«r

"I...”zq-a

   
  
 
 

5?; ,jj' I MW .‘.;:';‘W ""’I::"’I;I;‘I‘7:;;:gh‘HWrHI " $5113.31";
':;('i'J - 3 3 '33 3 3 ' 3333' ‘ M5 ‘ m3 HIE! 3333';
Ilt‘ . 3 .' 5"" '3'; ' . : I .‘~31:L"'a'33"$" 3" “'3'3'“ $33.34?

‘53:: 5 I: ’ ~ I;I'r;:5:rII-r-,-5I45:
3"'I.I:~.Ivr.I$:..;I- I': "v” j.zI‘"

3":3' $30333'33'3"v"""7""‘I\33"I"3""')1'333'330'333'3'M‘I 3.I '3 '3 3"
4:31p: 3%! III-*7 3-51. I IILIIII III: ‘;:;;',’ : IIII. ,.I-’,I,. :

.3; "3231:3431. 3"":3'33333333'I'M”33"" E'3'333 ' 3"'I"3' 3‘ 3"" I 1"3-7333'3’3g‘3u):;3n ' :‘3‘3: '3" 3 [3" '3'33'1'r

can . I: I'Ir; "rrIIII III’I'I“ ,iji ;I.'I:I.«':I"_. IIHIm an I II {'“II I.” 4:53" I:
l‘ t “E .. 9",“ ‘ L.I,I.a)p, _.,.‘o5.3 5 1 5 )- 551:3“.
rIImIHQ%¢%Hmum¢uI2yh W ~1Mfi“'flwwu d3P3'"
at; 519-33” .rw‘f '3 453:3" I I' H H. .‘" ‘ ‘ 3 ' I;
- .,-5. 3,1; If‘I‘ 555-35, In}; .555, 355.5(1- '3111' 3 $05,. ‘33:: IHIJIIFJIIIFJI I51.' ”'3'”
;.'. L3 -' -.’ 3:; I ,' ‘ I . ‘ ," 5'35“ ‘II‘HJI'
. )5” . . 5" 3h . . .
5 Mn,“ .‘ 3333,33 :\||' 3.. . ii": 3-:
r 5 'L' 3 ° 3 ' {35393:
. 13‘; ' , I "-3 "91 1‘33”: f
.3 3:3: ' 555.55.75155333‘33323W55511 1 .1 52:“.‘3 3:55 .‘_-’
i ' ’3‘“ ““3354: 5 . ' ' a“ 7.

a . -.
‘vv
> o-

'7"

,.-

2 .
. . ,
nap»:-

  
 
  

-.-.

  
     

I 1‘
, II"?!
I 33"" “3'37, $5.3": Hi"
I ~ , '5‘: '
'~“I‘3::3:§""£“3" I; ‘II "‘3
u \ .33 i 3" ‘3

fm‘ _

--<o"¢'

uh 3":‘5'5 N'”: “5
I 5 5 “"3 .9! l
[3331:5593 3I3'IK

    
 
 

w:

.9...

m " ' ' ' '*'
’ 3'..'......'_
;“'.‘.t_'

~.~
:‘W L ’.
fI-‘. ~ .
4.. .II, .
. n
‘ v

   
 
  
 

     
     
 

 

 
  
 
 
 
   
  

   
 
 
  
 

     

 

 
  
  
    
   
 
 
 
 
 
 
 
 

    
   
  
   
  

        
 
 
 

    
 

   
  

       
     
 

   

 
 

   

 

v
I I 'j 5 I
0.5 5.3 .3“ “5‘; I i; -2. I, .
“ "I 4313 :I I'U: .1 43h? " -."‘
IIHgIfinrNIIIIH' H 5 '* JHHW WMH H”
" ’II'JI 3&3Ié!'rj23:3":53.315-4I I243: '13-:- 3 3'$II.I- ' ‘ I. ‘ {Ié'éé'r'IeIzII i '
25:5 33333313'33 4.555335555555455!“ Y". 5511:,5'235 '3""" 5,355 3M5“ . : '-I' ,1. ”55533555555555 .
my -.I .-5-' ~':-“:; 1-, 8);“, , \H1ISQI'3HI-W " “£31,435 _, . ._ -|' :; "1.1.5:. fir: ,.
55'55q33353535313,‘ '5"5351'5‘5433'I'f'li' 5&3 (332:3? Hump)!“ 55.31:: .. I; 5 33333:“ ‘9' 5’ 5415565555255, ' 3... " '3' ';-‘
Isa: . 1.7 ‘I':’ III: W: ' :er .. I15 55,935 . {I1 I55 _. 2w -,r. .3
l" -’ I3'I '3 ' #:1223533. U I ;';";3£J:53 I :i; ”4‘33? A I" [:3 '3
:il " I; ( 3' '33-}? .H ‘ 3 .
g! 3 33.? 9515 (3):?“ “34"" 3'33 5! 5: I. 5." -
I . 5~ , I \ ‘ ' I 5 I .
[2: 1-3" \§'\}.5§jfl "H.353; £3” '7' ’5I:’ 5h! c‘ 1‘
i I H ‘ s
33a"- t"? 43:32:}! ' "3 333,33 ,‘33 [41"3" ”"33 A313”: 53.5! )5; 55...] 39.3.:33' :33."
555 ' 1‘» IIIII}II5I "5,5,5; ,5 33; .3'5 I: :5.- 5 -, 3:3 ,5}; _ . 4; .- ,.‘2
3-33 ’ 'Iu'k 3" I-" ' ' 3" '3 "3'3" "313" Inst r 3,3: '33}?! ' 333333133; ’3 "' .‘ IIIz‘I" H‘
5 5 551:3.“ 4 m3”: 1‘" Ln”: 5-,: 3:2" :5 I55 535255555555} 'kfpl'é-f if )-.-"
’5 ~35. " ' ' 3 "' 3'3: ‘j': r 'L 197"? 333'. EL" fli'uqiffl‘ll 3:
IH' ’ }"" ' ‘ #9339339. 'I3 ’I'IfIr'I‘I (3 9 WW
.I' ' 3'" .I' 83"!“ "a"! ‘3'? ’3' 3 .'

    

III, 553,55...
H3;

   

35"
i"?

     
 
   

W ‘

 

, .7!

135553555933 “'3

E 335333333 {5:15 23, ’5 5513355555
{5555333,53'g515535555335555'53 '

.. .3“:

  
   
    

   

   
 
   

   
   
 
  
 

 
   
     

‘i
f ;
.52.
m -
In — X :- -v
¥ ~‘ " Ow - -vv-
‘1'“: ‘3. .4‘ .. -: r - .. wh'
.* I. 'I “3:2: I‘ fit: 3" v I -173: ' v
.vr- 5“ ’ - r v “ I
tut-r .. ‘ .. w; vfi'f' - ‘v‘
. ~ I'-
‘ 4.... ,.. 23.3.7 I. .h I ,. m
a I .,
M _ ‘ M 40:" ,' 7-2. ‘
Vt”: a“ m
, . w , A
a... .,
~— , ‘ ~
r» -

5
. u
.._ ~—

V‘J' .-.. < ‘z 'l ..
- - 1‘ m
.n- .. .

.. ”$.73 . 5 . .p
w .

 
    
    
 
 
 
 
  

 

   
  
 
 

 

   
       
   
 
   

 
 
   

 

   
  

         
 
 

v“

i55f333'1bu {33‘3'23313'33 ‘- ,I
3,5134: 33:33,; 3'3"; 3:33 3' '3 t 3
333:3? 3333333333“; giaéi, . ::,: .‘I I, s
git: :: my I,I=I,I 5;. :pz - - ‘
.I. I: .2 :‘232Ir33'If::if;shis‘$.. . ‘ 2 ~ I ,
:"i3f : :I IzI'Iv'EIII' "I’M? ~ NIKE“ “'3'
if z '-I ’ l'f' '35}: 3 ,3;$f3;:~,s ~ , 5‘ :5'.
RI ‘ I'M" 353:: MI 1 ‘ 3'3" 3'
II' IHH I~—I : I; uma- I
' nfi ‘ h} ”I, 'P ‘3 e '53; r'l",lf,.‘ ‘ d I
' a“; ”355 1,3: 1,. 5‘» ' ’ g ;3 r}‘_ $115k??? ' I
3 5 i l' ‘7 v, b4.” I . (Er) : 1'. ,9- ,I ,
" 3‘ 3’3 3' 3'." I' " - in 313‘3'3-4 3'3'3'53“ 3» 13:4
I 3 g 5“ 3. 0‘ - , 3 1'! “-55" .t‘ I -
3 0 311%.33 3-5! . ,' ' 3 f f ' ‘35 533*]; . up"!
I ‘ 1% 33’3"}; 3343} I 33% 3 3’; 9.23;" ' ' .533: 33
I 35:5 ")3 181 , 5 , R 3,3522 t' j I' 5
d ’ Igéfljfiwxw ' 'mfiy,k- I‘353
. ,, : III“ III; .; IEEHI‘III. - f2:
:5 51 .5: . 5. .5 53(555 . .' £15m ~55? $3. 5:25:15
' s 33 . ’1 ‘5 $3335: 5;;
:-I I

W.-
“m-
--.‘~'~ u
52:.
m-
I “‘3' :n

' Thaw
‘u; :3
u.- a
w;-
I WW
- o
. “I“
w u.
-‘ -_. '
' w
M0- ‘
0“
Q
M.
- .,
m
an

 

I ‘ . ‘ : i A 31:;
' ' 1* 3 I ' .1 ._“"B;‘
m. o 5 3f 3 .. 3'3?! ’1? s 3 13"" .333
'3' 3 F' 3 ' ”'1'? '3 '7" ' ’
3:3.“ 13I}.'..!.. . in“, 3 ' 33'31313'33133'3i."i: ' I

MICHIGAN sure

I MINIMUM/(I“WWIWt

3 1293 00790 3234

W

 

 

 

 

 

 

 

 

 

 

 

This is to certify that the

dissertation entitled
THE RETINO—HYPOTHALAMO-SPINAL PATHWAY WHICH
MEDIATES GONADAL RESPONSES TO PHOTOPERIOD
IN SYRIAN HAMSTERS
presented by

TIMOTHY GRANT YOUNGSTROM

has been accepted towards fulfillment

of the requirements for
Neuroscience/
PH . D .

degree in Psychology

 

Major professor

Date DECEMBER 18, 1989

 

MS U is an Affirmative Action/Equal Opportunity Institution

0-12771

 

—_

 

 

LIBRARY
Michigan State
University
‘ A

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE
MMARM_'tQI99§ *

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ll

 

 

 

 

 

 

 

 

 

 

 

 

fiF—fi—I

MSU Is An Affirmative Action/Equal Opportunity Institution
cWDMS-OJ

 

 

 

THE RETINO-HYPOTHALAMO-SPINAL PATHWAY WHICH MEDIATES
GONADAL RESPONSES TO PHO'POPERIOD IN SYRIAN HAMSTERS

By

Timothy Grant Youngsz

A DISSERTATION

hi SubsmlttettiJ to

Mic an tate mverstty

in partial higllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Psychology and Neuroscience Program

1989

lib

C) L) LT) C.

ABSTRACT

THE RETINO-HYPO’l‘HALAMO-SPINAL PATHWAY WHICH MEDIATES
GONADAL RESPONSES TO PHOTOPERIOD IN SYRIAN HAMSTERS
by
Timothy Grant Youngstrom

The reproductive system of the Syrian hamster is sensitive to
changes in photoperiod. The pineal hormone, melatonin. is involved in
many of the changes observed in the reproductive system. However,
the interpretation of functional data obtained in the study of the
multisynaptic pathway mediating the circadian rhythm of melatonin
production and photoperiodism has relied on anatomical data obtained
from the nonphotoperiodic rat. The present study examined the
anatomical features of two segments of this circuit at the light
microscopic level in hamsters.

Intraocular injections of horseradish peroxidase conjugated to
Cholera toxin (CT-HRP) stained retinal ganglion cell (RGC) efferents in a
number of structures of the hypothalamus (including the
suprachiasmatic nuclei [SCN]). anterior thalamus and piriform cortex.
Retinal input to selected sites was confirmed by retrograde labeling of
86:03 following injections of fluorescent dye-labeled latex microspheres
aimed at these targets. Not all effects of photoperiod in the hamster are
mediated by pineal melatonin. Some extra-SON sites receiving retinal
input may participate in the expression of pineal-independent
photoperiodism.

A second segment of the circuit is thought to involve the
hypothalamic paraventricular nuclei (PVN). Iontophoretically applied
Phaseolus vulgaris-leucoagglutinin labeled efierent fibers of PVN
neurons with apparent terminal fields in the spinal cord, including the
intermediolateral cell column. Spinal cord injections of CT-HRP or
wheat germ agglutinin-HRP were employed to retrogradely label PVN
neurons; these injections labeled fibers of PVN origin that organized
themselves into four tracts on their path to the spinal cord.

Electrolytic lesions of one of these tracts at an extra-hypothalamic
site failed to prevent gonadal regression in short-day male hamsters. in
contrast to the effects of lesions made in the hypothalamus. Thus, PVN
fibers in the dorsal midline tract do not appear to be involved in
mediating the effects of photoperiod on the reproductive system of the
male Syrian hamster. Bilateral parasagittal knife cuts made lateral to
the PVN are effective in preventing gonadal regression in Syrian
hamsters suggesting that the lateral PVN-spinal cord efferents labeled
by HRP conjugates are involved in the control of photoperiodism in this
species.

To my best love MARTHA.
I'M DONE!
NOW we may go.

ACKNOWLEDGEMENTS

I would like to thank the members of my committee. Drs. Antonio
A. Nunez, Cheryl L. Sisk. John 1. Johnson and Charles D. Tweedle for
their assistance and suggestions. The author would like to thank Dr.
Nunez for his support and the use of his facilities. The assistance of a
number of people was essential for the completion of these projects and
the author is grateful for each person's many hours of participation.
Special thanks to those graduate students I have often turned to with
questions and were able to provide me with useful comments. assistance
and support, Ken Smithson and Barb Modney. A significant contribution
to those studies employing CT- and WGA-HRP was made by Dr. Mark L.
Weiss of M.S.U. who made most of the conjugates with me and Dr.
Richard R. Miselis of the University of Pennsylvania who provided
laboratory facilities to create some of these conjugates. Mark has also
been a valuable source of information and help in several of these
projects. The author is grateful to Betti Weiss for her skillful drawing of
one of the figures used in this paper. The author also thanks Dr. Glenn
1. Hatton for permitting the use of his equipment for the evaluation of
fluorescent dye labeled tissue and his service on the dissertation
committee. This work was supported in part by NIMH Grant MH 37877
to Dr. A.A. Nunez and by Biomedical Research Funds from Michigan
State University.

TABLE OF CONTENTS

LISTOFTABLES ....................................... v
LISTOFABBREVIATIONS ............................... xiii
INTRODUCTION ....................................... l

EXPERIMENT Ia-Retinal Efferent Projections to the Hypothalamus
and Basal Forebrain Identified by Anterogradely

Transported CT-HRP ............................. 7
METHOD ......................................... 9
RESULTS ........................................ 1 1

EXPERIMENT Ib-Retinal Efi'erent Projections to the Hypothalamus
and Basal Forebrain: Retinal Neurons Retrogradely

Labeled by Fluorescent Dye ....................... 33
METHOD ........................................ 34
RESULTS ....................................... .37
DISCUSSION ..................................... 43

EXPERIMENT Ila-The PVN Efferent Pathway to the Spinal Cord
Identified by Iontophoretically Applied Phaseolus vulgaris-

Leucoagglutinin ................................ 51
METHOD ........................................ 53
RESULTS ....................................... .54

EXPERIMENT lib-The PVN Efierent Pathway to the Spinal Cord:
Retrograde labeling of Neurons and Fibers in the PVN and

LHA. ........................................ 66
METHOD ........................................ 69
RESULTS ........................................ 7 1

III

DISCUSSION ..................................... 84
EXPERIMENT III-Do Extrahypothalamic Lesions of a PVN to Spinal
Cord Pathway aract 1) Block the Testicular Response to

Photoperiod? ................................. .89
METHOD ........................................ 91
RESULTS ....................................... .94
DISCUSSION ..................................... 94

GENERAL DISCUSSION ................................ 100
PHOTOPERIODISM ............................... 100
RGC INPUT TO THE BASAL FOREBRAIN ................ 102
PARAVENTRICULO-SPINAL CORD CONNECTIONS ......... 103
NEURAL CONTROL OF THE CIRCADIAN RHYTHM OF PINEAL

MEL SYNTHESIS .............................. 105
SPECIES SPECIFIC ANATOMY ....................... 107

BIBLIOGRAPHY ...................................... l 10

iv

LIST OF TABLES

Table l . Sites Receiving Retinal Ganglion Cell Efi'erents in the
Syrian Hamster ................................ 12

Table 2. Areas of the CNS Containing PHA-L Stained PVN
Efi‘erents ..................................... 6 2

Table 3. Areas of the Brain Containing Retrogradely Labeled
Neurons Following Injections of CT- and WGA-HRP into
the Spinal Cord ................................ 73

Table 4. Final Mean Width (1 sem) of Right Testis of Male Syrian
Hamsters in Experiment III ....................... 95

LIST OF FIGURES

Figure 1. Camera lucida drawings of coronal sections through the
brain of a Syrian hamster that received a unilateral
intraocular injection of CT-HRP (3 ml) into the left (L) eye.

The distribution of staining was asymmetrical with the
highest density of label in the hemisphere contralateral
(viewer's left) to the injected eye. All of the illustrations

are drawn to the same scale (Bar [A 8: E): 500 um). For a
definition of the abbreviations used in this and

subsequent figures refer to the List of Abbreviations. p.

xiii .......................................... 13

Figure 2. Photomicrograph (A) of a coronal section (50 mm)
through the caudal half of the SCN of an animal that
received a unilateral injection of CT-HRP into the left eye.
Labeling was present throughout each nucleus. including
the dorso-medial crescent (arrows) adjacent to the ventral
wall of the third ventricle which was relatively less densely
labeling. Some labeled fibers continued dorsally through
and around the SCN terminating in the overlying anterior
hypothalamic area (AHA; darkfield illumination; Bar =
200 um). The box in (B) outlines the area depicted in (A). . . .16

, Figure 3. Schematic drawing (A) of a coronal section through the
hypothalamus depicting the area (BOX) shown in (B). B -
CT-HRP labeled fibers passed dorsally through the SCN
and AHA to terminate in the AHA ventral to the PVN, while
other fibers coursed along the ventral PVN. Some fibers
ended along the ventral border while others ended in the
PVN or continued dorsally into the 21. Photomicrograph of
a coronal section (50 um) through the middle of the PVN.
(darkfield illumination; Bar = 200 um) ................ 19

Figure 4. Photomicrograph of the same coronal section shown in
Figure 2. Labeled fibers emerged from the lateral 0C and
coursed dorso-medially. Some fibers approached the
fornix and PVN but could not be traced into the PVN.
(darkfield illumination; Bar = 100 um) ................ 2 1

Figure 5. Photomicrograph of a coronal section through the OVLT.
CT-HRP labeled fibers emerged from the 0C contralateral
to the injected eye near the SCN and turned rostrally to
enter the POA. Some fibers continued rostrally into the
horizontal band of the DBB. ending in the vicinity of the
OVLT. A few continued dorsally in the vertical limb of the
DBB and ended near the MS (darkfield illumination; Bar =
100 m). The location of the fibers shown in (A) is
outlined in the schematic drawing (B) ................. 23

Figure 6. Camera lucida drawing of a tissue section cut in a
horizontal plane and stained for CT-HRP. Some labeled
fibers emerged from the lateral half of the optic chiasm
and coursed rostro-medially to the level of the OVLT. (Bar
= 200 um) ..................................... 26

Figure 7. A fascicle of CT-HRP labeled RGC etferents emerged from
the lateral DC at the level of the caudal SCN and turned
ventro-laterally, entering layer In of the Pyr (A; Bar =
200 pm). At higher magnification (B) individual fibers with
occasional varicosities along their length were seen
turning dorsally and ending in small swellings in layer 1A
(Bar = 20 um). (brightfield illumination) ............... 27

Figure 8. Photomicrograph (A) of a coronal section through the
SCN. In the rostral hypothalamus. CT-I-IRP labeled fibers
emerged from the lateral OC and passed laterally over the
SCN where some fibers terminated in the dorsal SON.

vii

This bundle of fiber terminated in the dorsal SON. This
bundle of fibers extended ventro-laterally into the Pyr
(darkfield illumination: Bar = 20 m). The box in (B)

indicates the location of the fibers shown in (A) .......... 29

Figure 9. Photomicrograph of a coronal section through the rostral

pole of the anterior thalamic nuclei. A small bundle of CT-
HRP labeled fibers joined the stria terminalis (ST) as they
continued from the LGN and ended in the TAM/TAV
(darkfield illumination; Bar = 50 pm; see Figure 10 for
location of fibers) ............................... 3 1

Figure 10. Photomicrograph of representative RGCs retrogradely

labeled with RLM. The cells are from an animal that

received a bilateral injection of RLM into the PVN (A) or

POA (B) and represent the range in size of somata that

were labeled. (epifluorescence illumination: Bar =20 um) . . .38

Figure l 1. Camera lucida drawing of a coronal section through the

hypothalamus at the level of bilateral RLM injections into

the area of the PVN/ZI in a Syrian hamster. The center of
each injection is marked by a dot. The total number of
labeled cells counted in the retinae of each animal ranged
between 2 and 12 ............................... 39

Figure 12. Schematic representation of the POA at the level of

bilateral RLM injection into the brain of a hamster. The
center of each injections is marked by a dot. The total
number of cells labeled in each animal ranged between 3

and 15. (Illustration adapted from Maragos, et aL. 1989) . . .41

Figure 13. Schematic representation of the brain of a Syrian

hamster at the level of bilateral RLM injections into the
rostral thalamus. The center of each injection is marked

VIII

by a dot. The total number of cells labeled in each animal
ranged between 2 and 9. (Illustration adapted from
Maragos, et aL. 1989) ............................ 42

Figure 1 4. Schematic drawings of the center of PHA-L injections to
show their location in the brain. The numbers in the
lower left corner refer to the case number (Cor -coronal
section: Sag-sagittal section). ...................... 55

Figure l 5 . Photomicrograph of a coronal section through the
center of a PHA-L injection site confined to the PVN (G16).
(brightfield illumination: Bar = 50 m) ................ 59

Figure 16. Camera lucida drawings of coronal sections through the
brain of a hamster (G 1 6) that received an iontophoretic
injection of PHA-L into the lateral PVN. Some fibers
emerging dorsally (la 81 1b) and ventro-laterally (2) from
the PVN and formed tracts that followed separate
trajectories to the periaqueductal grey (PAG). (Bar =
300 um) ...................................... 6 1

Figure 1 7. Schematic drawings of horizontal sections of the caudal
brainstem and spinal cord (cervical and rostral thoracic)
from an animal that received an iontophoretic injection of
PHA-L into the left (L) PVN (G 16). Drawings are arranged
dorsal (A) to ventral (C) and are composites of 4 sections
taken from each third of the spinal cord ............... 64

Figure 1 8. Schematic drawing demonstrating the vertebra. dorsal
spines and spinal cord of the upper thoracic and cervical
spinal column (a) of a Syrian hamster depicting the
location of the glass pipette-tipped 1 ml Hamilton syringe
(b) used to inject the HRP conjugates into the lateral
spinal cord (c) .................................. 7 0

IX

Figure 19. Camera lucida drawings of coronal sections through the
brain of a male Syrian hamster that received a unilateral
injection of WGA-HRP into the spinal cord between
vertebra C7 and T1 . Labeling in the brain was
predominantly ipsilateral to the injection site. Four tracts
(labeled '1'. '2'. '3' 8: '4' in (Bl) of stained fibers were seen
to emanate from the PVN. (Bar (A 81 D] = 500 um) ........ 74

Figure 20. Photomicrograph of a coronal section through the
rostral PVN of a Syrian hamster. The dorsal and ventro-
lateral PVN ipsilateral to an injection of WGA-HRP into the
spinal cord contained retrogradely labeled neurons and
fibers. had 3 was prominent as fibers from the PVN
course ventro-laterally around the fornix. A few fibers
were seen over the roof of the third ventricle (”act 1),
lateral to the PVN ('Iract 2) and ventral to the PVN “Tact
4). Scattered labeled cells were also seen in the ipsilateral
hypothalamus and contralateral PVN. (darkfield
illumination; Bar = 200 um) ........................ 75

Figure 2 l. Photomicrograph of a coronal section near the middle of
the PVN. A collection of retrogradely labeled cells were
observed in the dorsal and ventro-lateral PVN. Fibers of
'h'act 3 remained evident around the fornix. Some fibers
of the dorsal cell group were traced to the area over the
third ventricle. (darkfield illumination; Bar = 200 um) ..... 76

Figure 22. Photomicrograph of a coronal section through the
caudal third of the PVN showing the nucleus ipsilateral to
the injection site. The lateral parvocellular subdivision (as
defined for the rat by Swanson & Kuypers[19801) was
intensely labeled and few individual cells could be seen
distinctly from the background of labeling. The ventro-
medial portion of the PVN contained fewer labeled cells
than the lateral division while the fewest number of cells

X

were observed in the medial PVN. (darkfield illumination:
Bar = 200 um) .................................. 78

Figure 23. Photomicrograph of a coronal section through the
caudal PVN. WGA-HRP labeled cells were present in the
PVN and scattered through the LHA. At this level, fibers
of Tract 2 split into two bundles. One fascicle turned
ventrally (b) to join Met 3 in the caudal hypothalamus
while the second turned dorso-laterally over the CP (a).
(darkfield illumination;Bar=200 um) .................. 79

Figure 24. Photomicrograph of a coronal section near the level of
Figure 21. The PVN is just out of the field depicted
dorsally and to the viewer's left. Numerous fibers emerged
from the ventral border of the PVN and surrounded or
passed through the fornix. (darkfield illumination: Bar =
50 um) ....................................... 81

Figure 25. Photomicrograph of a coronal section near the level of
Figure 20, showing labeled fibers of Tract 4 near the wall
of the third ventricle. (darkfield illumination; Bar = 50 um)
83

Figure 26. Mean width of the right testis (measured trans-scrotally)
of animals receiving electrolytic lesions aimed at the PAG
or sham—lesioned. Animals were lesioned (LES; n = 8) or
sham lesioned (SDS; n = 7) and placed in a 8L: 16D
photoperiod. or sham lesioned and placed in a 16L:8D
photoperiod (long-day sham [LDS]; n = 4). The mean
testicular width for each short photoperiod housed groups
was significantly smaller than that of the LDS group by
the end of Week 8 and remained so until the end of the
experiment .................................... 96

Figure 27. Mean body weight of animals receiving electrolytic
lesions aimed at the PAG or sham-lesioned. Animals were
lesioned (LES; n = 8) or sham lesioned (SDS; n = 7) and
placed in a 8L:16D photoperiod. or sham lesioned and
placed in a 16L: 8D photoperiod (long-day sham [LDS]; n =
4). A nonsignificant decline in the mean body weight was
observed in both short-photoperiod housed groups. Week
zero represents the mean presurgical body weight of each
group ........................................ 97

LIST OF ABBREVIATIONS

3V Third Ventricle

AC Anterior Commissure

ADT Anterior Dorsal Thalamic
Nucleus

AHA Anterior Hypothalamic
Area

AMB Nucleus Amblguus

AP Area Postrema

BLX Olfactory Bulbectomy

BNST Bed Nucleus of the Stria
Terminalis

C7 Cervical Vertebra 7

Cor Coronal

CP Cerebral Peduncle

CT-HRP Cholera Toxin-Horseradish
Peroxidase

DBB Diagonal Band of Broca

DMN Dorsomedial Nucleus

DMNpc Dorsomedial Nucleus. pars
compacta

DMT Dorsomedial Terminal
Nucleus

DNV Dorsal Nucleus of the
Vagus

DV dorso-ventral

F Fornix

FR Fasciculus Retroflexus

FSH Follicle Stimulating
Hormone

H Habenula

IC Internal Ca sule

IGL Inter enic ate Leaflet

III Thir Ventricle

IML Intermediolateral Cell
Column

L Left

LC Locus Coeruleus

LD Light-Dark

LGN Lateral Geniculate Nucleus

LH Luteinizing Hormone

LHA Lateral Hypothalamic Area

LL Lateral Lemniscus

LOT Lateral Olfactory 'h'act

LPoA Lateral Preoptic Area

LTN Lateral Terminal Nucleus

MEL Melatonin

MFB Medial Forebrain Bundle

MH(a) Medial Habenula

ML media-lateral

MoV

MPoN
MS
MT

NST

OC
ON

OVLT

PAG
PB

PEG
PHA-L

SCG
SCN
SM
SN
SO
SOL

SON
ST

'n'igeminal Motor Nucleus
Medial Preoptic Area
Medial Preoptic Nucleus
Medial Septum
Mammillothalamic 'IYact
Medial Terminal Nucleus
Nucleus of the Solitary
'h‘act
Optic Chiasm
Optic Nerve
Optic Tract
Or anum Vasculosum of
e Lamina Terminalis
Periaqueductal Grey
Parabrachial Nucleus
Posterior Commissure
Polyeth lene Glycol
Phaseo vulgaris-
leucoagglutinin
Pons
Preoptic Area
Prolactin
Paratenial Nucleus
H othalamic
araventricular Nucleus
Thalamic Paraventricular
Nucleus
Piriform Cortex
Piriform Cortex. layer I
rostro-caudal
Retinal Ganglion Cells
Retino-Hypothalamic 'h‘act
Rhodamine Labeled Latex
Microspheres
Raphe Pallidus Nucleus
Raphe Magnus Nucleus
Raphe Obscurus Nucleus
Sagittal
Superior Cervical Ganglion
Suprachiasmatic Nuclei
Stria Medullaris
Substantia Nigra

' Subcommissural Organ

Nucleus of the Solitary
'h'act

Supraoptic Nucleus

Stria Terminalis

Thoracic Vertebra 2

TAM Antero-Medial Thalamic
Nucleus

TAV Antero-Ventral Thalamic
Nucleus

Tu Olfacto Tubercle

V III Vestib ocochlear Nerve

VMN Ventromedial Nucleus

Vs Triggrlninal Sensory fiact

WGA-HRP eat Germ
Agglutinin-Horseradish
Peroxidase

21 Zone Incerta

xiv

INTRODUCTION

The reproductive systems of a large number of seasonally
breeding mammals are sensitive to changes in daylength (photoperiod).
The response of the reproductive system to changes in photoperiod
results from an interaction of the endogenous circadian system and the
light-dark (LD) cycle (Elliot 8: Goldman. 1981: Bittman. 1984). In the
Syrian hamster. prolonged exposure to short photoperiods (< 12.5 hr)
induces a marked alteration in the reproductive system expressed. in
part. as gonadal regression and sexual quiescence (Gaston 81 Menaker.
1967 ; Seegal & Goldman. 1975). The pineal hormone. melatonin (MEL).
is intimately involved in the control of these photoperiodic responses
(Bittman & Karsch. 1984; Bittman. et aL. 1985; Carter & Goldman.
1983; Goldman 81 Darrow. 1983; Bartness & Goldman. 1988). The
concentration of MEL in the pineal gland and in the systemic
circulation displays a circadian rhythm with peak levels occurring
during the dark phase (night) of the LD cycle (Tarnarkin. et aL. 1979;
Bittman. et aL. 1983; Rollag, et aL. 1978; Klein. 1979). The duration of
elevated MEL is inversely proportional to the prevailing photoperiod.
Long durations of elevated MEL are associated with short photoperiods
while long photoperiods are associated with MEL elevated for shorter
intervals (Rollag, et aL. 1978; Goldman. et aL. 1981; Illnerova &
Vanecek, 1 982b). Of the several parameters of MEL release that may
provide information to the nervous system about photoperiod. duration
of elevated MEL has been shown to be a critical factor determining the
reproductive state of the Siberian hamster. a long photoperiod breeder
(Carter 8: Goldman. 1983; Bartness & Goldman. 1988) and the sheep. a
short photoperiod breeder (Bittman. et aL. 1985).

1

2

The effects of light on pineal MEL concentration appear to involve
direct retina-hypothalamic tract (RHT) input to the hypothalamic
suprachiasmatic nuclei (SCN; Tamarkin. et aL. 1985; Klein 82 Moore.
1979). The role of the SCN in the pathway controlling pineal MEL
production is thought to be primarily that of a circadian oscillator.
generating a circadian rhythm of neural activity that controls MEL
synthesis and release (Goldman & Darrow. 1983; Klein. 1979). The SCN
and the RHT also mediate the entrainment of the MEL rhythm to the
prevailing photoperiod (Klein & Moore. 1979; Rollag & Niswender.
1976). Exposure of the animal to light at night. when MEL levels are
normally elevated. results in a precipitous drop in both pineal gland
and serum MEL concentrations to daytime levels (Tamarkin. et aL.
1979; Vanecek & Illnerova. 1982; Illnerova & Vanecek, 1982a.b).
Depending upon the phase of the MEL rhythm at the time of light
exposure. this manipulation results in a phase advance. a phase delay
or no phase shifting of the MEL rhythm (Illnerova 8: Vanecek, 1982a.b)
in animals released into constant dark conditions. The absence of
phase shifts in the MEL rhythm at certain times in the cycle suggests
that. independent of light's efi‘ect on the circadian oscillator. light also
has a direct effect on the neural circuit mediating the elevated level of
MEL production in the pineal gland. Recent work in the rat and
hamster. using intraocular injections of Cholera toxin conjugated to
horseradish peroxidase (CT-HRP). has revealed a more extensive
distribution of retinal efferents in the brain than previously described
(Levine. et aL. 1986; Johnson. et a1. 1988). Although the SCN of the
hamster is the principal target of RHT input. the pattern of retinal input

3

to the remainder of the hypothalamus and forebrain (Johnson. et al.
1988; Pickard 8r Silverman. 1981) deserves further attention.

A model of the neural circuit involved in the control of the pineal
gland and MEL production has been proposed. It consists of a
multisynaptic pathway connecting the SCN to the pineal gland by way
of the hypothalamic paraventricular nuclei (PVN). intermediolateral cell
column (IML) of the spinal cord and the superior cervical ganglion (SCG;
Tamarkin. et al.. 1985). The efferent connections of the SCN have been
examined in the rat using iontophoretic injections of Phaseolus vulgaris-
leucoagglutinin (PHA—L). Projections of the SCN include a prominent
group of fibers passing dorsally through the medial PVN where some of
the fibers terminate (Watts. Swanson & Sanchez-Watts. 1987). In the
Syrian hamster. horizontal knife cuts placed between the SCN and PVN
abolish short-photoperiod-induced collapse of the reproductive system
of males and females (Badura. et aL. 1987; Eskes & Rusak. 1985:
Inouye 8r 'lurek. 1986; Nunez. et al.. 1985). Electrolytic or axon-
sparing chemical lesions of PVN neurons block the nocturnal rise in
MEL synthesis and photoperiodic responses (Bittman 81 Lehman. 1987:
Brown. et al.. 1988; Eskes & Rusak. 1985; Lehman. et aL. 1984:
Pickard 8: 'Iurek. 1983; Hastings 8: Herbert. 1986). The available
evidence suggests that the PVN is a relay nucleus in the pathway
mediating photoperiodism. and that the pertinent information is sent
through PVN descending projections to the spinal cord. In rats. PHA-L
labeled fibers are found in the IML of the spinal cord following injections
of this lectin into the PVN (Luiten. et aL. 1985). In hamsters. injections
of wheat germ agglutinin conjugated to horseradish peroxidase (WGA-
HRP) into the spinal cord between cervical vertebra 7 (C7) and thoracic

4

vertebra 2 (Ta) label neurons throughout the rostro-caudal extent of the
PVN (Don Carlos 8: Finkelstein. 1987). However. functional evidence
supporting the direct involvement of a mono-synaptic PVN-spinal cord
connection in the neural circuit mediating the circadian rhythm of
pineal MEL synthesis and release is currently unavailable. Thus,
brainstem nuclei receiving PVN efi'erents and projecting to the spinal
cord may participate in the pathway. either partially or to the exclusion
of PVN efi'erents terminating in the spinal cord. In hamsters.
preganglionic fibers originating at the level of CB-Ts in the IML terminate
in the SCG (Reuss. et aL. 1989). Some fibers from SCG neurons. in
turn. terminate in the pineal gland where sympathetic activation results
in elevated MEL production (Bower 8: Zigrnond. 1982; Wurtrnan. et aL.
1964: Kappers. 1960).

In the rat. two PVN-to-spinal cord pathways have been described
(Luiten. et al.. 1985). One path exits the PVN dorsally (Tract 1). and
projects caudally. running near the roof of the posterior hypothalamic
portion of the third ventricle (III). The second set of PVN efferent fibers
(Tract 2) exits the PVN laterally passing beneath the zona incerta (ZI)
and dorsal to the internal capsule (IC) and medial forebrain bundle
(MFB). In the mesencephalic Periaqueductal grey (PAG). and continuing
caudally to the level of the area postrema (AP). fibers making up Tract 1
appear to turn ventro-laterally joining 'IYact 2 in the lateral lemniscus
(LL) of the brain stem (Luiten. et al.. 1985). Recent evidence from
experiments using Syrian hamsters indicates that male and female
hamsters respond differently to horizontal knife cuts placed dorsal to
the PVN. In the male hamster. these knife cuts. like cuts placed
between the SCN and PVN. prevent gonadal regression perhaps by

5

disrupting Tract l (Nunez. et aL. 1985; Inouye & Turek. 1986). In
contrast. similarly placed cuts in females do not prevent short-day-
induced anestrus (Badura. et al.. 1989).

In male hamsters. horizontal knife cuts dorsal to the PVN are
associated with increased levels of plasma follicle stimulating hormone
(FSH; Badura. et aL. 1988). Similar changes in FSH levels have been
observed following transection of the lateral olfactory tract (LOT) or
removal of the olfactory bulbs (BLX) in the male Syrian hamster
(Clancy. et aL. 1986; Pieper. et aL. 1989). Exposure of BLX hamsters to
short photoperiods reduces both luteinizing hormone (LH) and FSH
levels. but FSH levels remain above those of control animals and
apparently at sufficiently high levels so as to maintain large testes.
Thus. BLX and knife cuts dorsal to the PVN appear to remove a tonic
inhibition on FSH secretion. but do not abolish the photoperiodic
control of release of the hormone. A similar change in FSH
concentrations. if present in the female. may not be sufficient for the
production of estrous cycles during extended exposure to a short
photoperiod. In the normal 4-day estrous cycle of the female hamster.
serum FSH levels are low on diestrus 2 and early on proestrus (Lisk.
1985). A large surge in FSH secretion occurs on the afternoon of
proestrus and is associated with increased LH release. A second surge
in FSH release occurs on estrus. which is involved in initiating the
development of the next set of follicles. LH and FSH levels have not yet
been measured in female hamsters possessing horizontal knife cuts
dorsal to the PVN. however. continuously elevated levels of FSH may be
incapable of maintaining estrous cycles in the face of reduced levels of
serum LH associated with exposure to short photoperiods.

6

Studies of the anatomical pathway mediating photoperiodism and
the circadian rhythm of pineal MEL production in the Syrian hamster
have had to rely upon anatomical data obtained from the
nonphotoperiodic rat for the interpretation of the effects of lesions and
knife cuts. While there is evidence to suggest that the pathways
controlling MEL production are similar across species. it is.
nevertheless. important that the anatomy of the pathway for the
hamster be described and the postulated neural connections verified.
The present series of experiments examined the anatomical features of
two segments of this pathway for the hamster at the light microscopic
level. First. the distribution of retinal ganglion cell efferents to the
forebrain was examined. Second. the trajectory of PVN-spinal efi'erents
through the brain was described and the location of neurons in the
hypothalamus with projections that terminate in the spinal cord
identified. A final experiment assessed the role played in
photoperiodism by PVN efferent fibers that pass through the PAG in
their way to the spinal cord.

EXPERIMENT Ia-Retinal Efferent Projections to the Hypothalamus and
Basal Forebrain Identified by Anterogradely 'h'ansported
CT-HRP.

A RHT projecting primarily to the SCN has been found in a large
number of mammalian species (Moore. 1978; Cassone. et aL. 1988).
While the SCN is the principal recipient of RHT input. it is increasingly
evident that other areas of the hypothalamus and basal forebrain
receive direct retinal ganglion cell (RGC) input to varying degrees.
Recent studies employing HRP as a tract-tracing tool have reported that
the RHT projects to several areas of the hypothalamus. including the
PVN. and suggest that the piriform cortex (Pyr). preoptic area (POA) and
rostral thalamus are also targets of RGC efferents (Pickard & Silverman.
1981: Johnson. et al.. 1988: Levine. et al.. 1986: ltaya. et al.. 1981:
ltaya. et al.. 1986).

Several pineal-independent effects of photoperiod have been
found to occur in the Syrian hamster (see Bartness & Wade. 1985:
Badura 81 Nunez. 1989) and some of the areas of the forebrain receiving
direct retinal input may be involved in the control of these photoperiod-
sensitive features. A recent report noted a pineal-independent decline
in behavioral sensitivity to steroid hormones in females maintained in a
short photoperiod (Badura 8r Nunez. 1989). In the female. the medial
POA provides an inhibitory influence on reproductive behavior (Pfafi’ &
Modianos. 1985). and in both male and female hamsters. the POA
receives direct RGC input (Johnson. et al.. 1988). Also in both sexes.
pineal-independent changes in feed efficiency and metabolism occur
(Bartness 8r Wade. 1985) and some sites in the hypothalamus involved

8

in the control of energy balance. including the PVN and lateral
hypothalamic area (LHA). have been found to contain RGC efi'erents.

Cholera toxin-HRP is a more sensitive neuroanatomical tract-
tracer than free (unconjugated) HRP. In the rat and hamster.
intraocular injections of CT-HRP have revealed a pattern of labeled RHT
fibers more widely distributed than had previously been reported using
other tract-tracing methods (Levine. et al.. 1986). The distribution of
RGC efferents outside the SCN on the Syrian hamster remains
controversial. For instance. Pickard 81 Silverman (1981) reported
evidence of input to the diagonal band of Broca (DBB) rostrally. Pyr
laterally and the dorsomedial (DMN) and ventromedial (VMN) nuclei
caudally. More recently. a study using CT-I-IRP did not describe RGC
input to some of these terminal sites for RGC efferents (e.g. - to the
DMN; Johnson. et al.. 1988) or provided a different interpretation of
their results (e. g. - input to the amygdala) compared to those of Pickard
and Silverman (1981; e.g. - fibers in the piriform cortex). Johnson. et
al.. (1988) also noted a direct projection to the anterior thalamus but
could not detect the path taken by the retinal fibers to this site. The
present experiment examined the RHT in the Syrian hamster. focusing
on the extra-SCN projections using intraocular injections of CT-HRP.
Experiment Ia. was completed shortly before the report of Johnson. et
al. (1988). containing similar results. appeared in the literature.

9
moo

Mala

Adult male (n = 10) and female (n = 10) Syrian hamsters
(Mesocricetus auratus; 120 - 200 g) were obtained from Charles Rivers
Laboratories and group-housed (n = 6 males or females/cage; cage size-
.16.5 x 30.5 x 36 cm) in a stimulatory photoperiod (l6 hour/8 hour LD
cycle). Food (Wayne Mouse Breeder Blox) and water were available ad
libitum.

m

Animals were anesthetized with Equithesin (4.5 ml/ kg) and
received unilateral (n = 1 7) or bilateral (n = 3) intraocular injections (1 .5
- 3.0 ul) of CT-HRP (OBS-0.49%) using a IO-ul Hamilton syringe. The
conjugate was prepared in the laboratory of Dr. R.R. Miselis (University
of Pennsylvania) using methods previously described (Shapiro &
Miselis. 1985).

P i n 1

Following survival periods of 12-48 hrs. the animals were deeply
anesthetized with Equithesin and perfused transcardially (using an IV
cannula) with heparinized (2 U/ ml) 0.05 M phosphate buffered (pH 7.4)
physiological saline rinse (250 ml over 10 min) followed by 0.05 M
phosphate buffered (pH 7 .4) fixative containing 0.5 % (W/v)
paraformaldehyde and 1 .25 % (V/v) gluteraldehyde infused at the same
rate. All solutions were infused using a peristaltic pump (Master-flex;
Cole-Parmer). The brains were post-fixed overnight at 4'0 in the same
fixative. Rozen coronal (n = 19) or horizontal (n = 1) sections (50 m)
were collected in cold 0. 1 M phosphate buffer (pH 7.4) and
subsequently processed using the tetramethylbenzidine protocol of

10

Mesulam (1982). The sections were arranged into three sets of adjacent
serial sections. each set consisting of every third section. All sections
were mounted on previously gelatinized slides. 'leo sets were air dried.
cleared in xylenes and coverslipped using Permount. One set of tissue
was air dried and then counterstained with cresylecht violet prior to
coverslipping.

figuration

The uncounterstained tissue was evaluated immediately using
brightfield and darkfield optics. Photographs of reaction product in
areas of interest in the brain were made at this time. Subsequently. the
labeled fibers were plotted on camera lucida drawings of sections from
representative cases. Structures of the brain were identified by
referring to the adjacent cresylecht violet stained sections.

W

The tissue was examined for evidence of hemal or trans-synaptic
transport of the CT-HRP. First. the magnocellular PVN and supraoptic
nuclei (SON) were examined to determine if retrogradely labeled cells
were present. In a number of species of rodents. magnocellular
neurons within these two nuclei are retrogradely labeled when HRP is
injected into the systemic circulation (Broadwell & Brightman. 1976;
Youngstrom 82 Nunez. 1987). Second. the primary visual cortex was
examined for evidence of orthogradely transported CT-HRP. Labeling in
this area of cortex would be indicative of trans-synaptic transport in at
least one target nucleus of the optic tract (Itaya & Van Hoesen. 1982).

The distribution of labeling in the basal forebrain from the level of
the organum vasculosum of the lamina terminalis (OVLT) through the
posterior hypothalamus is depicted schematically in Figure l and listed
in Table 1. There was no evidence of trans-synaptic transport of CT-
HRP in the visual cortex of any animal and the distribution of CT-HRP
staining was similar regardless of sex. The intensity of staining
increased with longer survival but the pattern remained unchanged.
Except for the SCN. this pattern of labeling was typified by the
asymmetrical distribution of staining. with structures contralateral to
the injected eye more intensely labeled than their ipsilateral
counterparts following unilateral injections.

SCN - The SCN were the most densely labeled structures of the
basal forebrain (Figures lD-F & 2) and the pattern of staining was
essentially the same as reported previously, so it will be reviewed
briefly. Near the rostral pole of the SCN. each nucleus. separated from
one-another by the third ventricle. possessed nearly equal amounts of
stained fibers following a unilateral injection. Proceeding caudally. as
the floor of the third ventricle separated from the optic chiasm (DC) the
medial borders of the nuclei met at the midline. A less intensely
stained dorso-medial crescent became evident in each nucleus (Figure
1E). At about the same level the intensity of stain in the ventral half of
the nuclei increased and the pattern of labeling became slightly
asymmetrical. the nucleus contralateral to the injected eye being more
intensely stained. In the caudal one-third of the nucleus, the
asymmetry persisted to a point near the caudal pole. where the

12

Table 1

Sites Receiving Retinal Ganglion Cell Efi'erents in the Syrian Hamster

MXQDALA Suprachiasmatic Nucleus
Supraoptic Nucleus
Medial Arnygdaloid Ventromedial Nucleus
Nucleus flSENCEPwN
QQRTEX Dorsomedial Terminal
Nucleus
Piriforrn Cortex. layer I Lateral Terminal Nucleus
‘ Medial Terminal Nucleus
HYQIMUSZPREQPTIC Pretectum
Superior Colliculus
Anterior Hypothalamic
Area W
Bed Nucleus of the Stria
Terminalis Antero-Medtal Thalamic
Diagonal Band of Broca Nucleus
Dorsomedial Nucleus Antero-Ventral Thalamic
Lateral Hypothalamic Nucleus
Area Lateral Geniculate
Lateral Preoptic Area Nucleus
Medial Preoptic Area Zona Incerta

Paraventricular Nucleus
Retrochiasmatic Area

13

Figure 1 . Camera lucida drawings of coronal sections through the brain
of a Syrian hamster that received a unilateral intraocular injection of CT-
HRP (3 pl) into the left (L) eye. The distribution of staining was
asymmetrical with the highest density of label in the hemisphere
contralateral (viewer's left) to the injected eye. All of the illustrations are
drawn to the same scale (Bar [A & E]: 500 pm). For a definition of the
abbreviations used in this and subsequent figures refer to the List of
Abbreviations. p. xiii.

l4

 

 

 

 

Figure l

l5

 

 

Figure 1 (Continued)

16

Figure 2. Photomicrograph (A) of a coronal section (50 um) through the
caudal half of the SCN of an animal that received a unilateral injection of
CT-HRP into the left eye. Labeling was present throughout each nucleus.
including the dorso-medial crescent (arrows) adjacent to the ventral wall
of the third ventricle which was relatively less densely labeling. Some
labeled fibers continued dorsally through and around the SCN
terminating in the overlying anterior hypothalamic area (AHA; darkfield
illumination; Bar = 200 m). The box in (B) outlines the area depicted in

(A).

17

 

 

 

 

 

Figure 2

18

declining intensity of staining prevented identification of any
asymmetry.

Extra-SCN Hypothalamus - A prominent feature of the RHT was a
bundle of fibers passing through the SCN (Figures 1E.F 8r 2). This
bundle first became evident as it emerged dorsally from the middle one-
third of the SCN. increasing in prominence in the caudal one-third of
the nuclei. After the SCN. the anterior hypothalamic area (AHA)
possessed the highest density of stained fibers in the hypothalamus.
Dorsal to the SCN. the fibers spread dorso-laterally into the AHA. At a
level near the rostro-caudal mid-point of the PVN. some RHT fibers
approached the nucleus ventrally (sub—PVN area) and traced a lateral
path along the boundary (Figures 1E.F 81 3). Other fibers penetrated
the PVN. where a few fibers appeared to terminate. while the remainder
continue dorsally into the Z1. Those fibers terminating in the PVN were
located primarily in areas corresponding to the periventricular-. lateral-
and dorsal-parvocellular regions of the PVNas described for the rat
(Swanson 81 Kuypers. 1980) with the heaviest concentration of reaction
product found in the ventro-lateral PVN. Additional fibers emerged
from the lateral OC adjacent to the SCN and contributed to the RHT
input to the lateral AHA (Figures 1E.F. 2 8: 4).

At the level of the caudal SCN. stained fibers emerged from the
lateral half of the optic tract (OT) coursing dorso-medially toward the
fomix (F) and lateral PVN (Figures lE.F & 4). It was not clear if these
fibers penetrated the PVN or were restricted to the LHA. Caudally. at
the level of the retrochiasmatic area. as the OT continued dorso-
iaterally. fibers emerged nearly horizontally from the tract.

19

Figure 3. Schematic drawing (A) of a coronal section through the
hypothalamus depicting the area (BOX) shown in (B). B - CT-HRP
labeled fibers passed dorsally through the SCN and AHA to terminate in
the AHA ventral to the PVN. while other fibers coursed along the ventral
PVN. Some fibers ended along the ventral border while others ended in
the PVN or continued dorsally into the 21. Photomicrograph of a coronal
section (50 um) through the middle of the PVN. (darkfield illumination;
Bar = 200 um)

20

 

 

 

 

 

Figure 3

21

 

Figure 4. Photomicrograph of the same coronal section shown in Figure

2. Labeled fibers emerged from the lateral OC and coursed dorso-
medially. Some fibers approached the fornix and PVN but could not be
traced into the PVN. (darkfield illumination; Bar = 100 um)

22

coursing medially toward the Z1 and PVN (Figures 1F.G). Some fibers
passed through the dorsal half of the cerebral peduncle (CP) on their
way to the dorsal hypothalamus. This input to the dorsal
hypothalamus was quite sparse and was absent in the hypothalamus
ipsilateral to the injected eye. At this same level stained fibers were also
seen in the medial hypothalamus. The fibers were primarily confined to
the region between the VMN and DMN bilaterally and the third ventricle
(Figure 16). although individual fibers were also evident in the lateral
hypothalamus. A small plexus of fibers appeared to terminate in the
DMN pars compacta (DMNPC; Figure 11-1). The number of stained fibers
in the hypothalamus declined rapidly in tissue caudal to this level.
Preoptic Area - At the level of the anterior SCN, a plexus of fibers
emerged from the lateral OC and turned rostrally. entering the ventral
POA contralateral to the injected eye (Figures 1C.D). Stained fibers
were found in the medial and lateral POA and medial preoptic nucleus
(Figures 1B.C). Continuing rostrally. some fibers were evident near the
LOT and olfactory tubercle (Tu) but did not enter these structures.
Most of the fibers reaching the level of the organum vasculosum of the
lamina terminalis (OVLT). entered the horizontal limb of the DBB. and
proceeded medially toward the OVLT (Figures 1A. 5). Of the fibers
traced rostrally to this level. many ended in the transition area between
the horizontal and vertical limbs of the DBB adjacent to the OVLT. while
a few remaining fibers continued dorsally to end near the ventral
boundary of the medial septum (MS). On the side ipsilateral to the
injected eye. fibers entering the POA emerged from the OC at a point
further rostral relative to contralateral fibers. The ipsilateral fibers

23

Figure 5. Photomicrograph of a coronal section through the OVLT. CT-
HRP labeled fibers emerged from the OC contralateral to the injected eye
near the SCN and turned rostrally to enter the POA Some fibers
continued rostrally into the horizontal band of the DBB. ending in the
vicinity of the OVLT. A few continued dorsally in the vertical limb of the
DBB and ended near the MS (darkfield illumination; Bar = 100 um). The
location of the fibers shown in (A) is outlined in the schematic drawing
(B).

24

 

 

 

 

 

25

were similarly distributed but did not extend as far laterally and were
fewer in number. The course of all rostrally projecting fibers through
the POA was particularly salient in tissue sectioned in a horizontal
plane and material cut in this fashion facilitated the interpretation of
the trajectory of these fibers (Figure 6).

Piriform Cortex - Approximately 200 um caudal to the Tu. laterally
projecting fibers emerged from the OC forming a fascicle along the
ventral surface of the brain and extended laterally toward the rhinal
fissure (Figures 13—6. 7). This set of axons was generally confined to
layer In of the Pyr (Devor. 1976; Haberly & Feig. 1983). Fibers were
never observed in the pyramidal cell layer of the Pyr. Moving caudally
through this region of the brain. the fascicle persisted to the level of the
caudal SCN. At points along the fascicle. fibers turned and entered the
overlying layer 1a where many ended in terminal swellings (Figure 7B).
Occasionally. individual stained fibers were observed in the medial
amygdaloid nucleus and anterior amygdaloid area.

In the hypothalamus. intermingled with the fibers destined for the
Pyr were other fibers that appeared to terminate in the dorsal SON.
Some fibers possessed varicosities suggestive of en passant terminals
while others appeared to give off short branches that ended in swellings
(Figure 8).

Extra-geniculate Thalamus - The antero-medial (TAM) and antero-
ventral (TAV) thalamus received direct retinal input (Figure 9). The
small bundle of stained fibers reaching this level appeared as a rostral
continuation of the large RGC input to the lateral geniculate nucleus
(LGN; Figure lC-F). At the level of the rostral LGN. some labeled fibers
separated from those terminating in the nucleus and continued into the

26

Rosita)

 

 

 

 

 

Figure 6. Camera lucida drawing of a tissue section cut in a horizontal
plane and stained fOr CT-HRP. Some labeled fibers emerged from the
lateral half of the optic chiasm and coursed rostro-medially to the level
of the OVLT. (Bar = 200 um)

27

Figure 7. A fascicle of CT-HRP labeled RGC efferents emerged from the
lateral DC at the level of the caudal SCN and turned ventro-laterally.
entering layer 1a of the Pyr (A; Bar = 200 m). At higher magnification
(B) individual fibers with occasional varicosities along their length were
seen turning dorsally and ending in small swellings in layer 1A (Bar =
20 um). (brightfield illumination)

28

 

 

Figure 7

29

Figure 8. Photomicrograph (A) of a coronal section through the SON. In
the rostral hypothalamus. CT-HRP labeled fibers emerged from the
lateral OC and passed laterally over the SON where some fibers
terminated in the dorsal SON. This bundle of fibers extended ventro-
laterally into the Pyr (darkfield illumination; Bar = 20 m). The box in (B)
indicates the location of the fibers shown in (A).

30

 

 

Figure 8

31

 

Figure 9. Photomicrograph of a coronal section through the rostral pole
of the anterior thalamic nuclei. A small bundle of CT-HRP labeled
fibers joined the stria terminalis (ST) as they continued from the LGN
and ended in the TAM/TAV (darkfield illumination: Bar = 50 um: see
Figure 1C for location of fibers)

32

medial ST taking a position near its medial border on the floor of the
lateral ventricle. The small set of axons coursed forward to the rostral
thalamus where it left the ST ventro-medially. The fibers terminated in
a triangle of tissue. containing the TAM and TAV. formed between the
ST and stria medullaris (SM).

EXPERIMENT Ib-Retinal Efi'erent Projections to the Hypothalamus and
Basal Forebrain: Retinal Neurons Retrogradely Labeled by
Fluorescent Dye.

As described in Experiment Ia. several areas of the medial
forebrain receive direct RGC input. In the hypothalamus. the AHA.
DMN. PVN and SON. in addition to the SCN. receive RHT input. as does
the POA rostrally and the Pyr laterally. Dorsally. the 21 and the rostral
pole of the thalamus (TAM/TAV) also receive optic input. Despite the
absence of staining in the visual cortex that might indicate trans-
synaptic transport, the concern remains that the observations in
Experiment Ia were the result of multisynaptic transport. To
corroborate these data. selected sites found to contain CT-HRP stained
fibers in Experiment Ia received micropressure injections of rhodamine
fluorescent dye labeled latex microspheres (RLM; Katz. et aL. 1984).
The retinae were then examined for evidence of retrogradely labeled

ganglion cells.

33

34

METHOD
Male
Animal and housing conditions were identical to Experiment Ia.
Sam

Adult male (n = 8) and female hamsters (n = 5) received bilateral
stereotaxically placed injections of RLM into one of several areas of the
brain found to possess stained fibers in Experiment la. The sites
chosen to receive injections were: 1) POA/DBB. 2) TAM/TAV. 3)
PVN/ZI. Based on the data from Experiment Ia. each area that was
selected represented the terminus of a retinal efferent path. The
interpretation of the results from injections into these sites would be
less likely to be confounded by RGC fibers-of-passage located near the
injection taking up the tracer dye. Also. the pattern of CT-HRP labeling
in the selected sites following ocular injections was relatively discrete.
It was anticipated that the relatively circumscribed distribution of RGC
fibers would maximize the opportunity for uptake and retrograde
transport to RGC. No of these sites (POA/DBB 8r PVN/ZI) were of
interest as structures implicated in the control of sexual behavior and
photoperiodism. In the hamster. the TAM/TAV had been found to
contain RGC efferents once previously (J ohnson. et al.. 1988) but the
trajectory of the efferents to this area of the brain had not been
described until the present report (Experiment Ia). These structures
received injections to confirm the retinal efferent projection to each.

In preliminary cases (n = 10 males) RLM injections spread to
include other structures receiving RGC input (e.g.. LGN. SON. AHA) or
invaded the optic nerve or chiasm. These cases were used to assess the

ability of RLM to retrogradely label ganglion cells.

35

Animals were anesthetized with Equithesin (4.5 ml/ kg) and
placed in the stereotaxic apparatus. The skull was exposed and a burr
hole drilled. The stereotaidc coordinates were empirically determined
for each site receiving an injection (from Bregrna. POA/DBB: rostro-
caudal (RC) +2.2 mm. medic-lateral (ML) :1: 0.6 mm. dorso-ventral (DV) -
6.7 mm: PVN/ZI: RC +0.3 mm. ML 1 0.5 mm. DV -6.4 mm: TAM/TAV:
RC +0.6 mm. ML :1: 1.3 mm. DV -5.3 mm; DV coordinates are from the
surface of the brain with the nose bar -2.0 mm from ear-bar zero).

A 1 -pl Hamilton syringe with a glass micrOpipette attached (tip
dia.: 40-50 m) to the needle was used to make the injections. The
assembly was first backfilled with mineral oil to displace all air in the
assembly. then backfilled with the fluorescent dye and mounted on the
stereotaxic. The micropipette was lowered in the brain at the selected
coordinates and, after a 5 minute delay. the tracer (60-200 nl RLM) was
injected over a 5 minute period using a micrometer drive. The syringe
was left in place for another 5 minutes after the injection was made.
then removed. The skin wound was closed and the animal returned to
the long—day photoperiod in an individual cage.

rf 1 Hi 1

Brain: After survival periods of 7 to 15 days. the animals were
perfused transcardially using the method described in Experiment Ia
with a heparinized (2 U / ml) 0.05 M phosphate buffered (pH 7.4)
physiological saline rinse followed by 4 % (V/v) formaldehyde in 0.05 M
phosphate buffered (pH 7 .4). The eyes were removed during the saline
rinse. They were stored in petri dishes labeled for left or right side of the
body. and. kept in the same rinse solution until the retinae were
removed. The brains were stored overnight at 4'C in the same fixative.

36

then cut (50-80 um sections) in a coronal plane. The sections were
mounted to gelatinized slides. cleared in xylenes and coverslipped using
DPX. a non-fluorescing mounting media. The injection sites were
verified by microscopic examination of the tissue using an epi-
fluorescent illumination attachment (510-560 nm excitation / 590 nm
barrier filter).

Retinae: The retinae were removed from the eyes and in some
cases were flattened. ganglion cell layer-side up. on gelatinized glass
slides. Radial cuts were made in the tissue to reduce distortion. A
coverslip was placed on top of the tissue and the slide placed in a petri
dish containing 0.05 M phosphate buffered 4% formalin for 6 hours to
fix the tissue. The slide was subsequently taken from the petri dish.
the coverslip removed. and the tissue rinsed with physiological saline.
The retinae were allowed to air-dry to the slides then cleared in xylenes
and coverslipped using DPX. The retinae were then examined for
“evidence of retrogradely labeled ganglion cells.

Some of the retinae were fixed free-floating in the buffered fixative
6- 18 hours. then mounted to gelatinized slides as described above.
before allowing them to air-dry. This procedure prevented tearing of the
tissue caused when the coverslip was removed following fixation.

Evaluation: One investigator counted the cells in each retina that
were labeled by the injections. An independent observer examined
selected retinae without knowledge of treatment to confirm the presence
and number of retrogradely labeled cells. Photographs were made of
representative labeled cells in the retinae and drawings were made of

the injection sites.

37
w

Retinal ganglion cells were consistently labeled. in either one or
both retinae. with RLM following injections placed bilaterally into the
three selected areas shown to receive retinal efi'erents (Experiment Ia).
No differences between the sexes were evident. Representative retinal
ganglion cells are shown in Figure 10. In all cases. the injection site
was composed of a brilliantly labeled core centered on the ventral tip of
the needle tract. Surrounding the core was a halo of dispersed RLM
and intensely labeled glia. neurons and processes. Beyond this ring.
labeled neurons and dendrites were evident but axons beyond a short.
proximal portion were rarely observed except when a significant tract
was included in the injection site (e.g. - anterior commissure [AC]. OT or
ST).

PVN/Z! - Injections (n = 4 animals) included in this group did
not extend into the hypothalamus ventral to the PVN or laterally to the
LGN by spreading through the surrounding tissue or along blood
vessels (Figure 1 1). In each case a short line of fluorescent dye marked
the needle tract through the overlying tissue but did not pass through
structures believed to receive retinal input. Labeled cells were located
in either one or both retinae. The total number of labeled retinal cells
ranged between 2 and 12 / animal and did not appear to be distributed
through the tissue in any particular pattern. Nor was there an obvious
pattern to the size of labeled somata (range: 10 um - 19 m). The
intensity of labeled RGC varied considerably from sparsely labeled
somata to brightly labeled cells and their associated proximal dendrites.
RGC axons could not be clearly identified in the retinae in any case.

38

 

Figure 10. Photomicrograph of representative RGCs retrogradely labeled
with RLM. The cells are from an animal that received a bilateral
injection of RLM into the PVN (A) or POA (B) and represent the range in
size of somata that were labeled. (epifluorescence illumination; Bar =
20 m)

39

 

SON

 

Figure l 1. Camera lucida drawing of a coronal section through the
hypothalamus at the level of bilateral RLM injections into the area of
the PVN/ZI in a Syrian hamster. The center of each injection is marked
by a dot. The total number of labeled cells counted in the retinae of
each animal ranged between 2 and 12.

40

POA/DBB - Injections that were placed in the POA/DBB (n = 4
animals) rostral to the SCN and medial to the LOT. including injections
that spread to the level of the OVLT. were counted in this group (Figure
'12). Similar to 'PVN/ZI' injections. a small number of RGC were
consistently labeled by RLM (range 3 - 15/ animal). One injection was
found placed particularly rostral in the brain anterior to the OVLT along
the ventral surface of the brain; this injection also labeled 3 RGC in
each retina. The low number of labeled cells and their bilateral
distribution suggests the tip of the micropipette attached to the syringe
had not pierced the ipsilateral optic nerve (ON) located medial to the
injection site at this level.

TAM / TAV- Injections in this area labeling RGC were effective only
if the rostral pole of the thalamus was included in the 'core' (Figure 13).
Injections that excluded this area but that did label the medial ST (i.e. -
injections of the bed nucleus of the stria terminalis [BNSTD were
ineffective in labeling RGC. Few cells in the retinae were labeled (range
2 - 9/ animal) and were not distributed in any obvious pattern.

I ryections outside the target regions: Retinal ganglion cells were
labeled by injections that included the LGN (n = 2 animals) or DC (n a 2
animals) or ON (n = 1 animal). These injections were associated with
many (>50) labeled retinal ganglion cells in one (an injection including
one ON) or both eyes (injections including the GO or LGN). Injections
that were confined to areas of the brain not observed to contain CT-HRP
stained fibers in Experiment Ia (e.g.. lateral septum. BNST. or medial
thalamus) did not label RGC.

In the remainder of the control cases the dye spread outside the
target area confounding the interpretation of the data. The most

41

 

 

 

 

 

 

 

 

 

Figure 12. Schematic representation of the POA at the level of bilateral
RLM injection into the brain of a hamster. The center of each injections
is marked by a dot. The total number of cells labeled in each animal
ranged between 3 and 1 5. (Illustration adapted from Maragos, et al..

1 989)

42

 

MPN

 

 

 

OC

 

 

Figure 13. Schematic representation of the brain of a Syrian hamster at
the level of bilateral RLM injections into the rostral thalamus. The
center of each injection is marked by a dot. The total number of cells
labeled in each animal ranged between 2 and 9. (Illustration adapted
from Maragos. et al.. 1989)

43

common mechanism of spread was dye following the needle tract
dorsally. But dye also spread along blood vessels passing through the
vicinity of the injection site. As a result. some injections centered in or
immediately dorsal to the PVN exhibited dye-spread either along the
large blood vessels that also passed through or near the SON and
lateral OC (n = 3 animals). or. along blood vessels that passed through
the LGN and optic tract (n = 2 animals). More RGC were labeled in
these cases compared to the cases in which injections were confined to
the intended target. In each case the relative contribution of each of the
affected areas could not be determined. Spread of the RLM was
reduced or prevented by decreasing the volume of dye injected and

decreasing the rate of injection.

I I N

The results of Experiment I confirmed the bilateral RHT input to
the SCN of the Syrian hamster and demonstrated the path of retinal
efferents to several extra-hypothalamic structures. including the DBB.
Pyr and rostral TAM/TAV. There is currently no evidence suggesting
that these particular regions of the brain receive direct input from the
SCN that might account for these inputs based on anterograde trans-
synaptic transport of CT-I-IRP (Watts. et aL. 1987). Nor have previous
studies employing CT-HRP as a tract-tracer reported evidence of such
transport (Johnson. et al.. 1988; Levine. et al.. 1986). The labeling of
RGC by injections of RLM into the PVN/ZI. TAM/TAV and POA/DBB
provides supporting evidence that retinal input reaches these

structures.

44

The SCN are the principal recipients of RHT input and were
labeled throughout their rostro-caudal extent. including the dorso-
medial crescent of each nucleus. Retinal input to this subdivision had
been in doubt until recently (Card & Moore. 1984) but seems to be a
consistent feature in the Syrian hamster (Experiment Ia; Johnson. et
al.. 1988). Next in density of RHT input in the hypothalamus was the
AHA: some fibers passed through this area to the PVN. In the rat. a
dense SCN projection ending in the AHA has been demonstrated by
injections of PHA-L while other fibers invade the PVN (Watts. et aL.
1987). Also. in the rat. some cells of the AHA and sub-PVN area project
to the PVN (Poulain 8: Carette. 1987). It seems that the distribution of
RHT fibers overlaps with that of some SCN projections: these inputs
may interact to modulate the output of the AHA and PVN and thus
influence the photoperiodic response in the hamster. Direct RGC input
may act to either work in concert with SCN input to enhance the
probability that a particular behavior or physiological mechanism is
activated. or to inhibit further conduction of SCN-generated activity and
thus prevent its activation (masking?).

The PVN is part of the multisynaptic pathway mediating the
circadian rhythm of MEL production and photoperiodism. but the
specific function of the PVN in this pathway is unknown. Rats and
hamsters exposed to pulses of light during darkness display a
precipitous decline in pineal and serum levels (Tamarkin. et al.. 1979
Vanecek 8: Illnerova. 1982; Illnerova 8: Vanecek. 1982a.b). In the rat.
comparisons of changes in the rhythm of MEL production in animals
released into constant darkness following exposure to a pulse of light at
night suggests that temporary suppression of pineal MEL production

45

may occur in the absence of long-term changes in the rhythm of MEL
production. Direct RHT input to the PVN. AHA (through its connection
to the PVN) or both may bypass the clock. thus. providing a mechanism
for modulating MEL production in the pineal gland at times during the
night when the circadian system is relatively insensitive to light.

A number of interconnecting circuits may exist between several
areas of the brain receiving retinal input. In the rat and hamster. the
intergeniculate leaflet (IGL) of the LGN participates in the entrainment
of circadian rhythms to the light-dark cycle (Pickard. et aL. 1987;
Johnson. et al.. 1989). The IGL projects to the SCN in the hamster
(Pickard. 1985). and. at least in the rat. this connection is reciprocated
(Card & Moore. 1989). In addition. the AHA and retrochiasmatic area
project to the IGL (Card 8: Moore. 1989) and receive direct retinal input
(Experiment la) in addition to input from the SCN (Watts. et aL. 1987).
These geniculo-hypothalamic circuits may participate in a feedback
mechanism to modulate the phase relationships of other SCN-
dependent rhythms to the photoperiod.

A projection to the POA labeled here and confirmed by RLM
injection has been previously recognized in both the rat and the
hamster (Pickard 8: Silverman. 198 1; Johnson et al.. 1988). Johnson
and coworkers (1988) reported labeled fibers that were traced to the
DBB and to the vicinity of the OVLT. These authors reported that the
fibers emerged from the optic nerve to provide input to the DBB and
caudally in the POA. ultimately terminating in the amygdala. The
interpretation of the present results give a different picture of this path.
In horizontally sectioned material the projection to the POA and DBB

was seen to emerge from the lateral OC and to follow a rostro-medial

46

course into the POA. In the rostral POA. fibers destined for the level of
the OVLT entered the horizontal limb of the DBB. A portion of the RGC
input to the rostral Pyr also coursed through the POA before turning
laterally.

In rodents. the POA participates in the control of sexual and
maternal behaviors as well as the release of gonadotropins
(Merchenthaler. et aL. 1984; Wiegand 8: Terasawa. 1982; Numan. 1988;
Kalra. 1986; Napoli. et al.. 1972; Powers 8r Valenstein. 1972; Larsson.
1979). This control involves connections that have been previously
described between the POA and the brainstem. 'limbic system' and
hypothalamus (see Swanson. 1987 : Berk & Finkelstein. 1981: Simerly
& Swanson. 1986; Kevetter & Winans. 1981). Gonadectomized
hamsters of both sexes. exposed to short photoperiods. are less
sensitive to the behavioral effects of exogenous gonadal hormones than
similarly treated animals housed in a long photoperiod (Badura. et aL.
1987b; Badura & Nunez. 1989; Campbell. et al.. 1978; Morin 8r Zucker.
1978). In the female. but not in the male. this effect is pineal-
independent (Badura. et al.. 1987b). In rodents. the POA participates in
the control of sexual and maternal behaviors as well as the release of
gonadotropins (Merchenthaler. et al.. 1984; Wiegand 8: Terasawa. 1982;
Numan. 1988; Kalra. 1986; Napoli. et al.. 1972; Powers 8: Valenstein.
1972; Larsson. 1979). This control involves connections that have been
previously described between the POA and the brainstem. 'limbic
system' and hypothalamus (see Swanson. 1987; Berk & Finkelstein.
1981; Simerly 8: Swanson. 1986: Kevetter 8r Winans. 1981). In the
male hamster. electrolytic or chemical lesions of the medial preoptic
area (MPoA) produces deficits in copulatory behavior (Powers. et aL.

47

1987). These deficits in the expression of male sexual behavior appear
to be due to disruption of medial amygdaloid nucleus connections to the
BNST and POA (Lehman. et al.. 1983; Lehman. et al.. 1980). In the
female rat. activation of maternal behaviors involves both the medial
preoptic area (MPoA) and lateral preoptic area (LPoA) and their
connections to the brainstem (Numan. 1988). Pinealectomized male
and female hamsters exposed to short photoperiod are reproductively
competent (Reiter. 1973/74). However. the maternal behavior of these
females is deficient when compared to that of females bearing litters in
long days. Thus. the effects of short-days on maternal behavior may be
pineal-independent. The POA and VMN are also involved in the control
of lordosis in the female rat and hamster (Dornan. et al.. 1989;
Takahashi 81 Lisk. 1988: Pfaff & Modianos. 1985). Thus. these
structures may integrate the information about ambient light with
chemosensory and endocrine input to mediate the pineal-dependent
and -independent photoperiodic efiects on sexual and matemal
reproductive behavior.

A circadian rhythm in body temperature has been reported for
rats and it is abolished by SCN lesions (Melanie & Kittrell. 1987). In the
rat. some neurons of the POA and anterior hypothalamus display
changes in their firing rate in response to thermal. osmotic and
cardiovascular challenges (Hori. et al.. 1988; Knox. et aL. 1973; Werner
& Bienck. 1985: Eisenman 81 Jackson. 1967). Retinal input to the POA
may provide a mechanism to modulate the circadian temperature
rhythm through connections with neurons responsible for the rhyihm's

generation.

48

In addition to RGC input to the VMN. retinal fibers reached the
DMN. 'lhese nuclei. along with the PVN. have been implicated in the
control of feeding and metabolism (Luiten. et al.. 1987). Recent work in
the hamster has shown pineal-independent photoperiod-induced
changes in various measures of energy balance and metabolism
(Bartness 8r Wade. 1985). Therefore. photic input to these structures
could be involved in the photoperiodic control of metabolic functions.

The retinal input into the piriform and periamygdaloid cortex
extended from near the Tu to the level of the anterior cortical nucleus of
the amygdala (Scalia 81 Winans. 1975). In contrast to a previous report
(Johnson. et al.. 1988) and in agreement with Pickard and Silverman
(1981). only a few fibers were evident in the amygdaloid nuclei. Like
olfactory fiber input. the retinal fibers terminated in the adjacent layer
Ia. In the rat and hamster. the main and accessory olfactory bulbs
project to layer In of the piriform and periamygdaloid cortex (Devor.
1976: Haberly 8: Feig. 1983). Thus. direct retinal input to the Pyr may
overlap olfactory projections. The olfactory cortex in turn provides some
of the input to the corticomedial amygdala and amygdalo-hippocampal
area (see Price. et al.. 1987). Fibers from these nuclei terminating in
the BNST and MPoA course through the stria terminalis (Maragos. et
aL. 1989) and modulate male sexual behavior as mentioned above. In
addition. amygdala projections reach the VMN and may modulate some
female sexual behaviors at this level (Krettek 8: Price. 1978).

Evidence of a RGC efferent projection ending in the rostral
thalamus has been obtained for several mammalian species (ltaya. et
al.. 1981: ltaya. et al.. 1986: Conrad & Stumpf. 1975: Johnson. et al..
1988: Levine. et al. 1986). The results of Experiment I confirmed the

49

RGC input to this thalamic region and extended our knowledge of this
path in the hamster by clarifying the course taken by these fibers to the
TAM /TAV. The anterior thalamic nuclei receive input from the
hippocarnpus formation (primarily the subiculum via the fornix). limbic
cortex and marnmillary nuclei (see Jones. 1985 for a review). Recently.
input to the anterior thalamic nuclei from several areas of the
brainstem associated with the auditory system has been identified
,(Sikes & Vogt. 1987 ; Gabriel. et aL. 1980a.b). In the rabbit. the TAV
have been implicated in the control of operantly conditioned responses
to auditory stimuli (Gabriel. et al.. 1980a.b). 'lhe electrophysiological
activity of the TAV and anterior cingulate cortex is positively correlated
with acquisition of discriminative behavior. When the relationship of
the conditional stimuli to the unconditional stimuli are reversed. only
the TAV displays a reversal in the electrophysiological response to the
'new' positive versus 'old' positive stimuli. Furthermore. this reversal
becomes significant only during the first session in which the training
criterion is met. Although there is currently no evidence for a similar
mechanism involving the control of responses to visual stimuli. the TAV
may participate in the learning of conditioned discriminatory behavior
through mechanisms that involve polymodal sensory input.

The combined results of Experiment Ia and lb suggest that the
retinal input to the hypothalamus and basal forebrain is more wide-
spread than traditionally thought. While the results of RLM injections
provided additional evidence in support of the extended distribution of
retinal input to certain areas of the brain. the total number of retinal
cells labeled per case was relatively small. The small number of labeled

RGC's may simply reflect the presence of a relatively small projection

50

from the retina to these sites. Alternatively. the small number of
labeled RGC's may indicate that the efficiency of latex microspheres to
label cells retrogradely is relatively low compared to anterograde
labeling of fibers by HRP conjugates. However. double-labeling studies
in the visual system have demonstrated that collaterals of some cells
reach widely separated terminal sites (Pickard. 1985. Giollo & Town.
1980; Jeffery. et aL. 1981: Yamadori. et al.. 1989). Thus. a subset of
ganglion cells may provide input to a variety of structures in the basal
forebrain simultaneously. Such input might serve to coordinate the
activity of various systems in response to the prevailing photoperiod in
addition to the information relayed by the SCN.

EXPERIMENT Ila-The PVN Efferent Pathway to the Spinal Cord
Identified by Iontophoretically Applied Phaseolus vulgaris-
Leucoagglutinin.

Lesions that destroy the PVN of the Syrian hamster prevent short-
day-induced gonadal regression in males and acyclicity in females
(Eskes 8: Rusak. 1985; Brown. et al.. 1988: Bartness. et al.. 1985:
Lehman. et aL. 1984; Pickard 8: Turek. 1983). Horizontal knife-cuts
placed above the PVN are similarly efiecfive in preventing short-
photoperiod induced gonadal regression in males (Nunez. et al.. 1985:
Inouye & Turek. 1986). The model of the multisynaptic pathway
mediating the expression of photoperiodism suggests that the PVN-
spinal cord connection is an essential segment of the pathway. PVN
efferents that course through the level of the knife cuts may be
disrupted by these lesions. It is clear that the PVN projects to the
spinal cord in the Syrian hamster (DonCarlos 8r Finkelstein. 1987;
Brown. et al.. 1987). but the trajectory of these fibers to the spinal cord
has not been described. In the rat. the course of PVN fibers terminating
in the spinal cord has been examined using PHA-L (Luiten. et al.. 1985).
TWO prominent bundles arising from the PVN were described: a dorso-
caudal (Tract 1) and a lateral (Tract 2) pathway. TYact l was comprised
of a relatively small bundle of fibers formed over the third ventricle and
Tract 2 developed from fibers coursing laterally from the nucleus.
However. similar data are not available for hamsters. The present
experiment aimed to describe the course of PVN efferents terminating in
the spinal cord of the Syrian hamster utilizing iontophoretic injections
of PHA-L into the PVN. The course of labeled fibers through the

51

52

diencephalon to the brain stem and spinal cord was studied in male

hamsters.

53
ME'I'HQD

Animals

Adult male Syrian hamsters were housed in the same conditions
as in Experiment 1.

Element

Animals were anesthetized (Equithesin 4.5 ml/kg) and placed in a
stereotaxic apparatus. The skull was exposed and a burr hole (dia.
3 mm) drilled through the bone. Each animal received an iontophoretic
injection of PHA—L (Gerfen 8r Sawchenko. 1984) aimed at the lateral PVN
(coordinates from Bregrna: RC 0.0 mm. ML 4» 0.3 mm and DV —6.6 mm:
nose bar —2.0 mm from ear-bar zero). Tip diameters for the glass
micropipettes ranged between 12-20 am. The lectin was ejected from
the micropipette by passing positive current through the solution using
a Midguard Constant Current device set to 5 (A and operated in the
alternating on/ off mode for 15 minutes. After removing the
micropipette. the skin over the skull was sutured closed. Each animal
was placed into an individual cage and retunied to the long—day
photoperiod for the remainder of the experiment.

P rf i n Hi 1

After a survival period of 6-8 days. the animals were perfused
transcardially following the protocol of Gerfen and Sawchenko (1984)
modified to contain buffer at 0.05 M rather than 0.1 M. This
modification. in addition to the use of an infusion pump. improved the
quality of the fixation. The brain and spinal cord (to the level of
vertebra T9) were removed together and post fixed in the final fixative for
at least 24 hours. Prior to sectioning. some of the tissue was embedded
in polyethylene glycol (PEG; Smithson. et al.. 1983).

54

Immunocytochemical staining was generally improved if the tissue was
embedded in this matrix prior to sectioning. Brains embedded in PEG
were sectioned with a rotary microtome (15 - 35 um) in a coronal or
parasagittal plane and the other brains were sectioned frozen (50 um) in
a coronal or parasagittal plane. The tissue was processed for
immunocytochemical visualization of PHA—L using a modification of the
protocol of Gerfen and Sawchenko (1984). The modification included
the incubation of the sections with normal rabbit serum (1:50) for
hour prior to incubation of the tissue with the primary antibody (1:500).
The tissue was mounted on gelatinized slides. lightly counterstained
with cresylecht violet and coverslipped.

mm

The tissue from each case was examined with brightfield
illumination for evidence of uptake of the lectin and anterograde
transport. Cases possessing PHA-L stained cells and fibers were
assigned to three groups: Group l-included cases with the injection
sites confined to the PVN. Group 2-injection sites included the PVN
plus tissue outside the PVN. Group 3 cases contained injection sites
located exclusively outside the PVN and were designated 'controls' for
this experiment. Camera lucida drawings of the tissue containing

labeled cells and processes were made of selected cases from each

group.

RESULII§

The centers of representative injections are shown in schematic

drawings in Figure 14. Four injections were centered in the PVN.

55

Figure 14. Schematic drawings of the center of PHA-L injections to show
their location in the brain. The numbers in the lower left corner refer to

the case number .(Cor - coronal section; Sag - sagittal section).

 

G50 (Cor) DMN

 

Figure l 4— (continued)

58

including two into the lateral PVN (G10 and G16) and one each into the
ventro-medial PVN (G28) and medial PVN (G67). Other injections
(Group 2) spread to include the PVN and adjacent structures. such as
the 21 (G133) and AHA (G39. G40). The remaining injections (Group 3)
pictured in Figure 14 were centered outside the PVN and did not spread
into the PVN (G18. G50. G52).

There were few differences in the overall pattern of staining in the
diencephalon and mesencephalon regardless of the injection's location.
The distribution of labeling obtained from the injections included in
Group 1 will be presented first. followed by a brief description of
particular variations that were evident in cases included in Groups 2 8:
3. Animal G16 received an injection into the proximal lateral PVN and
will be used to present the description of PVN efferents because of the
relatively large number of cells and fibers stained by this injection that
was confined to the PVN (Figure 15). Additional detail will be provided
from G 10 which was sectioned in a sagittal plane.

Labeled cells were found in the caudal two thirds of the PVN with
few cells located in the medial subdivision of the nucleus. Numerous
darkly stained fibers emanated from the vicinity of the injection.
oriented predominantly medic-laterally. Near the PVN these fibers were
relatively thick. with few varicosities. Shorter labeled fiber segments
emanated from the PVN in all directions. but the largest number were
distributed throughout the medial two-thirds of the hypothalamus
between the PVN and base of the brain. These segments were thin with
occasional clusters of swellings along the fibers or individual short
branches ending in swellings. The density of labeled fibers was
generally lower in the medial PVN than in immediately adjacent tissue

59

 

Figure 15 . Photomicrograph of a coronal section through the center of
a PHA-L injection site confined to the PVN (G16). (bflghtfield
illumination: Bar = 50 um)

60

(Figure 15. 16A.B). The fibers present in the medial PVN were fine
caliber with few branches.

Areas of the brain receiving PVN efi'erents are identified in Table
2. A fiber bundle of PVN origin was seen dorsal to the third ventricle.
Fibers contributing to this bundle (labeled 1a in Figure 16) coursed
medially over the PVN. then turned dorsally and traversed the thalamus
until reaching the thalamic paratenial (PT) and paraventricular nuclei
(PVT). Few varicosities or branches were evident on these fibers until
they entered these thalamic nuclei where some fibers appeared to
terminate in short branches with terminal swellings. The majority of
the fibers continued into the ipsilateral lateral habenula (LH in Figure
16) where short branches contained increased numbers of putative
terminals (Figure 16C). Occasional fibers were evident in the medial
habenula (MHa). A smaller number of fibers entered the contralateral
habenula. Data obtained from G 10 indicate that some fibers followed a
rostro-dorsal course over the anterior pole of the thalamus before
entering the PVT and habenula. Caudal to the PVN. few fibers of this
bundle were present in the thalamic and subthalarnic region between
the habenula and hypothalamus. Some fibers in this path continued
caudally to the level of the posterior commissure (PC) and
subcommissural organ (SO: Figure 16D). The increased branching of
labeled fibers and the number of swellings suggested this was also a
terminal field for this path. A ventral continuation of this path entered
the PAG where some of the fibers terminated. It is not clear whether
any fibers in this path continued on to the spinal cord.

A second dorsal. midline path (lb in Figure 16) was confined
closer to the roof of the third ventricle (Figure 16 A-C). At the level of

61

 

 

 

 

 

 

 

 

 

Figure 16. Camera lucida drawings of coronal sections through the
brain of a hamster (G 16) that received an iontophoretic injection of
PHA—L into the lateral PVN. Some fibers emerging dorsally (la 8r lb)
and ventro-laterally (2) from the PVN and formed tracts that followed
separate trajectories to the periaqueductal grey (PAG). (Bar = 300 um)

62

Table 2

Areas of the CNS Containing PHA-L Stained PVN Efferents

AMYQDALA Medial Preo tic Area
Septohypo alamic
Central Nucleus Nucleus
Corticomedial Nucleus Supraoptic Nucleus
pars anterior
BRAIN STEMlSPINAL CORD ars tuberal

entromedial Nucleus
Area Postrema

Penagueductal Gre SEPTAL AREA
Dors Nucleus of e
Vagus Lateral Septum
Intermediolateral Cell dorsal
Column intermediate
Locus Coeruleus Medial Septum
N'lticleus of the Solitary Thangular Septum
act
Parabrachial Nucleus W
HYEQIMUSZPREQPTIQ Lateral Habenula
Paraventricular Nucleus
Anterior Preoptic Area Nucleus Reuniens
Anterior Hypothalamic
Area
Dorsomedial Nucleus
Lateral Hypothalamic

Area

63

the mammillary bodies the fibers turned dorsally along the midline to
enter the PAG (Figure 16D). Some labeled fibers in the ventral. medial
and dorsal PAG branched into fine caliber short branches tipped by
stained swellings. A minority of these fibers then turned ventro-
laterally to join a lateral fascicle (described below) of stained fibers near
the lateral surface of the brain stem at the level of the parabrachial
nucleus (PB; Figure 16E). In the brain stem some midline fibers
terminated in the AP. locus coeruleus (LC). nucleus of the solitary tract
(NST) or dorsal nucleus of the vagus (DNV; Figure 17). While individual
fibers were observed near the midline caudal to the AP. there was no
evidence of a terminal field. Labeling was not detected near the midline
caudal to the raphe Obscurus (RPO) or in the central grey of the spinal
cord.

At the level of the PVN. diffusely distributed fibers were located
primarily in the medial two-thirds of the hypothalamus ventral to the
PVN (Figure 16A.B). Caudally. at the level of the VMN. a group of fibers
clustered near the fornix (bundle 2 in Figure 16C) and .other fibers were
evident in both layers of the median eminence. Further caudally. the
fibers in the ventro-lateral hypothalamus remained near the surface of
the brain. At the level of the mammillary bodies. fibers in the medial
hypothalamus provided input to the mammillary nuclear complex.
Fibers of the ventro-lateral bundle turned dorso-laterally taking a
position over the substantia nigra (SN). This group formed the fascicle
which joined with the dorsal path near the PB. Some of these fibers
terminated in the PB while the remainder appeared to continue into the
spinal cord. In the spinal cord. fibers were found in the lateral
funiculus (Figure 17). In the lower cervical spinal cord. labeled thin

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 1 7. Schematic drawings of horizontal sections of the caudal
brainstem and spinal cord (cervical and rostral thoracic) from an animal
that received an iontophoretic injection of PHA-L into the left (L) PVN

(G 16). Drawings are arranged dorsal (A) to ventral (C) and are
composites of 4 sections taken from each third of the spinal cord.

65

collateral branches could be traced medially to the IML. Some fibers in
the IML terminated in small clusters of swellings while others divided
again into several fine branches that surrounded individual cell bodies.
The pattern of labeling in tissue sections taken from other animals in
Group 1 was similar to that observed in G16.

. Several differences in the distribution of staining were noted in
cases included in Groups 2 and 3 compared to the cases in Group 1.
The dorsal bundles were more salient after injections that encompassed
both the dorsal PVN and 21 (G133) or AHA (G39. G40). In contrast. few
stained fibers were found in the habenula following injections restricted
to the Z1 (G52). the hypothalamus ventro-caudal to the PVN (G18). or
near the mammillo-thalamic tract (MT) dorso-caudal to the PVN (G50).
Injections into the AHA (G39. G40) and 21 (G52) labeled some fibers in
the PVN. but the density of labeling in the nucleus proper was clearly
less than in the hypothalamus immediately ventral to it. Staining was
not evident in the median eminence when the injection was confined to
the Z1 (G52). All of the injections resulted in labeled fibers in the spinal
cord with the exception of G40. in which no stained fibers could be
found in the spinal cord and G39 in which only one stained fiber was
found in the spinal cord. In three cases (G133. G18 & G52) stained fine
caliber fibers branched and surrounded individual cells in the IML
ipsilateral to the injection site.

EXPERIMENT IIb-‘lhe PVN Eiferent Pathway to the Spinal Cord:
Retrograde labeling of Neurons and Fibers in the PVN and
LHA.

Evidence suggesting that a direct paraventriculo-spinal projection
terminates in the IML was described for male hamsters in Ibrperiment
Ila. The data indicated that PVN efferents to the spinal cord may follow
several pathways. including the possibility of a dorsal tract passing
through the habenula. There is functional evidence to suggest a
possible sex difference in the circuit mediating the gonadal response to
photoperiod in Syrian hamsters. In males. horizontal knife cuts placed
above the PVN block testicular regression when the animals were placed
in a short-day photoperiod (Nunez. et al.. 1985; Inouye & Turek. 1985).
In contrast. similar cuts made in females were ineffective in preventing
uterine regression and acyclicity (Badura. et al.. 1989). One
interpretation of these results suggests that an anatomical difierence
exists between the sexes in the path taken by PVN efferent fibers
terminating in the spinal cord that are involved in the control of
photoperiodism. However. knife-cuts dorsal to the PVN in male
hamsters have been associated with elevated serum levels of FSH
(Badura. et aL. 1989). Regularly fluctuating levels of FSH are normal in
the female (Lisk. 1985) and persistently elevated levels of FSH resulting
from knife cuts placed dorsal to the PVN may be ineffective in
maintaining estrous cyclicity.

PHA-L offered several advantages over other anterograde tract-
tracing techniques (e.g. - autoradiography). The lectin is taken up by a
limited number of neurons and dendrites located at the injection site

but not by intact fibers-of-passage (Gerfen 81 Sawchenko. 1984). Thus

66

67

the extent of the effective injection site may be assessed. The resulting
Golgi-like filling of the stained cells and fibers. including fibers some
distance from their parent cells. permits easy identification of the
labeled fibers and putative terminals that can then be related to the set
of labeled neurons. However. in the present context the PHA-L
technique could not answer several questions regarding the PVN—spinal
cord connection in the hamster. Injections of PHA-L labeled cells
projecting to other areas of the brain in addition to those terminating in
the spinal cord. A distinction between those labeled paraventricular
fibers terminating prior to the spinal cord relative to those terminating
in the spinal cord was not possible. Also. there was no assurance that
the labeled cells were representative of the paraventriculo-spinal cord
connection. Retrograde tract-tracers can complement the results of
anterograde techniques such as PHA-L by providing information
regarding the source of input to the spinal cord. Conjugates of HRP can
retrogradely fill the neurons by way of fibers that extend Into the
injection site as well as the fibers themselves. thus. permitting
identification of the source neurons and the tracing of the fiber's course
from the parent cell toward the injection site. However. when several
sources of input exist. labeling of structures near the injection site can
be relatively dense and indiscriminate. reducing the possibility of
discerning the pattern of fiber termination in the vicinity of the injection
site.

The present experiment sought to confirm the path of axons of
PVN neurons that terminate in the spinal cord in male and female
hamsters. and to determine if sex differences exist in the anatomy of
these connections. using pressure injections of either WGA-HRP or CT-

68

HRP aimed at the level of segments C7-T, of the spinal cord to
retrogradely label PVN neurons.

69

W
Ann—nab
Animal and housing conditions were identical to EXPERIMENT Ia.
W

Animals were anesthetized with Equithesin (4.5 m1/ kg). Each
animal received unilateral or bilateral pressure injections of an HRP
conjugate between C-, and T2. Multiple injections (totaling 3 (11/ side) of
WGA- or CT—I-IRP were aimed at the IML. The conjugates were prepared
in the laboratory of Dr. RR. Miselis (Univ. of Pennsylvania). In some
early cases a laminectomy was performed (n = 5 males) at the level of C7
to T1 and the dura mater incised. To enhance uptake of CT-HRP the
spinal cord was hemisected (unilateral injection; n = 3) and the CT-HRP
injected just rostral to the area of insult. In cases where bilateral
injections were made. the lateral spinal cord was intentionally damaged
(crushed or cut transversely; n = 2) in the area which received the
injection.

In the remaining cases (n = 19 males 81 3 females). an incision
was made on the dura and pia mater after the spinal column was
exposed. and a glass tipped l-ul Hamilton syringe was used to place 3-1
pl injections of one of two HRP conjugates into the lateral spinal cord
(see Figure 18 and METHOD-Experiment Ib for details). Either WGA-
HRP or CT-I-IRP. prepared with the assistance of Dr. M.L. Weiss at
M.S.U.. was used. The protocol followed in the preparation of the
conjugates was essentially the same as used for Experiment Ia. In most
cases. the conjugate and free HRP mixture was used for injections into
the animals without first separating the conjugates from the precursors.
Prior

70

 

 

 

Figure 18. Schematic drawing demonstrating the vertebra. dorsal
spines and spinal cord of the upper thoracic and cervical spinal column
(a) of a Syrian hamster depicting the location of the glass pipette-tipped

l-ul Hamilton syringe (b) used to inject the HRP conjugates into the
lateral spinal cord (c).

71

to its use. a portion of the WGA-HRP was separated from the free HRP
on an FPLC column (Superose 12) provided by Dr. J Kaguni. Dept. of
Biochemistry. M.S.U. prior to its use. The various solutions produced
similar patterns of labeling in the hypothalamus. Following the
injections. the musculature and overlying skin was sutured closed. The
animals were individually housed and returned to the long-day
photoperiod following surgery until the time of sacrifice.
rf i n Hi 1

The animals were sacrificed after 2-3 days and the tissue
processed following the protocol described in Experiment la. The brains
were sectioned in either a coronal (n = 24) or sagittal (n = 3 males)
plane.

Evaluation

The sections were examined with bright— and dark-field optics
immediately after the reacted tissue was mounted and coverslipped.
Photographs were taken for documenting the presence of labeled fibers
and neurons within the PVN and lateral hypothalamus. Camera lucida
drawings of representative tissue sections were made during the same
examination of the tissue and stained fibers and retrogradely filled

neurons were plotted.

RESULT§
The pattern of labeling was similar in male and female hamsters. The
intensity of labeling increased with Increased survival time but the
pattern was not altered. Further. manipulation of the spinal cord near
the injection site did not alter the distribution of labeling. In addition
to the labeling in the hypothalamus that is reported below. an

72

occasional neuron (N < 4/ animal) in the SCN ipsilateral to the injection
site was observed. Labeling was not evident in the preoptic area or
hypothalamus rostral to the PVN. and no labeled cells were found in the
SCN. A list of sites in the brain that contained retrogradely labeled cells
is presented in Table 3.

The injections were clustered approximately half-way between the
midline and lateral edge of the spinal cord (Figure 18) and extended
approximately 1 mm rostro-caudally. spreading laterally to the surface
of the spinal cord. After unilateral injections a small amount of intra-
and extracellular reaction product was evident in the spinal cord
contralateral to the injection site. However. the contribution of this
staining to the results described below appeared to be minor based
upon the relatively sparse labeling found in the contralateral
hypothalamus.

Case 88-G84. an animal receiving a unilateral injection of WGA-
HRP. was representative of the results obtained and will be described in
detail. The hypothalamic projection to the spinal cord was
predominantly ipsilateral (Figure 19). Near the rostral pole of the PVN.
labeled cells and processes were scattered throughout the PVN
ipsilateral to the injection (Figure 19A. 20). At this level. the labeled
cells were generally small to medium sized. ranging between 9.5 and
16 pm on their long axis. An occasional cell was evident in the dorsal
and lateral hypothalamus. Moving caudally approximately 200 um from
this level. the cluster of stained cells in the ventro-lateral PVN expanded
laterally. The smaller cluster of cells in the dorsal PVN (Figure 19B)
was intensely labeled. Some horizontally directed fibers from this group
were traced to the midline (Figure 21).

Table 3

Areas of the Brain Containing Retrogradely Labeled Neurons Following
Injections of CT- and WGA-HRP into the Spinal Cord.

Telgnggphalgn

Basal Nucleus of Meynert
Pyramidal Cells

Parietal Cortex

Sensory Cortex
Substantia Innominata

Diengephggn

Dorsal Hypothalamus
Lateral Hypothalamus
Paraventricular Nuclei
Posterior Hypothalamus
Retrochiasmatic Area
Zona Incerta

Mammalian

Dorsal Raphe Nucleus
Lateral Dorsal Tegmentum
Nucleus of Darkschewitsch
Periaqueductal Grey

Red Nucleus

Metengephalgn

Locus Coeruleus

Medial Parabrachial Nucleus

Pontine Reticular Nucleus.
Ventral

Subcoeruleus

Reticulotegrnental Nucleus.
Pons

Pedunculopontine Tegmental

Nucleus

Mylengephalgn

Nucleus Amblguus

Reticular Nucleus of the
Medulla. ventral part

Nucleus of the Solitary
Tract

Raphe Caudal Linear Nucleus

Ventral Gigantocellularis
Nucleus

Ventral Medullary Reticular
Nucleus

 

 

Figure 19. Camera lucida drawings of coronal sections through the
brain of a male Syrian hamster that received a unilateral inj ection of
WGA-HRP into the spinal cord between vertebra C7 and T1. Labeling in
the brain was predominantly ipsilateral to the injection site. Four
tracts (labeled '1'. '2'. '3' 8: '4' in (Bl) of stained fibers were seen to
emanate from the PVN. (Bar [A 8r D] = 500 um)

75

 

Figure 20. Photomicrograph of a coronal section through the rostral
PVN of a Syrian hamster. The dorsal and ventro-lateral PVN ipsilateral
to an injection of WGA-HRP into the spinal cord contained retrogradely
labeled neurons and fibers. Tract 3 was prominent as fibers from the
PVN course ventro-laterally around the fornix. A few fibers were seen
over the roof of the third ventricle (Tract I). lateral to the PVN (Tract 2)
and ventral to the PVN (Tract 4). Scattered labeled cells were also seen
in the ipsilateral hypothalamus and contralateral PVN. (darkfield
illumination: Bar = 200 um)

76

 

Figure 21 . Photomicrograph of a coronal section near the middle of the
PVN. A collection of retrogradely labeled cells were observed in the
dorsal and ventro-lateral PVN. Fibers of TYact 3 remained evident
around the fornix. Some fibers of the dorsal cell group were traced to
the area over the third ventricle. (darkfield illumination: Bar = 200 um)

77

In the caudal third of the PVN. the lateral subdivision extending
over the fornix was so densely stained that individual features were
difficult to discern (Figure 190. 22). Those cell bodies which could be
distinguished from the background of dense staining were large (25 -
35 um on their long axis). fusiforrn and oriented in a horizontal plane.
At the same level. the number of labeled cells in the medial PVN
declined while the density of labeled fibers appeared to remain similar
to the previous level.

Near the posterior limit of the PVN. labeled cells were scattered
bilaterally and were more numerous in the dorsal half (Figure 19D. 23).
Contralateral to the injected side. the number of stained PVN cells and
fibers was quite small and mirrored the location of labeled cells in the
ipsilateral nucleus. Some fibers crossed the midline dorsally. but it was
not obvious from which half of the hypothalamus they originated
(Figure 19D,E).

Outside the PVN. stained fibers were generally located within one
of four bundles (see Figure 19 for a summary of the paths taken by
these bundles). T\vo minor bundles.’ one above and one below the PVN
remained near the midline. while two larger bundles emerged laterally.
,One immediately turned ventro-laterally approaching the lateral OC and
the other continued lateral over the fornix. To facilitate comparison
with Experiment Ila and the work of Luiten. et aL. (1985) in the rat. the
individual bundles will be identified as Tracts 1 through 4 with the
dorsomedial bundle as Tract l. the lateral bundle as Tract 2. the
ventro-lateral bundle as 'li'act 3 and the ventromedial bundle as TYact
4.

78

 

Figure 22. Photomicrograph of a coronal section through the caudal
third of the PVN showing the nucleus ipsilateral to the injection site.
The lateral parvocellular subdivision (as defined for the rat by Swanson
& Kuypersl19801) was intensely labeled and few individual cells could
be seen distinctly from the background of labeling. The ventro-medial
portion of the PVN contained fewer labeled cells than the lateral division
while the fewest number of cells were observed in the medial PVN.

(darkfield illumination: Bar = 200 um)

79

 

Figure 23. Photomicrograph of a coronal section through the caudal
PVN. WGA—HRP labeled cells were present in the PVN and scattered
through the LHA. At this level. fibers of TYact 2 split into two bundles.
One fascicle turned ventrally (b) to join TYact 3 in the caudal
hypothalamus while the second turned dorso-laterally over the CP (a).
(darkfield illumination: Bar = 200 um)

80

Tract I remained within 200 pm of the roof of the ventricle
through much of the caudal hypothalamus. Additional PVN fibers
joined Tract 1 from the posterior levels of the nucleus. Near the caudal
pole of the PVN. the tract became less salient. Comparison of coronal
sections with those cut in a sagittal plane suggests this tract continued
caudally through the posterior hypothalamus. where the bundle turned
dorsally. A dense cluster of labeled cells was grouped near the ventro-
medial MT and dorsal hypothalamus and fibers from these cells mingled
with Tract 1 fibers as they continued into the PAG (Figure 19E-H).
Retrogradely labeled cells were present in the PAG and the addition of
labeled fibers from these cells to those just described prevented tracing
Tract 1 further caudally.

Tract 2 arched dorsally over the fornix and continued into the
LHA (Figure 19A.B. 20 81 21). At the level of the caudal PVN. it divided 9
into two paths. a dorso-lateral and a ventro-lateral component (Figure
19D. 23). This division anticipated the appearance of the CP present in
subsequent caudal sections. Labeled cells were abundant from the Z]
ventrally to the OT (Figure 19D-F. 22). Retrogradely labeled cells were
also scattered in the 21 ipsilateral to the injection site and their axons
contributed to the dorsal limb of Tract 2. The dorsal component
continued caudally above the CP and SN while the ventral branch
merged with Tract 3 (described below) in the posterior hypothalamus.

Fibers in Tract 3 emerged from the lateral and ventro-lateral PVN
(Figure 24). surrounding and passing through the fornix then entered
the MFB as the tract coursed through the caudal hypothalamus (Figure
19). A few retrogradely labeled cells (3-4/ sections) were present in the
fornix and contributed to TYact 3. At the level of the caudal SCN. a

81

 

Figure 24. Photomicrograph of a coronal section near the level of Figure
21 . The PVN is just out of the field depicted dorsally and to the viewer's
left. Numerous fibers emerged from the ventral border of the PVN and
surrounded or passed through the fornix. (darkfield illumination: Bar =
50 run)

82

small group of cells and fibers were situated near the OC lateral to the
SCN (Figure 19B). Stained fibers from this cluster appeared to course
dorso-laterally to meet Tract 3. A second group of retrogradely labeled
cells was clustered in the retrochiasmatic hypothalamus adjacent to the
floor of the third ventricle further caudally (Figure 19D). Fibers from
this cluster also rose dorso-laterally to meet Tract 3. About 300 um
further posterior In the hypothalamus. Tract 3 collected around the
fornix (especially the ventromedial quadrant). Approximately 250 um
posterior. Ti'act 3 moved to a position lateral to the fornix. merging with
fibers of the ventral limb of 'n'act 2 and other fibers from the labeled
lateral hypothalamic neurons. Labeling in the LHA began at the level of
the caudal SCN and persisted through caudal sections. At the level of
the caudal PVN. labeled cells of the LHA formed a ventro-lateral
continuum with those labeled PVN neurons of the lateral subdivision
(Figure 23). Fibers were consistently seen in the OC crossing the
midline. Neurons of the LHA contributed some of these fibers and the
possibility of LHA contribution to Tract 2 or Tract 3 fibers cannot be
ruled out. In the posterior hypothalamus. at the level of the
mammillary bodies. Tract 3 moved dorso-laterally over the SN (Figure
19G.H). Fibers of this tract may have joined with fibers from the dorsal
limb of Tract 2 in this region.

The ventro-medially directed TYact 4. emerged from the ventral
PVN and following the wall of the third ventricle. tapering medially
toward the wall (Figure 25). The tract was first evident in sections
containing the SCN. The tract approached this nucleus. but did not
invade it. Additional fibers joined the tract at intervals from the PVN
but beyond the posterior PVN they turned caudally and could not be

83

 

Figure 25. Photomicrograph of a coronal section near the level of Figure
20. showing labeled fibers of TYact 4 near the wall of the third ventricle.
(darkfield illumination: Bar = 50 um)

84

detected in coronally sectioned material. Labeling in tissue sectioned in
a sagittal plane suggests these fibers continued caudally to the
posterior hypothalamus where fibers adjacent to the third ventricle wall
turned dorsally and were joined by fibers from retrogradely labeled cells
located nearby. It is possible that fibers of Tract 4 merged with fibers of
TYact 1 as they coursed through the rostral mesencephalon.

I I N

The PVN neurons terminating in the spinal cord were located
through the rostro-caudal extent of the nucleus (Experiment 11b). The
projections of the PVN are primarily uncrossed as reported previously
for the hamster (DonCarlos & Finkelstein. 1987) and rat (Swanson and
Kuyper. 1980). Most of the cells retrogradely labeled were in areas
corresponding to the parvocellular subdivisions described for the rat
(Swanson & Kuyper. 1980). however. a few labeled cell were located in
the magnocellular subdivision. Cell sizes ranged from those generally
considered parvocellular to magnocellular and. thus are not a good
predictor of neurons that terminate in the spinal cord.

The fibers of PVN neurons terminating in the spinal cord formed 4
descending tracts. Similar to the results obtained in Experiment Ila.
the WGA- or CT-HRP injections revealed the presence of fibers of PVN
origin that extended dorsally over the third ventricle (Tract 1) and
ventro-laterally in the hypothalamus (Tract 3). A small bundle of
laterally projecting PVN fibers had been discerned after PHA-L
injections to the nucleus. While individual fibers were detected in PHA-
L labeled material. the quantity of labeling was not suggestive of a fiber
path. In material stained with HRP conjugates. additional fibers in this

85

bundle were observed (Tract 2). This tract is similar in trajectory to the
TYact 2 described for the rat (Luiten. et aL. 1985). The presence of a
dorsal PVN-spinal cord projection passing through the habenula was
suggested by the pattern of labeled fibers after injections of PHA-L.
This projection was not confirmed by the results obtained with the HRP
injections. Following spinal injections. stained fibers were rarely
observed in the habenula or PVT. The method used detected a number
of paths originating in the PVN. including one that has not been
previously reported for PVN efferent fibers to the spinal cord (Iract 4).
In addition. the injections were capable of labeling other structures
previously reported to terminate in the spinal cord (Table 3). suggesting
this dorsal projection probably would have been labeled if it was
present. Thus the PVN projection passing through the thalamus and
epithalamus of the hamster appears to terminate prior to the level of
the spinal cord receiving the injections.

Of the two PHA-L labeled tracts. the injections predominantly
labeled fibers that appear to correspond to Tract 3 identified in
Experiment lIb. This may have been due to the placement of the PHA-L
injection within the proximal portion of the lateral PVN. Many of the
fibers in Tract 3 emerged from the medial and ventro-lateral PVN.

. The ventro-medial tract (i.e.. TYact 4) near the ventral wall of the
third ventricle was not detected in Experiment Ila. This may have been
because the tract was obscured by the staining found through much of
the medial hypothalamus after the PHA-L injections. Fibers of this tract
remained near the third ventricle through the caudal hypothalamus.
Horizontal knife cuts placed between the SCN and PVN prevent gonadal
regression in both male and female hamsters (Eskes 8: Rusak . 1985;

86

Nunez. et aL. 1985; Badura. et al.. 1987). These results have been
attributed to the disruption of SCN input to the PVN. But. Tract 4 may
provide supplemental input to the circuit mediating the circadian
rhythm of melatonin production. It is not possible at this time to
selectively sever this spinal cord connection as it emerges from the
ventral PVN without disruption of the SCN-PVN connection and thus. it
may prove difficult to isolate this tract and assess its contribution.
However. knife cuts made in a coronal plane and aimed ventro-caudal
to the PVN may be efi'ective in severing this tract without damaging
SCN-PVN projections.

Tract I remained relatively close to the roof of the third ventricle
throughout most of the hypothalamus. In male hamsters. horizontal
knife cuts placed dorsal to the PVN have been reported effective in
preventing gonadal regression in response to exposure to a short
photoperiod (Nunez. et al.. 1985; Inouye 8: Threk. 1986). However.
these same knife cuts do not prevent retrograde HRP labeling of PVN
neurons (Brown. 1986). Such lesions would likely have disrupted TYact
l but would probably not have caused major damage to Tract 2 or little
damage to Tracts 3 and 4. But. some of these cuts appeared to be
located in the thalamus dorsal to Tract l and thus would not have
disrupted the tract (Inouye & Turek. 1986). In castrated male
hamsters. these knife cuts dorsal to the PVN produce a rise in plasma
levels of FSH in excess of castrates receiving sham lesions (Badura. et
al. 1988). BLX and lesions of the olfactory system also cause elevated
levels of FSH and LH in male hamsters housed in a long photoperiod
(Clancy. et aL. 1986; Pieper. et al.. 1989). Long-term exposure of BLX
animals to a short photoperiod reduces the level of serum FSH. but to a

87

level that remains above that measured in long-day control animals.
and BLX males retain large testes. In the hypophysectomized rat.
infusions of FSH that produce serum levels in the physiological range
significantly reduces the decline in testicular weight and supports
spermatogenesis (Bartlett. et al.. 1989). Thus. large testes in short-day
housed males that have received knife cuts dorsal to the PVN may be
attributable to elevated levels of FSH. Although the knife cuts that
produce the persistent increase in FSH in males are generally found
near the PVN. there is no evidence to indicate that the PVN. per se. is
involved in the mechanism mediating the increase in serum levels of
FSH. These knife cuts may disrupt fibers-of-passage that course near
the PVN. Thus far. a systematic examination of this question has not
been conducted.

Coronal knife cuts aimed caudal to the PVN have also been
employed to examine the role of PVN efierents of Tract l in the pathway
mediating photoperiodism (Johnson. et al.. 1989; Smale. et al.. 1989).
Such knife cuts in male hamsters did not prevent gonadal regression.
Coronal knife cuts would have disrupted fibers of Tract I while sparing
most of the fibers oriented in a dorso-ventral plane that were damaged
by horizontal knife cuts suspected of being responsible for the rise in
serum FSH and large testes. These data support the notion that Tract
l is not an essential part of the pathway mediating photoperiodism in
the male hamster.

Bilateral parasagittal knife cuts in male and female hamsters that
are placed lateral to the PVN and that disrupt most or all of Tracts 2 &
3 are effective in preventing gonadal regression and acyclicity in short-
photoperiod housed animals (Badura. et aL. 1989; Smale. et al.. 1989;

88

Johnson. et al.. 1989). In males. other parasagittal knife cuts that do
not extend far enough ventrally so as to lesion Tract 3 and little. if any.
of Tract 2. do not prevent these short-photoperiod-induced effects
(Johnson. et aL. 1989). However. the interpretation of these results are
confounded by the placement of the cuts. SCN efferent fibers may
project directly onto lateral PVN neurons or participate in a circuit that
involves AHA neurons acting as intermediaries to lateral PVN neurons.
Those knife cuts found to be 'efi'ective' may have disconnected such a
circuit as well as disrupted most or all of Tract 2 8r 3. Whereas.
'ineffective' knife cuts may have damaged only a portion of these tracts
but failed to interrupt SCN input to the lateral PVN. and thus. allowed
the circuit to the pineal gland to remain intact. In the future. the
results of Experiment 11 might be used to guide the placement of lesions
in the path of each lateral tract and assist in evaluating gonadal

responses to exposure to short photoperiods.

EXPERIMENT III-Do Extrahypothalamic Lesions of a PVN to Spinal
Cord Pathway ('lract 1) Block the Testicular Response to
Photoperiod?

Data from Experiment 11 indicated that some fibers originating in
the PVN passed through the PAG in their course to the spinal cord.
These fibers project dorsally from the PVN before turning caudally. and
may correspond to Tract 1 as described in the rat (Luiten. et al.. 1985).
In the hamster. this tract aract l of Experiment lIb) follows the roof of
the posterior hypothalamic portion of the third ventricle caudally and.
at the level of the posterior hypothalamus, rises dorso-caudally to
penetrate the PAG. In the male hamster. knife cuts placed dorsal to the
PVN prevent gonadal regression in animals exposed to short-
photoperiods (Nunez. et al.. 1985. Inouye 8r Threk. 1986). Thus.
interruption of the dorsal efi'erent projections of the PVN. destined to
terminate in the spinal cord appears to block photoperiodic responses
in male hamsters. However. similar knife-cuts result in elevated serum
levels of FSH (Badura. et al.. 1988) thus complicating the interpretation
of the effects of such knife cuts on photoperiodism.

This experiment was completed prior to the inception of
Experiment HE and those recent studies describing the effects of
parasagittal knife cuts on the reproductive system of Syrian hamsters.
Thus it was designed to further investigate the importance of Tract l in
photoperiodism. by assessing the effects of electrolytic lesions aimed at
the rostral PAG on short-day induced testicular regression. Tract l and
the lateral efferents achieve maximum separation in the diencephalic-
mesencephalic transition where the fibers of Tract l are restricted to

the rostral PAG. Such lesions would interrupt T‘ract l at an extra-

89

90

hypothalamic site separated by some distance from the laterally
projecting PVN efi'erents and. presumably. from those connections

involved in mediating FSH release.

91
M21122

Animalé

Adult male Syrian hamsters were initially group housed (n = 5 or
6/ cage) and placed in a long photoperiod of 16 hours light and 8 hours
dark. Food and water were available ad libitum.

W'

At the end of the first week after their arrival to the colony, all
animals were separated into individual plastic cages. The right testis
width and body weight of each animal were measured at the end of the
second week. Testicular width has been shown to be positively
correlated with the ability to produce sperm in the male hamster (Threk
8: Losee. 1978; Berndtson 8: Desjardins. 1974). Animals were then
randomly assigned to three groups: lesion (n = 24). short day sham (n =
12) and long day sham (n = 12).

m

Tract 1 Lesion (LES): Animals were anesthetized with Equithesin
(4.5 ml/ kg) and placed in the stereotaxic apparatus. The skull was
exposed and two burr holes (coordinates from Lambda: RC +1.5 mm.
ML :1: 0.3 mm. DV -5.3 mm. DV taken from surface of the brain: nose
bar 2 mm below earbar zero) were made bilaterally in the skull over the
PAG. A steel insect pin (“00) insulated with Epoxy-lite. except for
approximately 0.2 mm of the tip. was used to make the lesions. The
electrode was aimed at the rostral portion of the PAG surrounding the
fourth ventricle bilaterally. An cathodal current (positive terminal of the
lesion maker attached to the electrode) of 2 mA was passed for
10 seconds. The electrode was immediately withdrawn and the skin
sutured closed. The animals were returned in individual cages to the

92

long-photoperiod room for at least two weeks after recovery from
anesthetic.

A large proportion of the animals with lesions (n = 19 of 24) did
not survive beyond 2 weeks post-surgically. Therefore. 7 animals
originally assigned to either sham group were reassigned to the lesion
group. The transfer of animals to the short-day room was delayed so
that additional surgeries could be performed to increase the number of
surviving animals with lesions (final n = 8).

Long day sham (LDS): Animals in this group (n = 6) were treated
in a similar manner as the lesioned animals. Sham surgery consisted
of lowering the electrode into each side of the brain using the same set
of coordinates. The electrode was left in place for 30 second. however
the Iesioning device was not turned on.

Short day sham (808): Animals in this group (n = 8) were treated
in a similar manner as long-day sham animals.

Measurements: Animals assigned to the short-day photoperiod
(sham and lesion) were transferred to a room set to a LD schedule of
6L: 18D after a minimum of two weeks postsurgical recovery. The right
testis width and length and body weights of all animals were obtained
at the time of transfer to the final photoperiod and every 1-2 weeks
thereafter for 13 weeks. The animals were lightly anesthetized with
ketarnine (1 mg/ kg) to measure testicular size. At the end of the
experiment. all short-day animals were anesthetized with an initial
injection of Equithesin (4.5 ml/ kg) and the testes removed. Final testes
weight. length and width were determined at this time. The animals
were then sacrificed by an overdose of Equithesin administered within
15 minutes of the first injection.

93
rfin 11

Animals with lesions were perfused transcardially with
heparinized (2 U/ ml) 0. l M phosphate bufiered (pH 7 .4) physiological
saline rinse followed by O. l M phosphate buffered (pH 7 .4) 4 % (V/v)
formaldehyde. The brains were stored in cold fixative at least overnight.
then sectioned in a coronal plane into distilled water. Sections were
mounted to gelatinized slides. counterstained with cresylecht violet and
coverslipped with Permount. The location of the lesion site was verified
by examining the tissue with brightfield optics.

mm

Camera lucida drawings of the lesion sites were made from the
counterstained tissue. Some of the drawings were traced onto photo
copies of an atlas of the rat brain (Paxinos 8: Watson. 1982). The atlas
plates used were selected by comparing structures that could be
identified in the sections obtained from the hamster brain to those
structures pictured in the atlas. The extent of damage to the PAG was
assessed by two observers who examined the counterstained sections
with brightfield illumination on a microscope. Complete lesions were
defined as the absence of tissue corresponding to the rostral PAG in
several sections containing the rostral mesencephalon. Incomplete
lesions left some portion of the rostral of the PAG intact.

@343

Group body weights and right testicular widths obtained at the
end of the experiment were analyzed by one-way ANOVA. Pair-wise
comparisons of mean testicular widths were made using the Least

Squares Difference test.

94
w

A total of three animals (I short-day sham and 2 long-day shams)
died prior to the completion of the experiment and thus were not
included in the data reported here. Examination of cresylecht violet
stained tissue obtained from the lesioned animals revealed that the
rostral PAG was destroyed in 3 cases and extensively damaged in the
other cases. Periaqueductal grey lesions (complete or incomplete) did
not prevent gonadal regression in animals maintained in the short
photoperiod compared to long-photoperiod-housed shams (Figure 26).
The onset of regression (statistically significant decline in mean width of
right testes as compared to the LDS group) in each of the two groups
housed in the short photoperiod was not different from one another and
occurred by Week 8. At the completion of the experiment the mean
testicular width of lesioned animals was not significantly different from
SDS animals (LES 7.29 m :t 0.29 mm [mean :t sem) vs SDS 7.45 m :t
0. 18 mm. p>0.05 [Table 4]) while the testes width of the sham-operated
animals in the long photoperiod (13.72 m :l: 0.20 mm) was significantly
greater than both short photoperiod-housed groups (p<0.05; Table 4).
There was no evidence of recrudescence in either group at the end of 13
weeks in the short photoperiod. The body-weight of animals in both
groups housed in the short photoperiod displayed a non-significant
decline (p>0.10) over the course of the experiment compared to long
photoperiod-housed shams (Figure 27).

DISQQSSION
Electrolytic lesions that destroyed the PAG did not prevent short-

day induced gonadal regression in male Syrian hamsters. Nor did these

lesions appear to alter the time course of gonadal regression compared

95

Table 4

Final Mean Width (1 sem) of Right Testis of Male Syrian Hamsters in

 

 

Experiment III
Group a Testes Width
(mm
Lesioned 8 7 .29 :t 0.29a
Short-day Sham 7 7.54 :t 0. 18b
Long-day Sham 4 13.72 :t 0.20

 

«1 Significantl different from long-day sham operated group
(p<0.001. f=10)

b Significantl different from long-day sham operated group
(p<0.001. f=9)

 

IE”. I .

 

 

 

E
5
:l:
[d
a
B
m
g.
tn
m .
E-
z 4"
< -
g 2_HLES
1I—ISDS
O HLDS
llllilllllill1
0 2 4 6 8 10 1214

WEEK

Figure 26. Mean width of the right testis (measured trans-scrotally)
of animals receiving electrolytic lesions aimed at the PAG or sham-
lesioned. Animals were lesioned (LES: n = 8) or sham lesioned
(SDS; n = 7) and placed in a 8L:16D photoperiod. or sham lesioned
, and placed in a 16L:8D photoperiod (long-day sham [LDS]: n = 4).
The mean testicular width for each short photoperiod housed
groups was significantly smaller than that of the LDS group by the
end of Week 8 and remained so until the end of the experiment.

97

 

 

150

140-

130-

MEAN BODY WEIGHT (gr)

 

120.3,... LES

l ' I ' l ‘ I ' I ‘ l

0 2 4 6 8 1 O l 2 14
WE E K

Figure 27. Mean body weight of animals receiving electrolytic

lesions aimed at the PAG or sham-lesioned. Animals were lesioned

(LES; n = 8) or sham lesioned (SDS; n = 7) and placed in a 8L:16D

photoperiod. or sham lesioned and placed in a 16L:8D photoperiod

(long-day sham [LDS]; n = 4). A nonsignificant decline in the mean

body weight was observed in both short-photoperiod housed groups.

Week zero represents the mean presurgical body weight of each

group .

98

to SDS animals. The lesions were severely debilitating to many of the
animals and were associated with a noticeable but non-significant
postsurgical weight loss. The present data indicate that the animals
with lesions tended to increase their body weight over most of the
experiment. appearing to match the rate of gain in body weight made by
the long photoperiod group. In the latter half of the experiment.
however. both short photoperiod group displayed a non-significant loss
of body weight. This trend is in contrast to reports of increased weight
gain shown by hamsters kept in short days (Wade. 1983: Bartness 8r
Wade. 1985: Hoffman. et al.. 1982). An explanation for the weight loss
is not available and may be due to differences in diet and housing
conditions between laboratories.

The results of Experiments Ila and 11b suggest that one or more of
the brainstem sites receiving PVN input (i.e. - PAG. PB. SOL and area
adjacent to the LC) and also containing neurons possessing axons
reaching the level of C7-Tl of the spinal cord. are in a position to
participate in the pathway mediating the reproductive system's
response to photoperiod. However. because PVN input to the PAG. LC
area and SOL would likely have been disrupted by rostral PAG lesions.
the direct involvement of these structures in the pathway appears
remote. The PVN efi'erent fibers projecting directly to the spinal cord via
a tract passing through the PAG (Tract I) also do not appear essential
for the display of gonadal response to short-day photoperiod. In the
female Syrian hamster, in contrast to the results in the male. horizontal
knife cuts placed above the PVN do not prevent the reproductive system
response to short-day photoperiod. Bilateral parasagittal knife cuts. on
the other hand. are effective in preventing gonadal regression in both

99

sexes when the animals are exposed to short-days. Taken together with
the results of Experiment lI. these results suggest that the efi'ectiveness
of horizontal knife-cuts above the PVN to prevent testicular regression
may not depend on disruption of dorsal PVN-spinal cord projections.
but may instead be due to elevated levels of FSH associated with this
type knife-cut. This rise in FSH levels may involve PVN efferents not
involved in the generation of the pineal melatonin rhythm. Thus. TYacts
2 and 3 (Experiment II) seem to provide an essential connection
between the PVN and the spinal cord in the pathway mediating
photoperiodism in male and female Syrian hamsters.

ENERAL DI I N

PHOTOPERI DISM

The present experiments examined the anatomical connections in
the Syrian hamster between the retina and forebrain. and between the
PVN and spinal cord. two segments of what is believed to be a
multisynaptic circuit ultimately controlling MEL secretion by the pineal
gland important for the display of photoperiodism (Tamarkin. et al..
1985). Photoperiodism is commonly associated with the seasonal
reproductive cycles of mammals such as the Syrian hamster. In the
hamster. seasonal cycles are eliminated by pinealectomy and mimicked
by timed infusions or injections of MEL (Elliott 82 Goldman. 1981:
Tamarkin. et aL. 1976; Watson-Whitmyre 8: Stetson. 1983). Seasonal
fluctuations in reproductive physiology. however. represent only one
aspect of mammalian photoperiodism.

Other aspects of the animal‘s physiology and behavior are
influenced by photoperiod in ways that prepare the animal for the
anticipated change in the environment. In the Syrian hamster.
exposure to a short photoperiod is associated with a number of
metabolic changes that are independent of gonadal secretions (see
Bartness & Wade. 1985). The feed efficiency (i.e.. weight gain/ calories
consumed). thermogenic capacity. brown adipose tissue mass and body
weight (primarily lipid content) are increased in short-day housed
animals with little change in their caloric intake. Further. thyroid
hormone levels and presumably metabolic activity. are reduced
(Vaughan. et al.. 1982). In Siberian hamsters. changes in pelage and
the onset of daily bouts of torpor are also associated with exposure to

100

 

101

short photoperiods (Bartness 8: Wade. 1985). All of these changes in
non-reproductive functions can be induced in animals kept in long days
by administration of timed daily injections of MEL (Bartness & Wade.
1984: Bartness 8r Wade. 1988). However. different from the effects of
photoperiod on reproductive physiolog. neither the pineal gland nor
MEL is necessary for the expression of some of these changes. For
example. Wade. Bartness and others have presented evidence that the
short-day efi'ect on body weight. energy balance and lipid content can
be expressed in the pinealectomized Syrian hamster (Bartness 8: Wade.
.1984; Hoffman, et al.. 1982).

The display of reproductive behavior is partially pineal-
independent in a sex-specific manner. Ovariectomized female
hamsters. kept in a short photoperiod. are less sensitive to activational
effects of estrogen and progesterone on sexual behavior than long day
housed animals. and. as in the case of changes in energy balance.
pinealectomy does not abolish this effect of photoperiod (Badura. et al..
1987: Badura & Nunez. 1989). In contrast to the female. however. the
short-photoperiod-associated reduced behavioral sensitivity of the male
hamster to testosterone is pineal-dependent (Miernicki. et al.. 1987).
Prolactin (PRL) release is also sensitive to changes in photoperiod
(Blask. et al.. 1986). Serum levels of PRL. a hormone associated with
reproductive and non-reproductive systems. fall in both males and
females when the animals are placed in short days. however, in the
female. pinealectomy is only partially effective in reversing the short-
photoperiod induced decline in the production and storage of PRL by
the anterior pituitary gland.

 

102
RGC IN? THE BASAL BRAIN

The RHT-SCN connection is considered an important component
of the circadian system and of the mechanism controlling
photoperiodism in reproductive physiology. But. as was seen in
Experiment I. the SCN is not the only area of the forebrain receiving
retinal efferents. A number of regions of the brain (in both sexes)
thought to be involved in the control of many of the regulatory and
behavioral functions that are affected by photoperiod (e.g. — POA. Pyr.
AHA, PVN) receive direct retinal input (Experiment I: J ohnson. et al..
1988). The extra-SCN retinal inputs. thus. may mediate some of the
effects of photoperiod on physiology and behavior that are independent
of the pineal gland. Photoperiodic responses in the Syrian hamster may
be pineal-dependent. -independent or some point in between and in
some instances there is a gender difference. It is not obvious why the
hamster should exhibit different degrees of involvement for MEL in the
control of photoperiodic responses in physiology and behavior. But. the
changes in physiology examined thus far appear to be adaptive to the
anticipated changes in the environment (the onset of "winter-like"
conditions).

Comparison of the target sites of SCN efferent projections and the
sites that receive RGC input in the hamster reveals a large degree of
overlap. It is, therefore. possible that this anatomical overlap provides a
basis for interactions between these two sources of light-dark
information. Direct retinal input might override the circadian control
provided by the SCN to a site. thus forming the basis of 'masking'. This
phenomenon occurs. for instance. when acute exposure to light causes

a decline in pineal MEL release at night while the SCN may or may not

103

be phase shifted by the same pulse of light (Illnerova 81 Vanecek.
1982a.b). A similar effect of light may be present in the phenomenon of
masking of locomotor activity and drinking rhythms observed during
transients when an animal is exposed to phase shifts of the LD cycle
(Sisk 81 Stephan. 1981).

Direct RGC input may interact with the SCN input to produce
pineal-independent photoperiodic effects. Thus. direct photic input
may affect different neural systems only during periods of
photosensitivity modulated in a circadian fashion by inputs from the
SCN. Such overlapping of inputs may provide an explanation for the

 

pineal-independent response to photoperiod noted in the production
and secretion of PRL. the behavioral sensitivity to steroids and the
control of metabolic functions in the hamster. The gender difierences
observed with respect to the effect of pinealectomy on some of these
photoperiodic responses may be due to sexually-dimorphic anatomical
or functional mechanisms in the regions receiving SCN and RHT input
or that are sensitive to MEL.

PARA - INAL NNE N

In contrast to what appears to be fundamental differences
between the sexes with respect to the degree that the pineal gland
participates in the behavioral and endocrine responses to photoperiod
mentioned above. the differential response by males and females to
horizontal knife cuts dorsal to the PVN does not appear to be due to a
sexual dimorphism in the anatomical pathways that mediate
photoperiodism. The anatomy of the paraventriculo-spinal connections
are similar in males and females (Experiment llb). Also. Experiment III
demonstrated that lesions in the PAG that presumably disrupt TYact I.

104

but do not damage fibers at the level of the PVN. do not prevent gonadal
regression in males. Other evidence suggests that the large testes
observed in animals with cuts dorsal to the PVN are the result of
elevated FSH levels (Badura. et al.. 1988). In hypophysectomized rats.
FSH replacement to physiological levels is sufficient to prevent complete
testicular regression even though spermatogenesis is incomplete
(Bartlett. et al.. 1989). Thus. the apparent sexually dimorphic response
to photoperiod after cuts dorsal to the PVN may be due to an increase
in FSH secretion that is sumcient to maintain large testicular size but
not female estrous cycles.

A number of questions remain to be answered regarding the
pathway mediating photoperiodism. While dorsally projecting PVN
efferents may not participate in the control of photoperiodism. the
course of PVN-spinal efferents that do remains to be clarified. Recent
evidence obtained in the male and female hamster suggests that lateral
PVN projections participate in this circuit (Badura. et al.. 1989: Smale.
et al.. 1989; Johnson. et aL. 1989). Discrete lesions placed in the path
of Tracts 2 or 3 may reveal that one or both of these tracts is essential
to the expression of photoperiodism. When these tracts are disrupted
in this manner particular care should be given to avoid damage to the
adjacent tract. The parasagittal cuts used in the previous studies may
have produced damage to both sets of PVN lateral projections. thus
preventing the evaluation of the role of each bundle in the control of
photoperiodic responses.

Since the pineal gland and MEL are intimately involved in the
expression of photoperiodism in the reproductive physiology of the
hamster. future studies involving effects of knife cuts and lesions on

105

photoperiodism should evaluate the effects of such insults upon the
rhythm of MEL production. Measurements of MEL levels in the serum.
ideally over a 24-hour period. should serve to establish whether the
circadian rhythm of MEL secretion has been disrupted by the brain
damage. Given the effects of knife cuts dorsal to the PVN and that of
damage to olfactory pathways. on the levels of serum FSH. it seems
important that the serum levels of FSH and LH be measured and the
histology of the gonads examined in addition to the use of testicular size
in males or cyclicity in females as indices of responsiveness to

photoperiodic changes.

NEURAL QONTROL QF THE QIRCADIAN RHYTHM QF PINEAL
NEIL SYNTHESIS ‘

 

 

 

A neurophysiological question also remains regarding the circuit
mediating the circadian rhythm of MEL production and photoperiodism.
Experiment II demonstrated a direct projection from the PVN to the
spinal cord and IML. But. the electrophysiological activity of the SCN
compared to pineal gland activity displays an inverse relationship. In
the rat, extracellular recordings of electrical activity in the SCN reveals
an increase in SCN activity during the day or when the optic nerve is
stimulated (Inouye & Kawamura, 1979. Shibata. et aL. 1984a.b). In
contrast. pineal gland activity (and melatonin production) is normally
reduced during the day or following acute exposure to light (Hudson 8:
Menaker. 1984; Tamarkin. et al.. 1979; Vanecek & Illnerova. 1982).
Also in the rat, activation of the SCG results in increased production of
MEL (Bowers 81 Zigmond. 1982; Bowers. et al.. 1984). Thus. at some
point in the pathway controlling the circadian rhythm of MEL

106

production. exposure of the animal to light must inhibit activity of the
SCG.

The PVN and AHA receive direct RGC input as well as input from
the SCN (Experiment 1; Johnson. et al.. 1988). In the rat. electrical
activity in the AHA. isolated with the SCN in a 'hypothalamic island'. is
low during the day and high at night (Inouye 8r Kawamura. 1979). This
observation is in contrast to the SCN which displays increased activity
during the day. Bilateral enucleation has no effect either on SCN or
AHA activity within the island but the lesions producing the islands
sever most or all RGC efferents. thus preventing light information from
reaching the isolated tissue. It would be informative to measure the
electrical activity of the same sites following acute exposure to light at
night in similarly prepared hamsters. providing the optic fibers were
preserved intact. AHA cells. in the hamster. receiving SCN input may
be similarly influenced by direct RGC input and participate. through its
connections to the PVN, in mediating the acute effects of light at night
on the rhythm of pineal MEL synthesis.

The PVN projects to the IML (Experiment 11; Luiten. et aL. 1985)
and could serve to inhibit further transmission of SCN output to distal
segments of the circuit. In the Guinea pig. a number of
neurotransmitters and peptides are present in the IML (Chiba &
Masuko. 1987). Among these is enkephalin. which has been reported
to produce inhibitory efi'ects in the CNS (Zieglgansberger. 1982). In the
rat, some PVN neurons projecting to the spinal cord contain
enkephalin-like irnmunoreactivity and may provide inhibitory input to
the IML (Cechetto 8: Saper. 1988). Thus. the PVN is in a position to

107

mediate inhibitory influences on the rest of the pathway. However. in
the rat. stimulation of the PVN for an extended period (3 hours) during
the day. but not during the night. results in a significant increase in the
concentration of the MEL metabolite. 6-hydroxymelatonin in the urine
(Yanovski, et al.. 1987). This suggests that the inhibition of MEL
production at night due to acute exposure to light is due to inhibition of
the PVN or another structure at a point in the circuit prior to the PVN.

Experiment 11 also revealed a number of nuclei in the brain that
receive PVN input and that. in the rat. terminate in the IML. Among
these nuclei are the locus coeruleus. nucleus of the solitary tract and
parabrachial nucleus (Tucker & Saper. 1985; Westlund. et aL. 1981:
Chiba & Masuko. 1987). The LHA and dopaminergic cells in the
posterior hypothalamus-rostral mesencephalon (A1 I) also project to the
IML (Chiba 8: Masuko. 1987 ; Lindvall. et al.. 1983; Tucker 8: Saper.
1985). Similar connections appear to be present in the hamster and
one or several of these nuclei may act to modulate the activity of
neurons in the IML. The results of Experiment III suggest the
parabrachial nucleus would be the most likely candidate as a brainstem
relay in the circuit. The possibility also exists that interneurons
present at one or more level of the pathway are responsible for this
phenomenon.

SPECIE§ §PEQIFIQ ANAIIQMY

The model of the multisynaptic pathway mediating the circadian
rhythm of pineal MEL secretion first obtained much of its anatomical
supporting evidence in data for the rat (Tamarkin. et aL. 1985). Studies
that have addressed the functional anatomy of this circuit in the
hamster have. in the main. also supported this model but have had to

108

rely upon the anatomical data for the rat to guide the interpretation of
the results (Inouye & Turek. 1985; Nunez. et al.. 1985: Eskes 8r Rusak.
1985: Lehman et aL. 1984: Pickard & Turek. 1983). However. the
present data and those from other studies suggest that differences exist
between neuroanatomical connections of the rat and hamster
(Experiments l 8: 11; Johnson. et al.. 1988; Levine. et al.. 1986;
Youngstrom 8r Weiss. unpublished results). Differences exist in the
pattern of labeling between the rat and hamster following intraocullar
injections of HRP conjugates. The hypothalamus of the Syrian hamster
is more densely stained at every level compared to the rat. In addition.
the input to the anterior thalamus and Pyr observed in the hamster was
absent in the rat. Other differences in the anatomy of these two species
with respect to the trajectory of PVN efferents to the spinal cord were
observed in Experiment 11. Thus. four fiber tracts originating in the
PVN were counted in the hypothalamus of the hamster following spinal
cord injections of HRP conjugates. In contrast. iontophoretic injections
of PHA-L labeled two tracts in the rat. These differences with respect to
paraventriculo-spinal connections must be viewed cautiously. given the
different tract-tracing techniques used and the results of Experiment Ila
compared to IIb. Nevertheless. it seems essential that the
neuroanatomy of the hamster be further addressed. Photoperiodism is
the product of an interaction between a circadian system displaying a
circadian rhythm of photoperiodic photosensitivity and the external
photoperiod (Elliott 81 Goldman. 1981). If we are to fully understand
this phenomenon it seems only reasonable that the anatomy of the
underlying structures in the hamster and other photoperiodic species

receive the same attention that the physiological responses associated

109

with photoperiodism have received. Otherwise the interpretation of the
results of functional studies using photoperiodic species will continue to
rest upon the neuroanatomical descriptions available for the essentially
non-photoperiodic rat.

BIBLIOGRAPHY

Badura. L.L.. Sisk. C.L. 8: Nunez. A.A. Neural Pathways Involved in the
Photo eriodic Control of Reproductive Physiology and Behavior in
Fem e Hamsters (Mesocricetus auratus). NeuroendocrinoL.
46:339-344. 1987a.

Badura, L.L.. Yant. W.R 8r Nunez. A.A. Photoperiodic Modulation of
Steroid-Induced Lordosis in Golden Hamsters. Physiol. Behav..
40:551-554. 1987b.

Badura. L.L.. Sisk. C.L. & Nunez. A.A. Knife Cuts Dorsal to the
Hypothalamic Paraventricular Nucleus Disinhibit Gonadotrophin
Secretion in Male Golden Hamsters Kept in Long or Short Days.
Soc. Near. Abstr.. 14:52. 1988

Badura, L.L & Nunez. A.A. Photoperiodic Modulation of Sexual and
Aggressive Behavior in Female Golden Hamsters (Mesocricetus
auratus): Role of the Pineal Gland. Harm. Behav.. 23:27-42. 1989.

Badura. L.L.. Kelly. K.K. 81 Nunez. A.A. Knife Cuts Lateral but not
Dorsal to the Hypothalamic Paraventricular Nucleus Abolish
Gonadal Responses to Photoperiod in Female Hamsters
(Mesocricetus auratus). J. Biol. Rhythms. 4:79-91, 1989.

Bartlett. J .M.S.. Weinbauer, G.F. & Nieschlag. E. Differential Effects of
FSH and Testosterone on the Maintenance of Spermatogenesls in
tlléesgidult Hypophysectomized Rat. J. EndocrinoL. 121:49-58.

Bartness. T.J. 8: Goldman. B.D. Peak Duration of Serum Melatonin and
Short Da Responses in Adult Siberian Hamsters. Am. J. PhysioL.
255zR81 -R822. 1988

Bartness, T.J. & Wade, G.N. Photoperiodic Control of Body Weight and
Energy Metabolism in Syrian Hamsters (Mesocricetus auratus):
Role of Pineal Gland. Melatonin. Gonads and Diet. EndocrinoL.
114:492-498. 1984.

Bartness. T.J. 81 Wade. G.N. Photo eriodic Control of Seasonal Body

Weight Cycles in Hamsters. eurosci. Biobehav. Revs. 9:599-612.
1985

110

111

Bartness. T.J.. Bittman. EL. 8: Wade. G.N. Paraventricular Nucleus
Lesions Exa :erate Die Obesity but Block Photoperiod-
Induced Wei at Gains an Suspension of Estrous clicity in
Syrian Hamsters. Brain Res. BuIL. 14:427-430. 198 .

Berk. M.L & Finkelstein. J A. Afferent Projections to the Preoptic Area
aggfjigosthalanuc Regions in the Rat Brain. Neurosct. 6:1601-

Berndtson. W.E. 8r Desjardins. C. Circulating LH and FSH Levels and
Testicular Function in Hamsters During Light Deprivation and
Sgbsequent Photoperiodic Stimulation. EndocrinoL. 95: 195-205.

4

Bittman. E.L. & Karsch. F.J. Nightly Duration of Pineal Melatonin
Secretion Determines the Reproductive Res onse to Inhibitory
Day Length in the We. Biol. Reprod.. 80:5 5-593. 1984.

Bittman. E.L. Melatonin and Photo eriodic Time Measurement:
Evidence from Rodents and uminants. In: Reiter. R.J. (Ed.). The
Pineal Gland. New York. Raven Press. pp. 155-192. 1984.

Bittman. E.L.. Dempsey. R.J. & Karsch. F.J. Pineal Melatonin Secretion
Drives the Reproductive Response to Daylength in the have.
EndocrinoL. 113:2276-2283. 1983.

Bittman. E.L.. Ka ard, AH. Olster. D.H.. Robinson. J.E.. Yellon. SM.
81 Karsch. .J. Pineal Melatonin Mediates Photoperiodic Control
of Pulsatile Luteinizing Hormone Secretion in the Ewe.
NeuroendocrinoL. 40:409-418. 1985

Bittman. E. L. & Lehman, M.N. Paraventricular Neurons Control
Hamster Photoperiodism by a Predominantly Uncrossed
Descending Pathway. Brain Res. Bull, 19: 687 -694. 1987.

Blask. D.E.. Leadem. C.A.. Orstead. KM. 81 Larsen. B.R. Prolactin Cell
Activiatfi in Female and Male Syrian Hamsters: An Apparent
Sexu y Dimo hic Res onse to Li t De rivation and
Pinealectomy. euroe rinoL. 4 :15-2 , 1986.

Bowers. C.W.. Baldwin. C. 8r Zi ond, R.E. Sympathetic Reinnervation
of the Pineal Gland After ostganglionic Nerve Lesion Does Not
Restore Normal Pineal Function. J. Newosct. 4:2010-2015, 1984.

 

Bowers. CW. 8: Zigmond. RE. The Influence of Frequen and Pattern
of Sgpathetic Nerve Activity on Serotonin N-ace transferase in
the t Pineal Gland. J. PhysioL. 330:279-296. 1 82.

Broadwell. R.D. & Brightman. M.W. Entry of Perorddase into Neurons of
the Central and Peripheral Nervous Systems from Extracerebral
and Cerebral Blood. J. Comp. Neurol. 166:257-284. 1976.

112

Brown. M.H. H othalamic Neural Pathways that Mediate Photoperiodic
Control 0 Reproduction in the Syrian Hamster (Mesocricetus
auratus). A Dissertation. Michigan State University. 1986.

Brown. M.H.. Badura. L.L. 8r Nunez. A.A. Evidence that Neurons of the
Paraventricular Nucleus of the Hypothalamus with Projections to
the Spinal Cord are Sensitive to the Toxic Effects of N- ethyl
Aspartic Acid. Neurosci. Lett.. 73:103-108. 1987.

Brown. M.H.. Badura. L.L. & Nunez. A.A. Axon-Splaring Lesions of the
Hypothalamic Paraventricular Nucleus Abo sh Gonadal
Responses to Photoperiod in Male Syrian Hamsters. J. Biol.
Rhythms. 3:59-69. 1988.

Campbell. C.S.. Finkelstein. J .S. & Turek. F.W. The Interaction of
Photoperiod and Testosterone on the Development of Co ulatory
Behavior in Male Hamsters. Physiol. Behav.. 21:409-41 . 1978.

Card. J.P. & Moore. R.Y. Organization of Lateral Geniculate-
Ilizpotlléaalgmic Connections in the Rat. J. Comp. Neurol. 284: 135-

Card. J.P. 8r Moore. RY. The Suprachiasmatic Nucleus of the Golden
Hamster: Immunohistochemical Analysis of Cell and Fiber
Distribution. Neurosci. 13:415-431. 1984.

Carter. D8. and ED. Goldman. Anti onadal Effects of Timed Melatonin
Infusion in Pinealectomized M e Djungarian Hamsters
(Phodopus sun orus sungorus): Duration is the Critical
Parameter. ocrinoL. 113:1261-1267. 1983.

Cassone. V.M. Speh. J .C.. Card. J .P. 8: Moore. R.Y. Com arative
Anatomy of the Mammalian Hypothalamic Suprac asmatic
Nucleus. J. Biol. Rhythms. 3:71-91. I988.

Cechetto. D.F. & Saper. C.B. Neurochemical Organization of the
Hypothalamic Pro ection to the Spinal Cord in the Rat. J. Comp.
Neurol. 272:579- 04, 1988.

Chiba. T. 8: Masuko. S. Syna tic Structure of the Monoamine and
Peptide Nerve Terrnin s in the Intermediolateral Nucleus of the
Géiénea Pig Thoracic Spinal Cord. J. Comp. NeunoL. 262:242-255.
1 7.

Clancy. A.N.. Goldman. B.D.. Bartke. A. & Macrides. F. Reproductive
Effects of Olfactory Bulbectomy in the Syrian Hamster. Biol.
Reprod.. 35:1202-1209. 1986.

Conrad. CD 8: Stum f. W.E. Direct Visual Input to the Limbic System:
Crossed Retin Projections to the Nucleus Anterodorsalis
Thalami in the Tree Shrew. Exp. Brain Res.. 23:141-149. 1975.

Devor. M. Fiber Trajectories of Olfactory Bulb Efferents in the Hamster.
J. Comp. Neurol. 166:31-48. 1976.

 

113

Don Carlos. L.L. & Finkelstein. J .A. H othalamo-Spinal Pro ections in
the Golden Hamster. Brain Res. 1111.. 18:709-714. 19 7.

Dornan. W.A.. Block. G.J.. Priest. C.A 81 Micevych. P.E. Microinjections
of Chole stokinin into the Medial Preo tic Nucleus Facilitates
193.714‘10155889 ehavior in the Female Rat. P siol. Behav.. 45:969-

Eisenman. J.S 8: Jackson. D.C. Thermal Response Patterns of Se tal
and Preoptic Neurons in Cats. Exp. Neurol. 19:33-45. 196 .

Elliott. J .A. & Goldman. B.D. Seasonal Reproduction: Photoperiodism
and Biological Clocks. In: Adler. NT. (Ed). Neuroe o
Reproduction. New York. Plenum Press, pp. 377-423. 1981.

Eskes. G.A. 8r Rusak. B. Horizontal Knife Cuts in the Su rachiasmatic
Area Prevent Gonadal Responses to Photoperiod. eurosci. Lett..
61:261-266. 1985.

Gabriel. M.. Foster. K.. Orona. E.. Saltwick. Se. 8: Stanton. M. Neuronal
Activity of Cingulate Cortex, Anteroventral Thalamus. in
Hippocampal Formation and Discriminative Conditioning:
Encoding and Extraction of the Significance of Conditional
Stimuli. In: Sprague. J .M. & Epstein. A.N. (Eds). Progress in
Psychobiology and Physio ical Psychology, Vol. 9. New York.
Academic Press, pp. 125-2 2. 1980a.

Gabriel. M.. Orona. E.. Foster. K. & Lambert. R.W. Cingulate Cortex
and Anterior Thalamic Neuronal Correlates of Reversal Learning
in Rabbits. J. Comp. Physiol. Psych. 94:1087-1 100. 1980b.

Gaston. S. 81 Menaker. M. Photoperiodic Control of Hamster Testes.
Sci. 158:925-928. 1967

Gerfen. C.R. & Sawchenko. P.E. An Anterograde Neuroanatomical
Tracing Method that Shows the Detailed Morpholo of Neurons.
Their Axons and Terminals: Immunohistochemical calization of
an Axonally T‘ransported Plant Lectin. Phaseolus vulgaris.
Leucoagglutinin (PHA-L). Brain Res.. 290:219-238. 1984.

Giolli. R.A. & Town. L.C. A Review of Axon Collateralization in the
Mammalian Visual System. Brain Behav. EvoL. 17:364-390.
1980.

Goldman. B.D. & Darrow. J .M. The Pineal Gland and Mammalian
Photoperiodism. NeuroendocrinOL. 37 :386-396. 1983.

Goldman. B.D.. Hall. V.. Hollister. C.. Rep ert. S.. Roycurdhury. P..
Yellon. S.M. 8r Tamarkin. L. Dium Chan es in Pineal Melatonin
Content in Four Rodent Species: Relations p to Photoperiodism.
Biol. Reprod.. 24:778-783. 1981.

114

Haberlg)’. L.B & Feig. S.L. Structure of the Piriforrn Cortex of the
possum. II. Fine Structure of Cell Bodies and Neuropil. J. Comp.
New-al.. 218:69-88. 1983.

Hastings. M.H. 8: Herbert. J. Neurotoxic Lesions of the Paraventricula-
Spinal Projection Block the Nocturnal Rise in Pineal Melatonin
Synthesis in the Syrian Hamster. Neurosci. Lett., 69: 1-6. 1986.

Hoffman. RA.. Davidson. K. 81 Steinberg, K. Influence of Photoperiod
and Temperature on Weight Gain. Food Consumption. Fat Pads
aggz'lhyroxine in Male Golden Hamsters. Growth. 48:150-162.

Hori. T.. Nakashima. T. Koga. H.. Ki ohara. T & Inoue. T. Convergence
of Thermal. Osmotic and Car ovascular Signals on Preoptic and
Anterior H thalamic Neurons in the Rat. Brain Res. Bull.
20:879-88 . 1988.

Hudson. D.J. 8r Menaker. M. Pineal Melatonin Synthesis in the
Hamster: Sensitivity to Light. Soc. Neurosci. Abstr.. 10:820. 1984.

Illnerova. H. 8: Vanecek. J. Complex Control of the Circadian Rhythm in
N -Acetyltransferase Activity in the Rat Pineal Gland. In: Aschofi'.
J .. et al.. (Eds.). Vertebrate Circadian Systems. Berlin. Springer-
Verlag. pp. 285-296. 1982a.

Illnerova. H. & Vanecek, J. Two-Oscillator Structure of the Pacemaker
Controlling the Circadian Rhythm of N-Acetyltransferase in the
Rat Pineal. J. Comp. Physiol. 145:539-548. 1982b.

Illnerova. H. Vanecek. J. Effect of Light on the N-Acetyltransferase
Rh in the Rat Pineal Gland. In: Reiter. RJ. 81 Karasek. M.
( 5.). Advances in Pineal Research. London. John Libbey & Co
Ltd.. pp. 69-76. 1986.

Inouye. S.-I. 81 Kawamura. H. Persistence of Circadian Rhythmicity in a
Mammalian Hypothalamic "Island" Containing the
83 £rsachiasmatic Nucleus. Proc. Natl. Acad. Sci. 76:5962-5966.
1 .

Inouye. S.-I. 8: Turek. F.W. Horizontal Knife Cuts Either Ventral or
Dorsal to the Hypothalamic Paraventricular Nucleus Block
Testicular Regression in Golden Hamsters Maintained in Short
Days. Brain Res. 370:102-107. 1986

ltaya. SK & VAN Hoesen. G.W. WGA-I-IRP as a Transneuronal Marker in
the Visusazl Pathways of Monkey and Rat. Brain Res.. 236: 199-
204. 19 .

ltaya. S.K. VAN Hoesen. G.W 8: Benevento. L.A. Direct Retinal Pathwa
t60 ghe bistable Thalamus of the Monkey. Exp. Brain Res.. 81:60 -
l . I .

115

ltaya. S.K.. VAN Hoesen. G.W. & Jenq. C.-B. Direct Retinal Input to the
Limbic System of the Rat. Brain Res.. 226:33-42. 1981.

Jeffery. G. Cowey. A. 81 Kuypers. H.G.J.M. Bifurcating Retinal Ganglion
Cell Axons in the Rat. Demonstrated b Retrograde Double
Labeling. Exp. Brain Res.. 44:34-40. 1 81.

Johnson. RF.. Morin. L.P. & Moore. RY. Retinohypothalamic
Projections in the Hamster and Rat Demonstrated Using Cholera
Toxin. Brain Res.. 462:301-312. 1988.

Johnson. RE. Moore. RY. 8: Morin. L.P. Lateral Geniculate Lesions
Alter Circadian Activity Rhythms in the Hamster. Brain Res. Bull.
33:422-422. 1989.

Johnson. R.F.. Smale. L. Moore. RY. & Morin. L.P. Paraventricular
Nucleus Efferents Mediating Photo riodism in Male Golden
Hamsters. Neurosci. Lett.. 98:85-9 . 1989.

J ones.llg.8(‘g The Thalamus. Plenum Press. New York. pp. 673-698.

Kalra. S.P. Neural Circuitry Involved in the Control of LHRH Secretion:
A Model for Preovulatory LH Release. In: Ganong. W.F. and
Martini. L. (Eds.). Frontiers inNeuraendacrinolagy. 9:31-75. 1986.

Kappers. J .A. The Develo ment. To ographical Relations and
Innervation of the iphysis erebri in the Albino Rat. Z.
Zellfarsch. 52: 163- 15. 1960.

Katz. L.C.. Burkhalter. A. & Dreyer. W.J. Fluorescent Latex
Microspheres as a Retrograde Neuronal Marker for In Vivo and In
Vitro Studies of Visual Cortex. Nature. 310:498-500. 1984.

Kevetter. G.A. & Winans. S.S. Connections of the Corticomedial
Arnygdala in the Golden Hamster. I. Efferents of the "Vomeronasal
Amygdala". J. Comp. Neural. 197:81-98. 1981.

Klein. D.C. & Moore. RY. Pineal N-Acetyltransferase and
Hydroxyindole-O-Methyltransferase. Control by the
Retinohypothalamic Tract and the Suprachiasmatic Nucleus.
Brain Res.. 174:245-262. 1979.

Klein. D.C. Circadian Rhythms in the Pineal Gland. In: Krieger. D.T..
(Egd.£)9. Endocrine Rhythms. Raven Press. New York. pp. 203-223.
I 7 .

Knox. G.V.. Cam bell. C. & Lomax. P. Cutaneous Temperature and Unit
A38? 3126 8%alamic Thermoregulatory Centers. Exp. Neural.
4 : 1 - . .

Krettek. J .E. & Price. J.L. Am gdaloid Projections to Subcortical
Structures Within the asal Forebrain and the Brainstem in the
Rat and Cat. J. Comp. Neural. 178:225-254. 1978.

116

Larsson. K. Features of the Neuroendocrine Regulation of Masculine
Sexual Behavior. In: Beyer. C. (Ed.). Endocrine Control of Sexual
Behavior. Raven Press New York. pp. 77-163. 1979.

Lehman. M.N.. Bittman. E.L. & Winans-Newman. S. Role of the
Hypothalamic Paraventricular Nucleus in Neuroendocrine
Responses to Daylength in the Golden Hamster. Brain Res..
308:25-32. 1984.

Lehman. M.N.. Powers. J .B. 81 Winans. S.S. Stria Terminalis Lesions
Alter the Temporal Pattern of Copulatory Behavior in the Male
Golden Hamster. Behav. Brain Res.. 8:109-128. 1983.

Lehman. M.N.. Winans. S.S. & Powers. J .B. Medial Nucleus of the
Amygdala Mediates Chemosenso Control of Male Hamsters
Sexual Behavior. Sci. 210:557-5 . 1980.

Levine. J .D.. Weiss. M.L.. Gosin. D.. Rosenwasser. AL. 8: Miselis. RR
Re-Examination of the Retina-Hypothalamic Projections of the Rat
Using HRP Conju ated to Cholera Tordn (CT- ). Soc. Neurasci.
Abstr.. 12:549. 1 86.

Lindvall. O.. Bjorklund. A. & Ska erburg. G. Dopamine-Containing
Neurons in the Spinal Cor : Anatomy an Some Functional
Aspects. Ann. Neural. 14:255-260. 1983.

Lisk. RD. The Estrous Cycle. In: Siegel. H.I. (Ed.). The Hamster:
Rgggaductian and Behavior. Plenum Press. New York. pp. 23-51.
I .

Luiten. P.G.M.. TER Horst. G.J.. Karst. H. 8: Steffens. AB. The Course of
Paraventricular Hypothalamic Efferents to Autonomic Structures
in Medulla an Spinal Cord. Brain Res.. 329:374-378. 1985.

Luiten. P.G.M.. TER Horst. G.J. 8r Steffens. AB. The H othalamus.
Intrinsic Connections and Outflow Pathways to e Endocrine
System in Relation to the Control of Feeding and Metabolism.
ng. Neurabial. 28:1-54. 1987

Maragos. W.F.. Newman. S.W.. Lehman. MN. «St Powers. J .B. Neurons
of Ori {in and Fiber TYajectory of Amygdalofugal Pro ections to the
Medi l Preoptic Area in Syrian Hamsters. J. amp. euraL.
280:59-71, 1989.

Melanie. E. 81 Kittrell. W. Lesions of the Su rachiasmatic Nuclei and
Surrounding Basal Hypothalamus olish the Circadian Rhythms
of Activity. DMg. and Bad; Temperature in Female Rats. Soc.
Neurosci. Abstr.. 13:419. 198 .

Merchenthaler. I.. Gores. T.. Setalo. G.. Petrusz. P. 8: Flerko. B.
Gonadotrogin-Releasing Hormone (GnRH) Neurons and Pathways
in the Rat rain. Cell 7155. Res.. 237:15-29. 1984.

 

117

Mesulam. M.-M. Principles of Horseradish Peroxidase
Neurohistochemistry and Their Applications for TYacing Neural
Pathways-Axonal Transport. Enzyme I-Iistochemistry and Light
Microscopic Analysis. In: Mesulam. M.-M.. (Ed.). Tmc Neural
Connections with Horseradish Peraxidase. New York. Jo Wiley
8: Sons Ltd.. pp. 1-151. 1982.

Miernicki. M.. Pos ichal. M.. Karp. J. 8: Powers. B. Photoperiodic
Effects $387318 Hamster Sexual Behavior. Can. Reprad. Behav.
str.. .

Morin. L.P. 8r Zucker. I. Photoperiodic Regulation of Copulato?
Behavior in the Male Hamster. J. Endocrinol. 77:249-2 8. 1978.

Moore. RY. Central Neural Control of Circadian Rhythms. In: Ganong.
W.F. 8: Martini. L.. (Eds.). [Wonders in Neuraendacrinolagy. New
York. Raven. Vol. 5. pp. 185-206. 1978.

Napoli. A.. Powers. V.B. 8r Valenstein. E.S. Hormonal Induction of
Behavioral Estrus Modified by Electrical Stimulation of
Hypothalamus. Physiol. Behav.. 9:1 15-1 17. 1972.

Numan. M. Neural Basis of Maternal Behavior in the Rat.
Psychoneuraendacrinol. 13:47-62. 1988.

Nunez. A.A.. Brown. M.H. 81 Youn strom. T.G. H thalamic Circuits
Involved in the Regulation 0 Seasonal an ircadian Rhythms in
Male Golden Hamsters. Brain Res. Bull. 15:149-153. 1985.

Paxinos. G. 8: Watson. C. The Rat Brain in Stereotaxic Coordinates.
New York. Academic Press. 1982.

Pfaff. D.W. & Modianos. D. Neural Mechanisms of Female Reproductive
Behavior. In: Adler. N.. Pfafi'. D. & Gay. RW. (Eds.). Handbook of
Behavioral Neurobiology & Reproduction. Plenum Press. New York.
pp. 423-480. 1985.

Pickard. G.E. & Silverman. A.J. Direct Retinal Projections to the
H othalamus. Piriform Cortex. and Accessory Optic Nuclei in the
Go den Hamster as Demonstrated by a Sensitive Anterograde
Hogserggish Peroxidase Technique. J. Comp. Neural. 196:155-
17 . l l .

Pickard. G.E. 8r Turek. F.W. The H thalamic Paraventricular Nucleus
Mediates the Photo eriodic ontrol of Reproduction But Not the
Effects of Light on e Circadian Rhythm of Activity. Neurosci.
Lett.. 43:67-72. 1983.

Pickard. G.E. Bifurcating Axons of Retinal Ganglion Cells Terminate in
the Hypothalamic Suprachiasmatic Nucleus and the
12nterggn81gulate Leaflet of the Thalamus. Neurosci. Lett.. 55:2 1 1-

17. 1

118

Pickard. G.E.. Ralph. M.R 8: Menaker. M. The Intergeniculate Leaflet
Partially Mediates Effects of Light on Circadian Rhythms. J. Biol.
Rhythms. 2:35-56. 1987.

Pieper. D.R. Newman. S.W.. Lobocki. C.A. 8: Go ola. G. Bilateral
T‘ransection of the Lateral Olfactory Tract ut not Removal of the
Vomeronasal Organs Inhibit Short-Photoperiod-Induced
gagicréléag Regression in Golden Hamster. Brain Res.. 485:382-

. 1 .

Poulain. P. & Carette. B. Low-Threshold Calcium Spikes in
Hypothalamic Neurons Recorded Near the Paraventricular
Nucleus In Vitro. Brain Res. Bull. 19:453-460. 1987.

Powers. B. & Valenstein. E.S. Sexual Rece tivi : Facilitation b Medial
Preoptic Lesions in Female Rats. . 17 :1003-1005. 1 72.

 

Powers. J .B.. Newman. S.W. 8r Ber andy. M.L. MPOA and BNST Lesions
in Male Syrian Hamsters: Di erential Effects on Co ulatory and
ngmoinvestigative Behaviors. Behav. Brain Res.. 3: 181-195.

1 7.

Price. J .L., Russchen. RT. 8: Amaral. D.G.. The Limbic Re on. 11: The
Am gdaloid Complex. In: B orklund, A.. Hokfelt. T. Swanson.
L. .. (Eds.). Handbook of hemical Neuraanatamy. Vol 5:
Integrated Systems of the CNS, Part I. Hypothalamus.

Iligirépzacampus. Amygdala. Retina. Elsivier. New York. pp. 279-388.

 

 

 

Reuss. S.. Johnson, RR. Morin. L.P. 8r Moore. RY. Localization of
Spinal Cord Preganglionic Neurons Innervating the Superior
Cervical Ganglion in the Golden Hamster. Brain Res. Bull.
22:289-293. 1989.

Rollag. M.D. & Niswender. G.D. Radioimmunoassay of Serum
Concentrations of Melatonin in Sheep Ergosed to Different Light
Regimens. Endocrinal. 98:482-489. 197 .

Rollag. M.D.. O'Callaghan. P.L. & Niswender. G.D. Serum Melatonin
' Concentrations During Different Stages of the Annual
Reproductive Cycle in Ewes. Biol. Reprod.. 18:279-285. 1978.

Scalia. F. & Winans. 8.8. The Differential Projections of the Olfactory
Bglbaangéliccggssory Olfactory Bulb in Mammals. J. Comp. Neural.
1 1: l- . l .

Seegal. RF. 8: Goldman. B.D. Effects of Photoperiod on Cgelicity and
Serum Gonadotrcgins in the Syrian Hamster. Biol prod.
12:223-231. 197

Shapiro. R. E. & Miselis. RR. The Central Neural Connections of the
Area Postrema in the Rat. J. Comp. Neural. 234:344-364. 1985.

119

Shibata. S.. Liou. S.Y.. Ueki. S.. Oomura. Y. Influence of Environmental
Light-Dark cle and Enucleation an Activity of Su rachiasmatic
Neurons in S ce Preparations. Brain Res.. 302:75- I. 1984a.

Shibata. S.. Oomura. Y.. Hattori. K. 8: Kita. H. Res onses of
Suprachiasmatic Nucleus Neurons to Optic erve Stimulation in
R38 thpothalamic Slice Preparation. Brain Res.. 302:83-89.
l 4 .

Sikes. RW. 8r Vogt. B.A. Afierent Connections of Anterior Thalamus in
Rats: Sources and Association with Muscarinic Acetylcholine
Receptors. J. Comp. Neurol. 256:538-551. 1987.

Simerl . RB. 81 Swanson. L.W. The Organization of Neural In uts to the
affilialglgrgoptic Nucleus of the Rat. J. Comp. Neural. 8:312-

Sisk. C.L. 81 Stephan. F.W. Phase Shifts of Circadian Rhythms of
Activity and Drinking in the Hamster. Behav. Neural Biol.
33:334-344. 1981

Smale. L.. Cassone. V.M.. Moore. RY. 8: Morin. L.P. Paraventricular
Nucleus Projections MediaflnfilPineal Melatonin and Gonadal
Res onses to Photoperiod in e Hamster. Brain Res. Bull.
22: 63—269. 1989

Smithson. K..E. MacVicar. BA. 8: Hatton. G.I. Polyethylene GI col
Embedding: a Technique Com atible with Immunocytoc emistry.
Enzyme Histochemistry. Histo uorescence. and Intracellular
Staining. J. Neurosci. Meth. 7:27-41. 1983.

Swanson. L.W. & Kuypers. H.G.J.M. The Paraventricular Nucleus of the
H othalamus: Ctfitoarchitectonic Subdivisions and Organization
a Projections to e Pituitary. Dorsal Vagal Complex. and Spinal
Card as Demonstrated by Retrograde Fluorescence Double-
Labelling Methods. J. Comp. Neurol. 194:555-570. 1980.

Swanson. L.W. The H othalamus. In: Bjorklund. A.. Hokfelt. T. 8r
Swanson. L.W.. Eds.). Handbook of Chemical Neuraanatamy. Vol
5: Integrated Systems a the CNS. Part I. Hypothalamus.
Hippocampus. Amygd Retina. Elsivier. New York. pp. 1-124,
1987.

Takahashi. L.K. 81 Lisk. RD. Dual Progesterone Action in Dience halon
Facilitates the Induction of Sexual Receptivity in Estrogen-gamed
Golden Hamsters. Physiol Behav.. 44:741-747. 1988.

Tamarkin. L.. Baird. C.J. & Almeida. O.F.X. Melatonin: A Coordinating
Signal for Mammalian Reproduction. Sci. 227:714-720. 1985.

Tamarkin. L.. Reppert. S.M. & Klein. D.C. Regulation of Pineal
Melagtonin in the Syrian Hamster. Endacrinol. 104:385-389.
197 .

120

Tamarkin. L.. Westrom. W.K.. Hamill. AI. 8: Goldman. B.D. Effect of
Melatonin on the Re roductive Systems of Male and Female
Syrian Hamsters: A iurnal Rhythm in Sensitivity to Melatonin.
Endacrinol. 99:1534-1541. 19 6.

Tucker. D.C. 81 Saper. C.B. Specifici of Spinal Projections mom
Hypothalamic and Brainstem eas which Innervate Ssympathetic
Preganglionic Neurons. Brain Res.. 360:159-164. 198 .

Turek. F.W. 81 Losee. S.H. Melatonin-Induced Testicular Growth in
Golden Hamsters Maintained on Short Days. Biol Reprod.
18:299-305. 1978

Vanecek. J. & Illnerova. H. Effect of Light at Night on the Pineal Rhythm
in N-Acetyltransferase Activity in the S an Hamster
Mesocricetus auratus. Experientia. 38: 13-514. 1982.

Vaughan. M.K.. Powanda. M.C.. Brainard. G.C.. Johnson. L.Y. & Reiter.
RJ . Effects of Blinding or Afternoon Melatonin Injections on
Plasma Cholesterol. Triglycerides. Glucose. TSH and Thyroid
Hormone Levels in Male and Female Syrian Hamsters. In: Reiter.
R.J. (Ed.). The Pineal and Its Hamranes. Alan R. Liss. Inc.. New
York. pp. 177-186. 1982.

Wade. G.N. Dietary Obesi in Golden Hamsters: Reversibility and
Effggts of Sex and P otoperiod. Physiol Behav.. 30:131-137.
19

Watson-Whitmyre. M. 8r Stetson. M.H. Simulation of Peak Pineal
Melatonin Release Restores Sensitivity to Evening Melatonin
Irgections in Pinealectomized Hamsters. Endocrinal. 112:763-
7 5. 1983.

Watts. A.G.. Swanson. L.W. and Sanchez-Watts, G. Efferent Projections
of the Suprachiasmatic Nucleus: I. Studies Using Anterograde
Transport of Phaseolus vulgarls-Leucoagglutinin in the Rat. J.
Comp. Neural 258:204-229. 1987.

Werner. J 81 Bienck. A. The Significance of Nucleus Raphi Dorsalis and
Centralis for Thermoafferent Signal TYansmission to the Preoptic
Area of the Rat. Exp. Brain Res.. 59:543-547. 1985.

Westlund. K.N.. Bowker. RM.. Ziegler. M.G. & Coulter. J .D. Origins of
Spinal Noradrenergic Pathways Demonstrated by Retrograde
Trans art of Antibod to Dopamine-B-Hydroxylase. Neurosci.
Lett.. 5:243-249. 1 81.

Wiegand. S.J. & Terasawa. 1. Discrete Lesions Reveal Functional
Heterogeneity of Suprachiasmatic Structures in Regulation of
Gonadotropin Secretion in the Female Rat. New'aendacrinol.
34:395-404. 1982.

 

121

Wurtman. RJ .. Axelrod. 82 Fischer. J .F. Melatonin Ssynthesis in the
Pineal Gland: Effect of Li t Mediated b the ympathetic
Nervous System. Sci. 14 :1328-1329. 964.

Yamadori. T.. Nakamura. T. 81 Takami. K. A Study on the Retinal
Ganglion Cell which has an Uncrossed Bifurcating Axon in the
Albino Rat. Brain Res. 488:143-148. 1989.

Yanovski. J .. Witcher. J .. Adler. N.. Mark . SP. 8: Klein. DC.
Stimulation of the Paraventricular ucleus Area of the
Hypothalamus Elevates Urinary 6-H droxymelatonin During
Daytime. Brain Res. Bull. 19:129-1 . 1987.

Yellon. S.M.. Bittman. E.L.. Lehman. M.N.. Olster. D.H.. Robinson. J.E.
8r Karsch. F.J.. Importance of Duration of Nocturnal Melatonin
Secretion in Determining the Reproductive Res onse to Inductive
Photoperiod in the We. Biol Reprod.. 32:523- 29. 1985.

Youngstrom. T.G. 8r Nunez. A.A. Neurons in the Su rachiasmatic Area
are Labelled After Intravenous In ections of orseradish
Peroxidase. Exp. Brain Res.. 87: 27-130. 1987.

Zieglgansberger. W. Actions of Amino Acids. Amines and Neuropeptides
on Target Cells in the Mammalian Central Nervous System. In:
Buijs. R.M.. Pevet. P. 8: Swaab. D.F. (Eds.). Chemical
TYansmission in the Brain. The Role of Amines. Amino Acids and
Pe tides. Prag. Brain Res.. Elsevier. Blamed. Press. Amsterdam.
5 :297-320. 1982.

"Illllillllilillll