian rhythms of gene expression in hamster peripheral organs:

effects of transplanting the pacemaker. J Neurosci 26:6406-6412.

Hakim H, DeBernardo AP, Silver R (1991) Circadian locomotor rhythms, but not
photoperiodic responses, survive surgical isolation of the SCN in

hamsters. J Biol Rhythms 6297-113.

Hamada T, Antle MC, Silver R (2004) Temporal and spatial expression patterns
of canonical clock genes and clock-controlled genes in the

suprachiasmatic nucleus. Eur J Neurosci 19:1741-1748.

Hamada T, LeSauter J, Lokshin M, Romero MT, Yan L, Venuti JM, Silver R
(2003) Calbindin influences response to photic input in suprachiasmatic

nucleus. J Neurosci 23:8820-8826.

147

Hastings MH, Roberts AC, Herbert J (1985) Neurotoxic lesions of the anterior
hypothalamus disrupt the photoperiodic but not the circadian system of the

Syrian hamster. Neuroendocrinology 40:316-324.

Hastings MH, Field MD, Maywood ES, Weaver DR, Reppert SM (1999)
Differential regulation of mPERt and mTIM proteins in the mouse
suprachiasmatic nuclei: new insights into a core clock mechanism. J

Neurosci 19:RC11.

Hiatt ES, Brunetta PG, Seiler GR, Barney SA, Selles WD, Wooledge KH, King JC
(1992) Subgroups of luteinizing hormone-releasing hormone perikarya
defined by computer analyses in the basal forebrain of intact female rats.

Endocrinology 130:1030-1043.

lnouye ST, Kawamura H (1979) Persistence of circadian rhythmicity in a
mammalian hypothalamic "island" containing the suprachiasmatic nucleus.

Proc Natl Acad Sci U S A 76:5962-5966.

Jiao YY, Lee TM, Rusak B (1999) Photic responses of suprachiasmatic area
neurons in diurnal degus (Octodon degus) and nocturnal rats (Rattus

norvegicus). Brain Res 817293-103.

148

Jobst EE, Allen CN (2002) Calbindin neurons in the hamster suprachiasmatic
nucleus do not exhibit a circadian variation in spontaneous firing rate. Eur

J Neurosci 16:2469-2474.

Jobst EE, Robinson DW, Allen CN (2004) Potential pathways for intercellular
communication within the calbindin subnucleus of the hamster

suprachiasmatic nucleus. Neuroscience 123:87-99.

Kalamatianos T, Kallo l, Goubillon ML, Coen CW (2004) Cellular expression of
V1 a vasopressin receptor mRNA in the female rat preoptic area: effects of

oestrogen. J Neuroendocrinol 16:525-533.

Katona C, Smale L (1997) Wheel-running rhythms in Arvicanthis niloticus.

Physiol Behav 61:365-372.

Katona C, Rose S, Smale L (1998) The expression of Fos within the
suprachiasmatic nucleus of the diurnal rodent Arvicanthis niloticus. Brain

Res 791227-34.

Klante G, Secci K, Masson-Pevet M, Pevet P, Vivien-Roels B, Steinlechner S,
Wollnik F (1999) Interstrain differences in activity pattern, pineal function,
and SCN melatonin receptor density of rats. Am J Physiol 276:R1078-

1086.

149

Konopka RJ, Benzer S (1971) Clock mutants of Drosophila melanogaster. Proc

Natl Acad Sci U S A 68:2112-2116.

Krajnak K, Dickenson L, Lee TM (1997) The induction of Fos-like proteins in the
suprachiasmatic nuclei and intergeniculate leaflet by light pulses in degus

(Octodon degus) and rats. J Biol Rhythms 12:401-412.

Krajnak K, Kashon ML, Rosewell KL, Wise PM (1998) Sex differences in the
daily rhythm of vasoactive intestinal polypeptide but not arginine
vasopressin messenger ribonucleic acid in the suprachiasmatic nuclei.

Endocrinology 139:4189-4196.

Kramer A, Yang FC, Snodgrass P, Li X, Scammell TE, Davis FC, Weitz CJ
(2001) Regulation of daily locomotor activity and sleep by hypothalamic

EGF receptor signaling. Science 294:2511-2515.

Kraves S, Weitz CJ (2006) A role for cardiotrophin-like cytokine in the circadian

control of mammalian locomotor activity. Nat Neurosci 9:212-219.

Kriegsfeld LJ, Silver R (2006) The regulation of neuroendocrine function: Timing

is everything. Horm Behav 49:557-574.

150

Kriegsfeld LJ, Silver R, Gore AC, Crews D (2002) Vasoactive intestinal
polypeptide contacts on gonadotropin-releasing hormone neurones

increase following puberty in female rats. J Neuroendocrinol 14:685-690.

Kriegsfeld LJ, Leak RK, Yackulic CB, LeSauter J, Silver R (2004) Organization of
suprachiasmatic nucleus projections in Syrian hamsters (Mesocricetus
auratus): an anterograde and retrograde analysis. J Comp Neurol

468:361-379.

Kubota A, lnouye ST, Kawamura H (1981) Reversal of multiunit activity within
and outside the suprachiasmatic nucleus in the rat. Neurosci Lett 27:303-

308.

Kurumiya S, Kawamura H (1988) Circadian oscillation of the multiple unit activity
in the guinea pig suprachiasmatic nucleus. J Comp Physiol [A] 162:301-

308.

Lambert CM, Weaver DR (2006) Peripheral gene expression rhythms in a diurnal

rodent. J Biol Rhythms 21:77-79.

Lambert CM, Machida KK, Smale L, Nunez AA, Weaver DR (2005) Analysis of
the prokineticin 2 system in a diurnal rodent, the unstriped Nile grass rat

(Arvicanthis niloticus). J Biol Rhythms 20:206-218.

151

Leak RK, Moore RY (2001) Topographic organization of suprachiasmatic nucleus

projection neurons. J Comp Neurol 433:312-334.

Lee WS, Smith MS, Hoffman GE (1990) Luteinizing hormone-releasing hormone
neurons express Fos protein during the proestrous surge of luteinizing

hormone. Proc Natl Acad Sci U S A 87:5163-5167.

Lee WS, Abbud R, Smith MS, Hoffman GE (1992) LHRH neurons express cJun
protein during the proestrous surge of luteinizing hormone. Endocrinology

130:3101-3103.

Legan SJ, Karsch FJ (1975) Daily Signal for Lh Surge in Rat. Endocrinology

96:57-62.

LeSauter J, Silver R (1999) Localization of a suprachiasmatic nucleus subregion

regulating locomotor rhythmicity. J Neurosci 19:5574-5585.

LeSauter J, Kriegsfeld LJ, Hon J, Silver R (2002) Calbindin-D(28K) cells

selectively contact intra-SCN neurons. Neuroscience 111:575-585.

Lu J, Zhang YH, Chou TC, Gaus SE, Elmquist JK, Shiromani P, Saper CB (2001)
Contrasting effects of ibotenate lesions of the paraventricular nucleus and
subparaventricular zone on sleep-wake cycle and temperature regulation.

J Neurosci 21 :4864-4874.

152

Mahoney M, Bult A, Smale L (2001) Phase response curve and light-induced fos
expression in the suprachiasmatic nucleus and adjacent hypothalamus of

Arvicanthis niloticus. J Biol Rhythms 16:149-162.

Mahoney MM, Smale L (2005a) A daily rhythm in mating behavior in a diurnal

murid rodent Arvicanthis niloticus. Horm Behav 47:8-13.

Mahoney MM, Smale L (2005b) Arginine vasopressin and vasoactive intestinal
polypeptide fibers make appositions with gonadotropin-releasing hormone
and estrogen receptor cells in the diurnal rodent Arvicanthis niloticus.

Brain Res 1049:156-164.

Mahoney MM, Nunez AA, Smale L (2000) Calbindin and Fos within the
suprachiasmatic nucleus and the adjacent hypothalamus of Arvicanthis

niloticus and Rattus norvegicus. Neuroscience 99:565-575.

Mahoney MM, Sisk C, Ross HE, Smale L (2004) Circadian regulation of
gonadotropin-releasing hormone neurons and the preovulatory surge in
luteinizing hormone in the diurnal rodent, Arvicanthis niloticus, and in a

nocturnal rodent, Rattus norvegicus. Biol Reprod 70:1049-1054.

Mahoney MM, Ramanathan C, Hagenauer MH, Smale L, Lee TM. A comparison
of mRNA rhythms in vasoactive intestinal polypeptide and vasopressin in

the suprachiasmatic nucleus of male and female grass rats. Program no.

153

P95. Society for Research on Biological Rhythms. Destin: Tenth Meeting,

Society for Research on Biological Rhythms, 2006.

Martinez GS, Smale L, Nunez AA (2002) Diurnal and nocturnal rodents show

rhythms in orexinergic neurons. Brain Res 95521-7.

Maywood ES, Hastings MH, Max M, Ampleford E, Menaker M, Loudon AS
(1993) Circadian and daily rhythms of melatonin in the blood and pineal
gland of free-running and entrained Syrian hamsters. J Endocrinol 136:65-

73.

McElhinny TL, Sisk CL, Holekamp KE, Smale L (1999) A morning surge in
plasma luteinizing hormone coincides with elevated Fos expression in
gonadotropin-releasing hormone-immunoreactive neurons in the diurnal

rodent, Arvicanthis niloticus. Biol Reprod 61:1115-1122.

Meyer-Bernstein EL, Jetton AE, Matsumoto SI, Markuns JF, Lehman MN,
Bittman EL (1999) Effects of suprachiasmatic transplants on circadian
rhythms of neuroendocrine function in golden hamsters. Endocrinology

140:207-218.

Moga MM, Weis RP, Moore RY (1995) Efferent projections of the paraventricular

thalamic nucleus in the rat. J Comp Neurol 359:221-238.

154

Moore-Ede MC, Sulzman FM, Fuller CA (1982) The clocks that time us:
physiology of the circadian timing system. Cambridge, Mass.: Harvard

University Press.

Moore RY, Eichler VB (1972) Loss of a circadian adrenal corticosterone rhythm

following suprachiasmatic lesions in the rat. Brain Res 42:201-206.

Moore RY, Card JP (1994) lntergeniculate leaflet: an anatomically and
functionally distinct subdivision of the lateral geniculate complex. J Comp

Neurol 344:403-430.

Moore RY, Silver R (1998) Suprachiasmatic nucleus organization. Chronobiol Int

15:475-487.

Moore RY, Leak RK (2001) Suprachiasmatic Nucleus. In: Handbook of
Behavioral Neurobiology, Vol. 12: Circadian Clocks, 1 Edition (Takahashi
JS, Turek, F.W., Moore, R. Y., ed), pp 141-179. New York: Kluwer

Academic/Plenum Publishers.

Moore RY, Danchenko RL (2002) Paraventricular-subparaventricular
hypothalamic lesions selectively affect circadian function. Chronobiol Int

1 9:345-360.

155

Morin LP, Blanchard J, Moore RY (1992) lntergeniculate leaflet and
suprachiasmatic nucleus organization and connections in the golden

hamster. Vis Neurosci 8:219-230.

Morin LP, Goodless-Sanchez N, Smale L, Moore RY (1994) Projections of the
suprachiasmatic nuclei, subparaventricular zone and retrochiasmatic area

in the golden hamster. Neuroscience 61 :391-410.

Morin LP, Shivers KY, Blanchard JH, Muscat L (2006) Complex organization of

mouse and rat suprachiasmatic nucleus. Neuroscience 137:1285-1297.

Mrosovsky N, Edelstein K, Hastings MH, Maywood ES (2001) Cycle of period
gene expression in a diurnal mammal (Spermophilus tridecemlineatus):

implications for nonphotic phase shifting. J Biol Rhythms 16:471-478.

Novak CM, Nunez AA (1998) Daily rhythms in Fos activity in the rat ventrolateral

preoptic area and midline thalamic nuclei. Am J Physiol 275:R1620-1626.

Novak CM, Albers HE (2004) Novel phase-shifting effects of GABAA receptor
activation in the suprachiasmatic nucleus of a diurnal rodent. Am J Physiol

Regul Integr Comp Physiol 286:R820-825.

156

Novak CM, Smale L, Nunez AA (2000) Rhythms in Fos expression in brain areas
related to the sleep-wake cycle in the diurnal Arvicanthis niloticus. Am J

Physiol Regul Integr Comp Physiol 278:R1267-1274.

Novak CM, Ehlen JC, Paul KN, Fukuhara C, Albers HE (2006) Light and GABA-A
receptor activation alter Period mRNA levels in the SCN of diurnal Nile

grass rats. In.

Nuesslein-Hildesheim B, O'Brien JA, Ebling FJ, Maywood ES, Hastings MH
(2000) The circadian cycle of mPER clock gene products in the
suprachiasmatic nucleus of the siberian hamster encodes both daily and

seasonal time. Eur J Neurosci 12:2856-2864.

Nunez AA, Stephan PK (1977) The effects of hypothalamic knife cuts on drinking

rhythms and the estrus cycle of the rat. Behav Biol 20:224-234.

Nunez AA, Casati MJ (1979) The role of efferent connections of the
suprachiasmatic nucleus in the control of circadian rhythms. Behav Neural

Biol 25:263-267.

Nunez AA, Bult A, McElhinny TL, Smale L (1999) Daily rhythms of Fos
expression in hypothalamic targets of the suprachiasmatic nucleus in

diurnal and nocturnal rodents. J Biol Rhythms 14:300-306.

157

Orpen BG, Steiner M (1994) Neural connections of the suprachiasmatic nucleus
and medial hypothalamus of the Syrian hamster (Mesocricetus auratus). J

Anat 184 (Pt 1):23-33.

Paxinos G, Watson C (2005) The rat brain in stereotaxic coordinates, 5th Edition.

Amsterdam; Boston: Elsevier Academic Press.

Perreau-Lenz S, Kalsbeek A, Garidou ML, Wortel J, van der Vliet J, van
Heijningen C, Simonneaux V, Pevet P, Buijs RM (2003) Suprachiasmatic
control of melatonin synthesis in rats: inhibitory and stimulatory

mechanisms. Eur J Neurosci 17:221-228.

Peterson GM, Moore RY (1980) Selective effects of kainic acid on diencephalic

neurons. Brain Res 202:165-182.

Porkka-Heiskanen T, Urban JH, Turek FW, Levine JE (1994) Gene expression in
a subpopulation of luteinizing hormone-releasing hormone (LHRH)
neurons prior to the preovulatory gonadotropin surge. J Neurosci 14:5548-

5558.

Ralph MR, Menaker M (1988) A mutation of the circadian system in golden

hamsters. Science 241:1225-1227.

158

Ralph MR, Foster RG, Davis FC, Menaker M (1990) Transplanted
suprachiasmatic nucleus determines circadian period. Science 247:975-

978.

Ramanathan C, Nunez AA, Martinez GS, Schwartz MD, Smale L (2006)
Temporal and spatial distribution of immunoreactive PER1 and PER2
proteins in the suprachiasmatic nucleus and peri-suprachiasmatic region
of the diurnal grass rat (Arvicanthis niloticus). Brain Res 1073-10742348-

358.

Reiner A, Veenman CL, Medina L, Jiao Y, Del Mar N, Honig MG (2000) Pathway

tracing using biotinylated dextran amines. J Neurosci Methods 103:23-37.

Reiter RJ, King TS, Richardson BA, Hurlbut EC (1982) Studies on pineal
melatonin levels in a diurnal species, the eastern chipmunk (Tamias
striatus): effects of light at night, propranolol administration or superior

cervical ganglionectomy. J Neural Transm 54:275-284.

Reppert SM, Weaver DR (2002) Coordination of circadian timing in mammals.

Nature 418:935-941.

Reppert SM, Schwartz WJ, Artman HG, Fisher DA (1983) Comparison of the
temporal profiles of vasopressin and oxytocin in the cerebrospinal fluid of

the cat, monkey and rat. Brain Res 261:341-345.

159

Salazar-Juarez A, Escobar C, Aguilar-Roblero R (2002) Anterior paraventricular
thalamus modulates light-induced phase shifts in circadian rhythmicity in

rats. Am J Physiol Regul Integr Comp Physiol 283:R897-904.

Saper CB, Lu J, Chou TC, Gooley J (2005) The hypothalamic integrator for

circadian rhythms. Trends Neurosci 28:152-157.

Sato T, Kawamura H (1984a) Effects of bilateral suprachiasmatic nucleus lesions
on the circadian rhythms in a diurnal rodent, the Siberian chipmunk

(Eutamias sibericus). J Comp Physiol 155:745-752.

Sato T, Kawamura H (1984b) Circadian rhythms in multiple unit activity inside
and outside the suprachiasmatic nucleus in the diurnal chipmunk

(Eutamias sibiricus). Neurosci Res 1:45-52.

Schwartz MD, Smale L (2005) Individual differences in rhythms of behavioral
sleep and its neural substrates in nile grass rats. J Biol Rhythms 20:526-

537.

Schwartz MD, Smale L The suprachiasmatic nucleus and the lower
subparaventricular zone project onto orexin-A neurons in a diurnal rodent

Atlanta 2006 Society for Neuroscience. 'Online.‘

160

Schwartz MD, Nunez AA, Smale L (2004) Differences in the suprachiasmatic
nucleus and lower subparaventricular zone of diurnal and nocturnal

rodents. Neuroscience 127:13-23.

Schwartz WJ (1991) SCN metabolic activity in vivo. In: Suprachiasmatic Nucleus:
The Mind's Clock (Klein DC, Moore, R.Y., and Reppert, S.M., ed), pp 144-

156. New York: Oxford University Press.

Schwartz WJ, Carpino A, Jr., de la Iglesia HO, Baler R, Klein DC, Nakabeppu Y,
Aronin N (2000) Differential regulation of fos family genes in the
ventrolateral and dorsomedial subdivisions of the rat suprachiasmatic

nucleus. Neuroscience 98:535-547.

Silver R, LeSauter J, Tresco PA, Lehman MN (1996a) A diffusible coupling signal
from the transplanted suprachiasmatic nucleus controlling circadian

locomotor rhythms. Nature 382:810-813.

Silver R, Romero MT, Besmer HR, Leak R, Nunez JM, LeSauter J (1996b)
Calbindin-D28K cells in the hamster SCN express light-induced Fos.

Neuroreport 7:1224-1228.

Simerly RB, Swanson LW (1986) The organization of neural inputs to the medial

preoptic nucleus of the rat. J Comp Neurol 246:312-342.

161

Simerly RB, Swanson LW (1988) Projections of the medial preoptic nucleus: a
Phaseolus vulgaris leucoagglutinin anterograde tract-tracing study in the

rat. J Comp Neurol 270:209-242.

Simonian SX, Spratt DP, Herbison AE (1999) Identification and characterization
of estrogen receptor alpha-containing neurons projecting to the vicinity of
the gonadotropin-releasing hormone perikarya in the rostral preoptic area

of the rat. J Comp Neurol 411:346-358.

Sisk CL, Nunez AA, Thebert MM (1988) Differential effects of electrolytic and
chemical hypothalamic lesions on LH pulses in rats. Am J Physiol

255: E583-590.

Smale L, Boverhof J (1999) The suprachiasmatic nucleus and intergeniculate
leaflet of Arvicanthis niloticus, a diurnal murid rodent from East Africa. J

Comp Neurol 403:190-208.

Smale L, Castleberry C, Nunez AA (2001) Fos rhythms in the hypothalamus of
Rattus and Arvicanthis that exhibit nocturnal and diurnal patterns of

rhythmicity. Brain Res 899:101-105.

Smale L, Lee TM, Nunez AA (2003) Mammalian Diurnality: Some Facts and

Gaps. J Biol Rhythms 18:356-366.

162

Smith MJ, Jiennes L, Wise PM (2000) Localization of the VIP2 receptor protein

on GnRH neurons in the female rat. Endocrinology 141:4317-4320.

Stephan FK, Zucker l (1972) Circadian rhythms in drinking behavior and
locomotor activity of rats are eliminated by hypothalamic lesions. Proc Natl

Acad Sci U S A 69:1583-1586.

Stephan FK, Nunez AA (1977) Elimination of circadian rhythms in drinking,
activity, sleep, and temperature by isolation of the suprachiasmatic nuclei.

Behav Biol 20:1-61.

Stewart GR, Price M, Olney JW, Hartman BK, Cozzari C (1986) N-
methylaspartate: an effective tool for Iesioning basal forebrain cholinergic

neurons of the rat. Brain Res 369:377-382.

Sumova A, Travnickova Z, Mikkelsen JD, lllnerova H (1998) Spontaneous rhythm
in c-Fos immunoreactivity in the dorsomedial part of the rat

suprachiasmatic nucleus. Brain Res 801 :254-258.

Swanson LW (2000) Cerebral hemisphere regulation of motivated behavior.

Brain Res 886:113-164.

163

Sylvester CM, Krout KE, Loewy AD (2002) Suprachiasmatic nucleus projection to
the medial prefrontal cortex: a viral transneuronal tracing study.

Neuroscience 114:1071-1080.

Takeo Y (1984) Influence of continuous illumination on estrous cycle of rats: time
course of changes in levels of gonadotropins and ovarian steroids until

occurrence of persistent estrus. Neuroendocrinology 39297-104.

ter Horst GJ, Luiten PG (1986) The projections of the dorsomedial hypothalamic

nucleus in the rat. Brain Res Bull 16:231-248.

Thompson RH, Swanson LW (1998) Organization of inputs to the dorsomedial
nucleus of the hypothalamus: a reexamination with Fluorogold and PHAL

in the rat. Brain Res Brain Res Rev 27:89-118.

Thompson RH, Canteras NS, Swanson LW (1996) Organization of projections
from the dorsomedial nucleus of the hypothalamus: a PHA-L study in the

rat. J Comp Neurol 376:143-173.

Urbanski HF (1991) Monoclonal antibodies to luteinizing hormone-releasing
hormone: production, characterization, and immunocytochemical

application. Biol Reprod 44:681-686.

164

van der Beek EM, Wiegant VM, van der Donk HA, van den Hurk R, Buijs RM
(1993) Lesions of the suprachiasmatic nucleus indicate the presence of a
direct vasoactive intestinal polypeptide-containing projection to
gonadotrophin-releasing hormone neurons in the female rat. J

Neuroendocrinol 5:137-144.

Van der Beek EM, Horvath TL, Wiegant VM, Van den Hurk R, Buijs RM (1997a)
Evidence for a direct neuronal pathway from the suprachiasmatic nucleus
to the gonadotropin-releasing hormone system: combined tracing and light
and electron microscopic immunocytochemical studies. J Comp Neurol

384:569-579.

van der Beek EM, van Oudheusden HJ, Buijs RM, van der Donk HA, van den
Hurk R, Wiegant VM (1994) Preferential induction of c-fos
immunoreactivity in vasoactive intestinal polypeptide-innervated
gonadotropin-releasing hormone neurons during a steroid-induced
luteinizing hormone surge in the female rat. Endocrinology 134:2636-

2644.

van der Beek EM, Wiegant VM, van Oudheusden HJ, van der Donk HA, van den
Hurk R, Buijs RM (1997b) Synaptic contacts between gonadotropin-
releasing hormone-containing fibers and neurons in the suprachiasmatic
nucleus and perichiasmatic area: an anatomical substrate for feedback

regulation? Brain Res 755:101-111.

165

Vidal L, Blanchard J, Morin LP (2005) Hypothalamic and zona incerta neurons
expressing hypocretin, but not melanin concentrating hormone, project to

the hamster intergeniculate leaflet. Neuroscience 134:1081-1090.

Vitaterna MH, King DP, Chang AM, Kornhauser JM, Lowrey PL, McDonald JD,
Dove WF, Pinto LH, Turek FW, Takahashi JS (1994) Mutagenesis and
mapping of a mouse gene, Clock, essential for circadian behavior.

Science 264:719-725.

Vrang N, Mrosovsky N, Mikkelsen JD (2003) Afferent projections to the hamster
intergeniculate leaflet demonstrated by retrograde and anterograde

tracing. Brain Res Bull 59:267-288.

Watson RE, Jr., Langub MC, Jr., Engle MG, Maley BE (1995) Estrogen-receptive
neurons in the anteroventral periventricular nucleus are synaptic targets of
the suprachiasmatic nucleus and peri- suprachiasmatic region. Brain Res

689:254-264.

Watts AG (1991) The efferent projections of the suprachiasmatic nucleus:
anatomical insights into the control of circadian rhythms. In:
Suprachiasmatic Nucleus: The Mind's Clock (Klein DC, Moore, R.Y., and

Reppert, S.M., ed), pp 77-106. New York: Oxford University Press.

166

Watts AG, Swanson LW (1987) Efferent projections of the suprachiasmatic
nucleus: Il. Studies using retrograde transport of fluorescent dyes and
simultaneous peptide immunohistochemistry in the rat. J Comp Neurol

258:230-252.

Watts AG, Swanson LW, Sanchez-Watts G (1987) Efferent projections of the
suprachiasmatic nucleus: I. Studies using anterograde transport of

Phaseolus vulgaris leucoagglutinin in the rat. J Comp Neurol 258:204-229.

Watts AG, Sheward WJ, Whale D, Fink G (1989) The effects of knife cuts in the
sub-paraventricular zone of the female rat hypothalamus on oestrogen-
induced diurnal surges of plasma prolactin and LH, and circadian wheel-

running activity. J Endocrinol 122:593-604.

Weaver DR (1998) The suprachiasmatic nucleus: a 25-year retrospective. J Biol

Rhythms 13:100-112.

Wiegand SJ, Terasawa E (1982) Discrete lesions reveal functional heterogeneity
of suprachiasmatic structures in regulation of gonadotropin secretion in

the female rat. Neuroendocrinology 34:395-404.

Yamazaki S, Numano R, Abe M, Hida A, Takahashi R, Ueda M, Block GD,
Sakaki Y, Menaker M, Tei H (2000) Resetting central and peripheral

circadian oscillators in transgenic rats. Science 288:682-685.

167

 

Yan L, Okamura H (2002) Gradients in the circadian expression of Per1 and
Per2 genes in the rat suprachiasmatic nucleus. Eur J Neurosci 15:1153-

1162.

Yan L, Takekida S, Shigeyoshi Y, Okamura H (1999) Per1 and Per2 gene
expression in the rat suprachiasmatic nucleus: circadian profile and the

compartment-specific response to light. Neuroscience 94:141-150.

168

lljliijjijiijjlijlllljljl