Page 1
CHARACTERIZATION OF THE ASCENDING MICTURITION PATHWAY TO THE PERIAQUEDUCTAL GRAY
Myto Duong
Submiüed in partial fiWhnent of the rqukmen& for the d e p of Master of Science
Dalhousie University Halifax, Nova Scotia
September, 1997
O Copyright by Myto Duong, 1997
Page 2
National Library Bibliothèque nationale du Canada
Acquisitions and Acquisitions et Bibliographie Services services bibliographiques
395 Wellington Street 395, nie Wellington OttawaON K 1 A W ûttawaON K1A ON4 Canada Canada
The author has granted a non- exclusive licence allowing the National Library of Canada to reproduce, loan, distribute or sell copies of this thesis in microform, paper or electronic formats.
The author retains ownership of the copyright in this thesis. Neither the thesis nor substantial extracts fkom it may be printed or otherwise reproduced without the author's permission.
L'auteur a accordé une licence non exclusive permettant a la Bibliothèque nationale du Canada de reproduire, prêter, distribuer ou vendre des copies de cette thèse sous la fome de microfiche/film, de reproduction sur papier ou sur format électronique.
L'auteur conserve la propriété du droit d'auteur qui protège cette thèse. Ni la thèse ni des extraits substantiels de celle-ci ne doivent être imprimés ou autrement reproduits sans son autorisation.
Page 3
I would like to dedicate this thesis to Arron,
and my family for their unconditional love and suppon
Page 4
TABLE OF CONTENTS
Page Number
Table of Contents ........................... ... ....................................... v ...
List of Figures ............ ..., ...................................................... vil1
List of Tables ......................................................................... x
............................................................................... Abstract xi ...
Ab breviations ....................................................................... ..xi11
Acknowledgements .................................................................. xvi
................................................................. INTRODUCTION. . - 1
Bladder Overview ............ .. ................................................ 1
Efferent Pathways ............................................................... 2
The Role of Barrington's Nucleus in Micturition .............. ......... 3
M-region and L-region of the Pontine Micturition Center .................. .5
Other Connections to and from Barrington's Nucleus ....................... 6
B ulbospinal Pathway for Micturition ......... .. ........................... 7
Suprapontine Influences on the Michirition Reflex Pathway .............. -8
Afferent Pathways ........... .,, ...................... ., ... 10
Bladder Afferent Terminais in the Sacral Spinal Cord ...................... 11
Spinobulbar Pathway for Micturition ......... ........... . . . 1 2
The Periaqueductal Gray (PAG) ................ .... ..................... 14
Evidence for PAG Involvement in Micturition ........... .,,, .............. 16
Researc h Proposal
.Rationale ................................................................ 17
-Airns and Surnmary of Experimental Preparation .................. 18
Page 5
MATERIALS AND METHODS ................................................. 21
Neuroanatomical Experiments:
............................. Asep tic surgicd preparation for tracer injection 21
.............................................. PAG retrograde tracer injections 21
Bladder anterograde tracer injections ......................................... 22
.................................................................... Post-surger y -22
................................. Experimen ta1 design: survival days (staged) 22
.................................. Surgical preparation for bladder distension 24
.............................................................. Bladder distension 24
..................... .............. Perfusion and tissue collection ....... 25
Sectioning the PAG and spinal cord .................... ... .............. 25
......................................................... Immunohistochemistry 26
................................................................... Data Handling 27
NeurophysiologicaI Experiments:
...................................... Surgical preparation .......... ...... 27
Mapping of the PAG - extracellular field potentials ......... .. ........... 29
Mapping of the Spinal Cord - single units ................................... 31
Histology ........................................................................ 33 . . ................................................................... Statistical Test 34
RES ULTS ............ - .................. ........ 35 .... Neuroanatomical Experiments:
....................... ............ PAG retrograde tracer injections .... 35
Page 6
Primary afferent terminal distribution in spinal cord ........................ 37
c-Fos induction by bladder distension ........................................ 37
Neurophysiologieal Experiments:
PAG extracellular field potentials ............... ... ....................... 50
Spinal Cord Units ............................................................... 63
Spinal neurons antidromicdy activated from the PAG ............ 66 Convergence of afferent input ......................................... 73 Effects of PAG stimulation on spinal neuron activity .............. 74
DISCUSSION ........................................................................ -78
Neuroanatomical Experiments:
c-Fos protein induction by bladder distension ............................... 85
Neurophysiological Experiments:
PAG extracellular field po ten tials ........... .... ......................... 88
............................................................. Spinal cord units -90
................................ ............................ Conclusion .. 97
........ Significance .. ...... .. ...................................................... 97
APPENDICES .......................................................................... 99
REFERE NCES ......................................................................... 103
Page 8
Figure 17: Distribution of Lumbosacral Neurons Activated by PAG ............. 68 Figure 18: Sacral Spinal Neuron Which Rojects to the PAG Conveys PLN
Input .......................................................................... 70
Figure 19: Spinal Neuron, Which Receives SFP Input, Projects to PAG ....... 72
Figure 20: PAG Orthodromically Activated Spinal Neuron ........................ 75
Figure 21: Distribution of Spinal First and Higher Order Intemewons in
Sacral Spinal Cord .......................................................... 76
Figure 22: PAG and PLN Interactions ............................................... 77
Page 9
LIST OF TABLES
Page Number
Table 1: Tracer Injections ............................................................. 23
.............................................................. Table 2: HRP CeU Counts 39
Table 3: Sum of c-Fos hmunoreactive Cells in Each Spinal Cord
................................................................... Segment 45
Table 4: Sum of c-Fos Immunoreactive Cells in Different Zones of Each
Spinal Cord Segments ................................................... 48
Table 5A: Median Latencies for Different Components of the Multiphasic Field
Potential Responses .............. .. .............................. 53
Table 5B: Median CVs for Different Components of the Multiphasic Field
Potential Responses ........... .. ............................... 54
Table 6A: Median Latencies for the 2nd Component of the Field Potentiai
........ Repsonse for Ipsilateral and Contralaterd PLN Stimulation 55
Table 6B: Median CVs for the 2nd Component of the Field Potenriai
........ Repsonse for Ipsilateral and Contralateral PLN Stimulation 56
Table 7: Counts of PAG Antidromicdy and Orthodromically Activated
Neurons in the Sacral Cord ........................................... 67
Page 10
ABSTRACT
The micturition reflex pathway is a supraspinai pathway. There is recent
anatomiciai evidence to suggest that the periaqueductal gray is involved in the ascending
limb of this reflex We propose to determine whether the periaqueductal gray conveys
bladder-relateci information and to characterive the (type of) information that is king
conveyed.
Injections of antemgrade tracer into the bladder wall revealed bladder eerent
terminah in S2 of the sacral spinal cord. These aEferent tenniLlals were within the
vicinity of spinal neurons retrogradely labelled from the periaqueductal gray. Bladder
distension induced c-fos exp-on in neurons located within the regions where
retmgradely labelled neurons were located. This provides some support for the
transmission of bladder inputs via axons of tbese periaqueductal gray -projecting
neurons.
Stimulation of pelvic nerves evoked maximum field potentials in dorsolateml and
ventrolaterai regions of caudal periaqueductal gray. Since the regions activated by pelvic
nerve stimulation differed from those activated by seasory pudendal and superficial
perineal stimulation, the possibilty of sWc pathways for different nerve inputs exists.
However, these fhdings are not conclusive.
Extmcellular single unit studies were undertaken to characte& the inputs
received by newons projecting to the periaqueductal gray. Seventeen c e h were
identiiïed which projected directly nom the sacral spinal cord to the periaqueductal gray,
and only one of these received pelvic nerve input. Although the specifïty of this pathway
was not detennined because this neuron was lost before the other nerves could be tested,
it was observed that at least a portion of this pathway conveyed pelvic nerve
information. This demonstrates that the periaqueductal gray is hvolved in the
Page 11
transmission of bladder inputs. However, the fact that only 4 out of the 17 cells with
direct projections to the periaqueductal gray received inputs h our test ne- implies
that the major part of this pathway is not dhctly related to lower urinary tract functions.
Page 12
ABBREVIATIONS
AP
CCF
CD
Qn
cm 5 0
CV
DAB
DC
hr
H R P
Hz
im.
i.v.
kg
kHz
L
L-PL
LDH
L W
M
m/sec
mA
caudai cutaneous femoral
cord dorsum
centimeter
centimeter of water
conduction velocity
diaminobemdine
dorsal commissure
hour
horseradish peroxidase
Hertz
in tramuscular
in travenous
kilogm
kilohertz
lit=
left pelvic
lateral dorsal horn
lateral ventral hom
molarity
meters per second
milliamperes
xiii
Page 13
MDH
mg
min
ML
mL
mm
H g
MOhms
mPUD
mSec
NaCl
NaNO,
4 p.0,
PAG
PL
PLN
P m
PUDN
R-PL
sec
SFP
SPN
SPUD
WGA-HRP
w
medial dorsal hom
milligram
minutes
mediolateral
rnilliliter
millimeter
millimeter of mercury
megaohms
motor pudendd
milli,ceconds
sodium chloride
sodium nitrite
oxm=n
oral
periasueducd gray
pelvic
peivic nerve
pudendd
pudendal nerve
right pelvic
seconds
superficial perineal
sacral parasympathetic nucleus
sensory pudendal
wheat gem agglutinin horseraàish pemxidase
microamperes
xiv
Page 14
CLL
CLm
SC
IC
NRM
Lm'
N.Aug.
CAq
CPed
FL
mimliter
micrometer
superior coUiculus
inferior coIlicuIus
nucleus raphe magnus
latedorsal tegmental nucleus
nucleus augustus
cerebral aqueduc t
cerebek peduncle
fasciculus longitudinus
Page 15
ACKNOWLEDGEMENTS
1 would lilie to take this opportuni~ to achowledge J e d e r Martin for allowing
me to use data c o k t e d during her rotation shidy with Dr. Downie. Some of these data
included 3 neurophysiologicai a h a l studies and 1 neuroanatornical (conml animai in
the c-fos induction) study.
1 would Iike to thank Dr. Huan-Ji Du and Scott Pmnych for their technicd
expertise. They were both instrumental in developing my understanding of the ber
aspects involved in my neurophysiological and neuroanatomical saidies.
To Dr. Kanie Semba and lab members, a sincere thank you, for the use of theV
microscope and cornputer softwares, and helpfd instnictions. 1 would also like to thank
Dr. Gary Ailen for his ideas, constructive cnticisms, and teaching.
On a personal level, 1 would like to thaak my good fi5ends Karen, Cindy, Mke,
Mostafa, Vidya, Leslie, Isabel, and the Sears family for their support, helpfid advices,
and companionship.
And las& but most importantly, 1 would iike to express my appreciation and
gratitude to my supenisor, Dr. John Downie for his patience and understanding. His
enthusiasm for science and vast scientinc knowledge have been inspiratiod. 1 am also
grateful for the many hours he devoted to reading and editing my work and training me.
xv i
Page 16
INTRODUCTION
Bladder Overview
The urinary bladder is a distensible hoDow organ cornpriseci of a smooth muscle
layer, the detmor muscle. lying between outer serous and inner mucous coats. Aithough
the bladder detrusor is anatomically continuous, it is neuromorphologically divisible into
body (the storage chamber) and base (inlet and outlet) regions. The junction of bledder and
urethra is referred to as the bladder neck (intemal sphincter). The two main functions of the
bladder are to store and to periodically eliminate urine (Elbadawi, 1996).
Urine is produced in the kidney and fiows to the bladder via the ureters. As the
bladder fills, the iension of the bladder wali adjusts to its capacity. so that there is minimal
change in the bladder pressure. Pudendal neme (l?UDN) activity keeps the extemal
sphincter closed Continued bladder f i lhg results in a progressive inmase in the afferent
nerve activity and a gradud increase in sensation. Fit there is a sensation of fullness and
then the desire to void.
The temis "micturition" or 'iirination" refer to the periodic cornplete emptying of the
bladder (involving coordinated bladder contraction and urethral sphincter relaxation). When
the bladder is full, mechanoceptors (stretch receptors) in the bladder neck and detrusor are
activated and impulses are sent to the sacral spinal cord via the pelvic nerve (PLN).
Anerent information from the bladder and the urethra reaches Barrington's nucleus
(Bmington. 19 15) or the M-region in the dorsolaîeral pontine tegmentum (Holstege et al.
1986). the organizing center for the coordination of bladder function The basic micturition
reflex pathway was thought to involve the propagation of information h m the bladder to
the sacml spinal cord to the Barrington enter, and then back to the sacral spinal cord to
elicit a bladder response.
Page 17
If mictuxition does not occur, continueci bladder fîiiing leads eventually to a feeling
of urgency followed by pain and severe discodort. The sensation of bladder W g . of
conscious desire to void, and of painful distension are mediated by aEerait fibers in the
PLN (Nathan and Smith, 1951).
Efferent Pathways
In order for the bladder and mthra to serve their functions (to store mine without
leakage at Iow pressures and to expel urine periodically through a relaxed outlet), the neural
events in the autonomie and somatic systems must be cooniinated. These processes are
controlled by the interplay of sympathetic and parasympathetic neurons and the somatic
motor neurons. A failure to coordinate these events results in increased postvoid residud
urine with high resting pressures in the bladder. Los of rend function occurs when high
intravesical pressures are transmitted to the upper urinary tract. Other manifestations
include irritative voiding or urinary incontinence (Chai and Steers, 1996).
The preganglionic neurons controlling the bladder are found in the intermediolateral
region of the sacral spinal cord and their axons project to the blad&r via the PLN. When
the bladder is empty the parasympathetic preganglionic neurons are quiescent but they are
activated by bladder distension via stretch receptors at micturition threshold (Vaughan and
Satchell, 1994).
Low levels of activity in the sensory afferents that mpond to bladder distension
will activate sympathetic preganglionic neurons in the lumbar spinal cord which resdt in:
(a) inhibition of parasympathetic activity in the pelvic ganglion via alpha-adrenoceptors
(dffiroat and Kawatani, 1985). (b) relaxation of the detnisor via beta-adrenoceptors
(Nergardh and Boreus, 1972; Levin and Wein, 1979; Edvardsen and Setekleiv, 1968;
Awad et al, 1974). and (c) excitation of the intemal urethral sphincter muscle via alpha-
adrenoceptors (Slack and Downie, 1983). Therefore, during bladder f i h g , sympathetic
Page 18
3 activity promotes continence by producing: (a) direct bladder wall relaxation, (b) indirect
blad&r relaxation by inhibithg parasympathetic activity, and (c) by direct contraction of the
intanal sphincter.
The PUDN innervates the extemal urethral sphincter, a muscle that forms part of
the pelvic fioor musculature, which is reguiated by the somatic motor system (Fedirchuk
and Shefchyk, 1991). There are projections h m the dorsolateral pontine tegmentum to
PUD motoneurons innervating the extemal urethral sphincter (Holstege et al, 1986;
Griffths et al, 1990; Dhg et al, 1995). located in the ventrolateral region of Ooufs nucleus
(Thor et al, 1989). Projections h m the dorsornediai part of Onuf's nucleus innemate the
anal sphincter (Thor et al. 1989). These motoneurons are activated at low levels of bladder
distension. At high leveis of distension, supraspinal newons that inhibit the nring of both
the sympathetic and the sornatic motor nemm are activated and result in the removal of the
sympathetic inhibitory effects on the parasympathetic system to elicit bladder contraction
and expulsion of urine.
The Role of Barrington's Nucleus in Micturition
Micnintion is mediated by activation of the sacral parasympathetic efferent pathway
to the bladder and reciprocal inhibition of the somatic pathway to the urethral sphincter.
Electrical stimulation in the rosaal brainstem in the region of the dorsolateml pontine
reticular formation elicits detnisor contractions and firing in the parasympathetic efferent
pathways to the bladder (Lalley et aI, 1972; DeGroat, 1975; McMahon and Spillane, 1982).
w h e m lesion in this area produces irreversible depression of bladder reflexes
(Barrington, 1925; Kum, 1965). Patients with spinal cord injuries above the sacrai cord,
have diniculty emptying their bladders because of uncwrdinated actions of the bladder and
sphincter (detnisor-sphincter dyssynergia) (Blaivas et al, 198 1). Patients with neurologie
lesions rostral to the pons, do not experience this disorder, indicating a coordinating center
Page 19
4
Iocated in the pontine tegmentum (Blaivas, 1982; Holstege, l m ) . Brain lesioning studies
in the cat demonstrated that neurons in the brainstem at the level of the inferior coUcuIus
have an essential role in the control of the parasympathetic component of micturition (Tang
and Ruch, 1956). IatercollicuIar decerebration facilitates micturition by eliminating
inhibitory inputs h m higher centers (Tang and Ruch, 1956). However, micturition is
abohhed by transections below the coilicdi (Tang and Ruch, 1956).
In the cat, bilateral lesions in the rostral pons in the region of the locus c d e u s
(Barrington, 1925; Griffiths et ai, 1990) abolish micturition, whüe electrical stimulation of
these sites aiggers bladder contractions and micturition (Kuru, 1965; McMahon and
Spillane, 1982; dffiroat, 1975). These observations led to the concept of a
spinobulbospinal micturïtion reflex pathway that involves a center in the rostral brainstern
(pontine mictufition center).
The micturition reflex center has been shown to be located in the dorsohteral
pontine tegmental region around the locus coeruleus (Barrington, 1925; Kuru, 1965) but
lesion studies to the locus coenileus demonstrated that micturition reflex œnter in the pons
did not include the locus coenileus region in rat (Satoh et al, 1978). Lesion to an area
rostral to the locus coedeus region in the rat (iaterodorsal tegmental nucleus) did resdt in
urinary disorder (Satoh et al, 1978). In the cat the pontine micturition center has b e n
identiaed with the locus coenrleus (deGroat, 1975), the locus coenileus alpha (Sugaya et
al, 1987) or with a distinct population of neurons located ventromedial to the locus
coedeus adjacent ?O the mesencephalic tract of the trigeminal neme (Tan and Hoistege.
1987)-
Barrington's nucleus, a pontine nucleus ventromedial to the locus coenileus, has
been implicated in micturition (Barrington, 192 1, 1925). Using an anterograde tracer
(Holstege et al, 1986; Blok and Holstege, 1997). this nucleus has been shown to project to
the intermediolateral column of the lumbosacral spinal cord in the region of preganglionic
Page 20
neurons that innervate the pelvic viscera. Furthermore, retrograde tracers injected in the
intermediolatera1 column of the lumbosarral cord labeled neurons in the area corresponding
to Barrington's nucleus (Hida and Shimini, 1982; Holstege et al. 1986; Loewy et al.
1979). Bladder injection with pseudorabies virus (tran~ynaptic tracer) also labeled these
neurons. implicating an involvement in micturition regdation (Nadelhaft and Vera, 1995).
Barrington's nucleus was desctibed in this snidy in the rat as an oval area 0.5 mm X 0.2
mm X 0.25 mm (dorsoventral X mediolateral X rostmcaudal dimensions) centered on a
point 1 mm lateral to the midliae and 0.5 mm below the border of the fourth ventricle
(Nadelhaft and Vera, 1995).
Elecnical stimulation of Barrington's nucleus in rats and cats resulted in bladder
contraction and increased bladder postgangiionic nerve activity (Noto et al, 199 1; Kabat et
al, 1936). Single unit recording in this pontine region identined neurons that responded to
changes in bladder pressure moto et al, 1989). Sugaya et al (1987) elicited micturition by
injecting carbachol to the poniine micturition center. Micturition cm be inhibited by lesions
(Barrington, 1925) or opiate administration to Barrington's nucleus (Wiette et al, 1988).
M-region and L-region of the Pontine Micturion Center
It has been observed that the dorsornedial part of the donolateral pontine
tegrnenaun (M-region) projects preferentially to the spinal parasympathetc nucleus while
the ventrolateral part of the dorsdateral pontine tegmentum (L-region) projects
preferentially to the nucleus of ûnuf (Holstege et al, 1986). The M-region &O overlaps
Bacrington's nucleus and is therefore regarded as the micturition center, while the L-region
is thought to be the control center for the storage of urine. Electrical stimulation of the L-
region elicited a prompt increase in the pelvic floor electmmyogram and urethral pressure
but had little effect on the intravesical pressure (HoIstege et al, 1986; Kun, and Yamamoto,
1964). Electrical stimulation of the pootine micturition refiex center in the cat inhibited
Page 21
6 extemal urethral sphincter activity, and elicited an increase in intravesicai presmre
(Holstege et al, 1986; K m and Yamamoto, 1964). Bilateral lesions in the M-region result
in a long penod of urinary retention, depression of detnisor activity and an increase in
bladder capacity (Grifnths et al, 1990). Bilaterai lesions of the Lregion result in an
inability to retaidstore urine, reduction in bladder capacity, and premature urine expulsion
by detmsor hyperactivity accompainied by urethral relaxation (Grfiths et al, 1990).
During the filhg phase, the L-region exerts a continuous excitatory effect on the nucleus of
Onuf to elicit contraction of the extemal urethral sphincter. mictuxition, the M-region
excites, via a direct pathway, the sacral parasympathetic motoneurom, while iahibiting the
Lregion to allow mictuntion to occur (HoIstege et al, 1986).
Other Connections to and Crom Barrington's Nucleus
Numerous regions such as the forebrain lirnbic structures which have been
implicated in fiuid balance or blood pressure reguiation (Valentino et al, 1994).
neuroendocrllie hinction and reproductive behaviour (Rizvi et al, 1994), project to
Barrington's nucleus.
Besides the sacral parasympathetic nucleus, Barrington's nucleus projects to other
regions such as the dorsal motor nucleus of the vagus nerve, rostral ventrolateral medulla
(Ruggiero and Reis, 1987), locus coedeus (Viuard et al, 1995) and the paravenmcular
thalamic nucleus (ûtake and Nakamura, 1996). Inhibition of reflex biadder contraction by
opioid receptor stimulation in the pontine micturition center results in concomitant inhibition
of increases in blood pressure, hem rate, tidal volume and respiratory rate n o m d y
associated with increases in vesical pressure, but no inhibition of basal cardioresphtory
parameters (WiUette et al, 1988). The inhibition of blood pressure increases associated with
bladder contraction suggests that Barrington's nucleus, although not an integrai participant
in the cardiovascular r d e x pathway, may modulate neurons responsible for
Page 22
cardiorespiratory functions (ie; rostral ventdateral rnedulla, an area essential for
cardiorespiratory regdation) (Ruggiem and Reis, 1987).
Barrhgton's nucleus pmjects to the paraventricular thalarnic nucleus. This nucleus
provides a "limbic" component of basal ganglia-thaIamocortical circuits implicaîed in the
neurocopitive, emotional or visceral concomitants of bebaviours (Otake and Nakarnura,
19%). This subset of neurons in Banington's nucleus which projects to the paraventricular
thalamic nucleus also projects to the sacral parasympathetic nucleus via axon cohterals in
rats (Otake and Nakamura, 19%). Retrograde double labelling aiso demonstrated fibers
h m Barrington's center projecMg to L6 of the spinal cord, tbalamic midline, locus
c o d e u s , subcoeruleus nucleus and sublaîerodorsal nucleus in rat (Otake and Nakamura,
1996).
Bulbospinai Pathway for Micturition
The neurons in the L- and M-regions send their axons to the sacral spinal cord via
specinc tracts. Kum (1965) lesioned areas which could elicit bladder motility upon
stimulation. His study on the locations of lesioned descending tracts resulted in
identification of three descending tracts: the laterai, ventral and medial retidospinal tracts
(Kuru, 1965). In cars, bladkr efferent fibers have also been shown to be located in the
dorsolateral funiculus (Kerr and Alexander, 1964) and lateral huuculus (McMahon and
Momison, 1982a; Fedirchuk and Shefchyk, 1991). Fedirchuk and Shefchyk (1991) found
that axons from the pontine micturition center travel in the lateral reticulospinal tract
occupying a ventral location at the œ ~ c a l levei but shiftng to a dorsolateral position at
thoracic and lumbar leveis. Axonal tracing studies demonstrated that this pathway pmjects
dVectly to the sacral parasympathetic nucleus and to lamina I on the lateral edge of the
dorsal hom Wolstege et al, 1986). an area bat contains dendntic projections fmm the
sacrai preganglionic neurons and pelvic atTerent terminais (Nadelhaft et al, 1980; Morgan et
Page 23
8 al, 1981; deGroat, 1986). Although, the site of tennination of descendhg projections from
the pontine micturition enter are optimally located to reguiate reflex mechanisnu at the
spinal level, it is not known if they make monosynaptic connections with the preganglionic
neurons or segmental interneurons (Yoshimura et al, 199ûa,b).
McMahon and Morrison (1982a) proposed that the descendhg excitatory pathway
h m the pont+ micnintion center is gated in the spinal cord by afferent input b r n the
lower urinary tract However, pontine mictinition center stimulation and stimulation of the
descendhg axons before and after transeciion of the lumbosacral dorsal mots revealed that
the excitatory action of the descendhg pathway was independent of facilitaion h m
primary afferent input (Shefchyk. 1989; Kruse and deGroat, 1992).
Suprapontine Innuences on the Michirition Reflex Pathway
Several discrete regions of the pons and medulla initiate and coordinate lower
urinary tract function (Barrington, 192 1, 1925; dffiroat, 1975; deGroat and Ryall, 1969;
Kum, 1965; Satoh et al, 1978; Tang and Ruch, 1956). In &als and humans with an
intact neuraxis, bladder distension is a primary stimulus for initiating a michrition reflex
involving a supraspinal pathway. Superimposeci on this reflex mechanism is input from
faciliatory and inhibitory centers in the brainstem, and pontine-mesenphalic reticular
formation.
For example, bilateml pamcentrai gyrus lesions cause inability to initiate rnictuntion
voluntarily (Torrens, 1995). Bilateral lesions of the superior frontal gyrus causes
incontinence in the fonn of coordinated voiding with loss of social awareness or sometimes
unconsciousness in the act of voiding (Andrew and Nathan. 1964).
Some studies suggest that the anterornedial amygdala has inhibitory influences on
micturition, and the posterolateral arnygdala and hippocampus are faditatory (Gjone, 1966;
Edvardsen and Ursin, 1968).
Page 24
9 The hypothahmus has predominantly faditatory (dmIaterai) and predominately
iahibitory (ventromedial) areas relating to micturition (Torrens, 1995). The medial preoptic
area of the hypothalamus has been shown to be involved in the conml of male semal
behaviour and pede erection, But this site does not project to the spinal cord (Simerly and
Swanson, 1988). Instead, it projects heavily to the PAG (Rizvi et ai, 1992) and
Barrington's nucleus (Rizvi et ai, 1994). It is possible that these projections regulate
mictuntion reflexes during reproductive behaviour by rnoduiating the activity of the extemal
urethrai sphincter since pseudorabies Wus injection into the external urethrai sphincter
resulted in labelhg in the preoptic area (Nadelhaft and Vera, 1996) while pseudorabies
Wus injection into the bladder did not (Nadelhaft and Vera, 1995).
Electricai stimulation of the pontine micturition center (alpha-locus coeruleus
nucleus) in the rat causai bladder contractions which were inhibiteci by electncal
stimulation of the nucleus reticuiaris pontis oralis (Kimura et al, 1995). Therefore, it was
concluded that the nucleus reticularis pontis oralis has an inhibitory influence on the
functions of the micturi tion center (either via a direct inhibitory pmjec tion from the nucleus
reticulaRS pontis ora l . to the pontine micturition center, to the sacral cord, or via a direct
excitatory projection to the locus subcoedeus nucleus (a urine storage center).
Micturition can occur without the cerebellum (Mshizawa et al, I989), but
stimulation of cerebeiïar nuclei has been shown to have both facilitatory (Chambers, 1969)
and inhibitory effects (Bradley and Teague, 1969; Martner, 1975). The fact that
cerebellectomy had no effect on urethral ekctromyogram, but decreased the threshold
pressure and volume in the collecting phase, and also decreased the micturition pressure
(N'shizawa et al, 1989) suggests that during filling, the cerebellum inhibits michrition but
facilitates it during emptying (Tomns, 1995).
Page 25
10 Afferent Pathways
The body of the bladder and bladder neck are innervated by parasympathedc pelvic
nerves and sympathetic hypogastnc nerves. The extemal sphincter is composed of sniated
muscle and is innervated by the PUDNs. The pelvic, hypogastric and PUDNs ail contain
sensory and motor fibers (afferent and efferent fibers).
The bladder wail ee ren t axons terminate as fke nerve endings within the muscle
fascicles and the perifascicular connective tissue of the smooth muscle and mucous layers
(Fletcher and Bradley, 1970). The feline PLN is composed of small myelinated A-delta
fibers and unmyelinated C-fibers. The srnail myehated biadder wall mechanoreceptor
fibers respond both to active contraction and passive distension (Bahns et ai. 1987).
The hypogastric afferent fibers f o m a bctionally homogenous population of
bladder wall mechanoreceptors, comprising equal numbers of myelinaîed A-delta and
unmyelinatted C-fibers in the cat (Bahns et ai, 1986).
W e the PLN afferent fibers innervate detnisor and bladder neck equally,
hypogastnc aEerent tenninations are largely restncted to the bladder base (Uemura et al.
1973). The PLN tenninals are more concentrated in the smooth muscle layers and the
hypogastric nerve afferent te& are more numerous in the mucosa and submucosa
(Uemura et al, 1974 and 1975).
The hypogastric afferents appear to have ongoing actvity when the bladder is
empty, unWre the PLN mechanoreceptor fibers which are silent when the bladâer is empty
(Bahns et al, 1987).
The function of the vast majority of the unmyelinated PLN bladder afferents
remains obscure (Janig and Koltzenburg, 1990). Some unmyelinated fibers are
chernosensitive, while others are mechanosensitive at high pressures (40-55 cmH20)
(Habler et al, 1990). These fibers are probably involved in mediating visceral pain (Janig
Page 26
11 and Koltzenburg, 1990). Some llnmyelinated fibers can act as cold receptors (Wntter,
197 1) while others are involved in the micturition reflex in spinal cats (de Groat et al.
198 1).
Bladder Afferent Terminais in the Sacral Spinal Cord:
Flurorescent tracers injected into the bladder and urethra resulted in labelhg
prirnarily of S2 dorsal root ganglia (Downie et al, 1984). but bladder aerents also enter
the fzst and third sacral segments (Mawe et al, 1984; Momson, 1987). Application of
horse-radish peroxidase (HRP) to the PLN demonstrated that afferents entering the dorsal
hom of the spinal cord at Lissauer's tract and lamina 1, extend via the lateralcollaterd
pathway II to IV) toward the sacral parasympathetic nucleus in lamina W (Morgan
et al, 198 1) and via the mediai-collateral pathway to extend into the dorsal gray commissure
(lamina X) (deGroat, 1986). The relative density of afferent projections to the lateral
collateral pathway and medial collaterai pathway varies according to the orgaa Projections
to the lateral collateral pathway are prominent for colon, blad&r and uterine pathways,
whereas projections to the medial collaterd pathway and dorsal gray commissure are
relatively weak for colon afferents in cornparison to bladder and uterine pathways
(deGroat, 1986). Injection of different tracers into the bladder and large intestine
demonstrated that sunilar nurnbers of dorsal root ganglion c e k (mainly in S2) innervate the
bladder and large intestine. It was also noted that 46% of this population of neurons was
doubly labeiled and presumably provided an innervation to both organs. However, the
hinctional signincance of a f f e ~ n t newons with receptive fields in two different orgam is
unlmown (deGroa~ 1986; Kawatani et al, 1985). There is no evidence for branching
afferents innervating different regions of the lower urinary tract (Downie et al. 1984).
The functions of the pelvic viscera are closely linked to hinctions of various somatic
stmchires innervated by the PUDN such as the urethrd and anal sphincters (NadeIliaft and
Page 27
12 Booth, 1984). Since the performance of excretory and sexual hinctions requires integration
of viscerai and somatic mechanisms. an overlap of PL and PUD afferent pathway might be
expected to and does occur at ceriain sites in the sacral spinal cord (üeyama, 1984. 1985).
PUD af5erent projections overlap with Msceral afferents in the lamina I and V of the laterai
coilateral pathway. the media1 collateral pathway, and in the dorsal gray conmisme- PUD
afferents atso project to lamiaa II and IV which do not receive significant nurnbers of
viscerd afferents (Roppolo et al. 1985). These projections represeat input from cutaaeous
afferents in the perineum (Brown and Fuch, 1975) whereas the projections to the lamina 1
and V of the lateral collateral pathway are Iikely to represent in part input b m pelvic flwr
muscle (Craig and Mense, 1983) and urethra (deGroat, 1986). Electmphysiological studies
have also shown a prominent convergence of viscerd and somatic aHerent inputs ont0
intemeurons and spinal tract neurons in lamina 1 and V and the dorsal commissure of the
lumbosacral spinal cord (deGroat et ai, 1981, McMahon and M o & u ~ , 1982a,c; Honda,
1985).
Spinobulbar Pathway for Micturition
Demy-Brown and Robertson (1933) proposed a local sacral reflex for rnictuntion,
mediated entirely by the PL and PUDNs. However, the micturition pathway has been
shown to involve a supraspinal component since electrophysiological saidies in cars and
rats indicate that the parasympathetic efferent outllow to the urinary bladder is activated by
a long latency supraspinal reflex pathway (KULU, 1965; deGroat and Ryall, 1969; deGroaf
1975). Electncal stimulation of afferents in the PLN elicits a long latency (120- 140 msec)
discharge measured on the bladder nerves. This corresponds to the sum of the ascending
latency to activation of neurons in Barrington's center (80 rnsec) and descendhg latency to
recording a discharge on the pelvic nerve when stimulahg in Barrington's center (40
Page 28
13 msec) (MaUory et al, 1989; Nom et al. 1991). These hdings are consistent with the notion
of a supraspinial micturition reflex.
nie spinal pathways that transmit sensory information h m the visceral afferent
tenninations in the spinal cord to more rosaal structures can be found in the dorsal, lateral
and ventral columns. Primary afferent cokterals carrying touch and pressure sensation in
the urethra and hocuous sensations fkom the PL floor muscle are conveyed via the dorsal
column (Nathan, 1956). The dorsal column also has a synaptic path, the postsynaptic
dorsal column pathway which has been shown to convey visceral (colocectal not bladder)
and noxious cutaneous inputs to the nucleus gracilius (Al-Chaer et al, 1996). These
ipsilaeral dorsal column projections synapse in the grade nuclei in the medulla (Morgan et
al, 198 1, Ueyama et al, 1985). The lateral columns transmit information concerning
temperature sensation in the urethra, the sensation of bladder fullness and desire to
micturate, sexual sensations, and pain sensations from the bladder, ureihra, lower ureter,
and skin. This pathway is thought to be the spinothalamic tract (Nathan and Smith, 195 1;
Nathan, 1956). Electrophysiological studies in the cat have identined a tract in the dorsal
half of the lateral column near the surface of the spinal cord and a tract in the ventrai and
ventrolateml column which convey infornation fbm the bladder tension receptors (Kuni,
1965; McMahon and Momison, 1982a; Fedirchuk and Shefchyk, 199 1; Fields et al, 1970).
Fibers in the dorsdateral tract in the thoracic spinal cord form the primary pathway for
activating the supraspinal michnition reflex as stimulation of this tract evokes coordinated
rnicturîtion (Fedirchuk and S hefchyk, 199 1).
Spinal tract neurons presumed to conmbute to the ascending limb of the micturition
reflex pathway exhibit long latency discharges (30-40 rnsec) foliowing P M afferent
stimulation (McMahon and Morrison, 1982a). This latency raises possibility that at Ieast
one intemeumn is interposed between primary Serent and spinal tract neurons. Sacral
intemeurons responding at shorter latency (5 msec) have k e n reported (Morrison, 1987).
Page 29
14 Interneu~ons responsive to bladder distension have been identined in both dorsai
c o ~ u r e and the intemediolateral regions of the sacral spinal cord (deGroat et ai, 1981;
McMahon and Morrison, 1982b; Honda, 1985).
In general, the basic rnicturition refiex pathway invoives: (1) activation of
mechanoreceptor in the bladder wail; (2) transmission of bladdex inputs to the sacral cord
via PLN; (3) termination of bladder afferents in the sacral cord (lamioa 1, II, IV, and X);
and (4) bladder impulses ascending in spinal columns to Barrington's nucleus where
micturition is elicited via descendhg tracts to the sacral puasympathetic nucleus, and
activation of efferent fibers in the PLN. The only problem is that there is a pauàty of sacral
spinal projections to Barrington's center @Blok et al, 1994). Therefore it has been proposed
that spinal projections tenninate in other brainstem area(s) before the information is
conveyed to Barrington's nucleus. This brainstem structure should be caudal to the
superior coiliculi (in accordance with lesion studies performed by Barrington, 192 1 and
Kuru, 1965) and have some connectivity with Barrington's nucleus.
There are many structures which are located caudal to the inferior colliculus and
receive numerous spinal projections. An example would be the parabrachial nucleus (Ma et
al, 1989). c-Fos saidies have implicated this nucleus in the processuig of visceral
nociceptive inputs h m the area of the sacral parasympathetic nucleus @ing et al, 1994).
Pseudorabies v ins injection into the bladder wall resulted in labelhg in Barrington's
nucleus, locus coeruleus, paraventricular nucleus of the hypothalamus, medial preoptic
nucleus, and the perhqueductal gray (PAG), but not in the parabrachial nucleus (Nadelhaft
et al, 1992).
The Periaqueductal Gray (PAG)
The PAG extends from the level of the posterior commissure and the oculomotor
nucleus to the level of the dorsal tegmental nucleus. There are different ways of dividing
Page 30
15 the PAG into subdivisions. Beitz (1985) divided the PAG into dorsal, dorsolateraî,
ventroiateral, and medial subdivisions. The medial zone is the region that immediately
surrounds the cerebral aqueduct and the remaining regions are based on the dorsal-vend
dimensions. The PAG is organized in columns: dorsal, dorsolateral, ventral and
ventrolateral columns (Depaulis et al. 1992). Electrophysiological studies provideci
evidence for longitudinal columns arrangements of neurons for defernive reactions in the
rat PAG in the dorsolateral and ventrolateral subregions (Depaulis et al, 1992).
The PAG is involved in at least five major bctions. These include: pain
processing and modulation, vocalization. autonomie regdation (i; cardiovascular control),
fear and anxiety, and lordosis. AU of these fiinctions interact and the PAG is involved in
the integration of these p m e s s e s (Behbehani, 1995).
The PAG is an important site in ascending pain transmission (Magoun et al. 1937).
It receives afferents fmm nociceptive neurons in the spinal cord and sen& projections to
thalamic nuclei that pmess nociception (Yezierski and Schwarz, 1986; Yezierski. 1988).
The PAG is also a major cornponent in the descending pain inhibitory pathway (Reynolds,
1969; Mayer et al, 197 1). Activation of this pathway can inhibit nociceptive neurons in the
spinal dorsal hom.
The dorsal PAG is a major site for processing fear and anxiety (Bander and
Canive, 1988; Bander and Depaulis, 1988). It ioteracts with the amygdala which is the site
responsible for generating fear and anxiety (Bander et al, 1985; Rizvi et al, 1991). Lesion
of the PAG alters fear and aflxiety produced by stimulatiug the amygdala. Lesion of PAG
wiII result in mutism (Mehck et al, 1958; Jurgens and Pratt, 1979 ) while stimulation of
the PAG produces vocdization (Magoun et al, 1937; Jurgens and Pratt, 1979). The PAG is
also involved in lordosis (Sakuma and Pfaff, 1979).
Page 31
16 Evidence for PAG Involvement in Micturition
Anatomical studies suggest that the PAG may be a major target of afferent input
from the s a d spinal cord (Blomqvist and Craig, 1991; Vanderhorst et al. 1996). Blok et
al (1995) found that there were more antemgrade iabeiled tenninals in the PAG than in
Barrington's nucleus after WGA-HRP injection into the lumbosacral cord Projections
from the lateral PAG to Barrington's nucleus has been established (Blok and Hoktege.
1994). Valentino et al (1994) demonstrated ushg a retrograde tracer, cholera toxb subunit
B, a projection h m ventdateral PAG and Banhgton's nucleus. The type of information
canïed by projections from spinal cord to the PAG and then PAG to Bamhgton's nucleus
is unlaiown. However, the injection of pseudorabies vim. a traasynaptic tracer, into the
bladder produced labelling in Barrington's nucleus and the PAG (Nadelhaft et al, 1992),
providing some evidence for PAG involvement in conveying bladder information. Further
support for PAG involvement in micturition can be found in electrophysiological studies.
Stimulation of the PAG produces bladder contraction (Kabat, 1936; Noto et al, 1991).
Noto et al (1991) demonstrated that PLN stimulation produced shorter latency negative
field poientials in the PAG (13-15 msec) than in laterd dorsal tegmentual nucleus
(Barrington's nucleus) (30-40 mec) in rats. The PAG may be a receiving area for
spinomesencephalic tracts carrying afferent information from the urinary bladder and
neurons in the PAG may in tum process and relay this information to the lateral dorsal
tegmental nucleus (Noto et al, 1991)-
Since the PAG is lcnown to be an inkgrative center for numemus autonomie
processes, and it does make the appropriate connections to elicit micturition, there is reason
to believe that the PAG is involved in the rnicturition reflex pathway.
Page 32
Research Proposa1
Rationate
Little is known about the supraspinal control of micturition. h the p a s it was
accepted that Bankgton's nucleus was the controlling center for micturition but there are
several hdings which supports the clah that the PAG is also involved in the processes of
michuition:
(1) PAG stimulation can cause biadder contraction (Kabat, 1936);
(2) PAG is a major target of neurons pmjecting from the sacral spinal cord. (Blomqvkt and
Craig, 199 1 ; Vanderhorst et al. 1996).
(3) There is a substantial projection h m the PAG to Barrington's nucleus (Blok and
Holstege, 1995).
(4) PW stimulation in the rat evoked potentials with shorter latency in PAG than in
Barrington's nucleus (Noto et al, 199 1).
The mode1 that is presented by Blok and Holstege is as follows: bladder afferents
enter the sacral spinal cord, and bladder information is conveyed by spinal projecting
neurons to the PAG. Barrington's aucleus then receives bladder input from the PAG. The
descending limb of the basic micturition pathway consists of neurom projecting from
Barrington's nucleus to the sacral spinal cor& to activate pelvic pregangiionic neurons and
elicit bladder contraction (figure 1).
Although it has been demonsûated that there are spinal cord projections to the PAG,
it is not b o w n whether these projections are involved in bladder or bladder-related
functions. Therefore we will atternpt to detennine whether bladder-reiated information is
carried on this pathway. The hypotheses to be tested are: (1) the sacral spinal cord
projection to the PAG carries bladder-related information (pediaps in addition to other
information); (2) a component of this ascending pathway (Le. some individual neurons)
conveys bladder-s@c infornation.
Page 33
Aims of Experimentai Preparations
Neuroanatornical and neurophysiological approaches will be used to locaiize.
identQ, and characterize bladder sensitive neurons projexting from the sacral spinal cord to
the PAG.
For the neumanatomical studies, three different techniques will be used to establish
the comectivity of neurons conveying bIadder input h m the sacral cord to the PAG:
(1) Retrograde m i n g techniques will be used to establish a direct projection h m the
sacral cord to the PAG.
(2) Induction of the imrnediate early gene, c-fos, by bladder distension @hier and
deGroat, 1992) will be used to identify spinal neurons related to bladder function.
Combining this technique with retrograde aacing wiU enable us to determine whether sacral
spinal ceUs projecting to the PAG are likeIy to carry bladder-related information.
(3) Antemgrade aacing techniques will be used to i d e n m the terminal field distribution of
bladder afferents in the sacrai spinal cord Although not definitive, the results of this type
of study may help in asssessing whether die comection between primary afferents and
projecting neurons is likely to be monosynaptic.
Extensive anterograde and retrograde tracing data were presented by Vauderhorst et
al (1996) while we were in the midst of our studies. Our neuroanatornical work was
abbreviated as a result.
Although, it was asssurned for a long t h e , 3 a t the ascending limb of the
micturition reflex probably synapsed directly in Barrington's nucleus. littIe
neurophysiological work has addressed this issue and the question of the routing of bladder
af£erent information. The characteristics of the neurons carrying bladder afferent inputs also
have not been adequately described.
Page 34
Figure 1 : The Basic Michrition Reflex Pathway. It was thought this pathway involved
Barrington's nucleus with an aEerent and efferent limb h m and to the sacral spinal cord
(A). Recently, Holstege et al (1994) proposed an involvement of the PAG in the basic
micturition rdex pathway (B). SC, superior collicdus; IC, infenor coiliculus; CAq,
cerebral aqueduct; CPed, middle cerebeiiar peduncle; N. Aug, nucleus augustus.
Page 35
For the nemphysiological studies, the four -c objectives are:
(1) to elicit field potentials in PAG with PLN stimulation.
(2) to construct a 3-D map of bladder-sensitive ce& located in the PAG.
(3) to characterize the cells in the sacral spinal cord that pmject to the PAG.
(4) to determine whether any cells projecting h m the sacral spinal cord to the PAG carry
PLN information exclusively. SPUD, SFP and CCF nerves were also stimulated to test the
specincity of this ascending pathway.
These objectives will be approached by using extracellular field potential recordings
in the PAG and extracellular single unit recordings in the sacral spinal cord
Page 36
MATERIALS AND MEXHODS
Neuroanatomicd Experiments
Aseptic surgical preparation for tracer injection
Ten adult cats of either sex, weighing 3.1-5.2 kg, were used for the
neuroanatomical tracing studies. The animais were sedated with a mixture of 2 mglkg
acepromazhe (Atrovet@), 20 mglkg meperidine HCI (Demerol9). and 0.12 mgkg atropine
(im). and anesthetued with 28 mgkg thiopental (i-v.). After tracheal intubation. the cat's
head was shaved and placed in a stereotaxic apparatus. Body temperature was monitored
h g a rectal pmbe and maintained at 37OC using a heating pad layered with blankets.
Durhg the aseptic surgical procedures, reflex statu, heart rate (using an esophageal
stethoscope), and pulse pressure were rnonitored. Anesthesia was maintained with 1 5 2 %
halothane (Fluothane) in Oz (total flow rate of 1.5 Umin). The scalp was scrubbed with
4% chlorhexidine gluconate (Steri-Stat), followed by alcohol (isopropyl) and then a 10%
provioduie solution. A midline scalp incision was made and the muscle was dissected to
expose one side of the skull. Using a hand-held electric dRU @remel. WI), a small
craniotomy was performed to expose a s m d area of the braui. The dura was then opened.
PAG retrograde tracer injections
A glas micropipette secured with dental wax on a Hamilton syringe which was
placed in a stereotmic holder was used to inject tracers. Two to three percent fluoro-gold
(Fluorocbrome Inc.) (n=7), 5% wheat gem agglutinin horseradish peroxidase (WGA-
HRP) (Sigma Inc.) (n=3), or acidic 10% biotin de- (Molecular Probes Inc.) (n= 1) was
Page 37
22 injected into various regions of the PAG (table 1). Some of these animais were ais0 used in
the anterograde tracer and bladder distension studies.
Bladder anterograde tracer injections
A midline abdominal incision was made to expose the bladder. HRP (2.5-1546) or
WGA-HRP (5%) was injecwi at 5-6 injection sites at the base of the right detrusor and
trigone. Injections were made below the serosa without puncturing the bladder. The total
volume injected into the bladder wall rauged fkom 5-44 pL. Seven animals were used in
this part of the study.
Pos t-surgery
Following tracer injection, the head and abdomen were closed in two layers.
Morphine (0.075 mgkg, Lm.) and cefazolin sodium (20 mgkg, i-m.) were given after the
surgery and again 6 hours later. Supplernental doses of morphine were given as required
during the mt pst-operative day. Cefazolin sodium was administered via intramuxular
injection on the fmt day. Cefadroxil monohydrate (Cefa-Tabs@, 22 mgkg, p.o.) was
given twice a day for the next 3-7 days for animals with long survival times.
Experimental design: survival (staged)
Originally the PAG and blad&r injections were staged (n=2) since different tracers
are transported at different rates. PAG was injected with fluoro-gold two weeks prior to
bladder injection with 5% WGA-HRP. The cat was allowed to recover and a week later,
the bladder distension experiment was performed. In later experiments (n=3), where
WGA-HRP and/or Biotin-Dextran were used for the retrograde studies, the s d v a l time
was shortened to 1 week.
Page 38
23
Table 1 : Tracer Injection Plans.
Various tracers and tracer volumes were injected into the PAG (retrograde tracers) and the bladder (anterograde tracers). For cases #773 and W85. the injection rimes for the different tracer injections were staged
Case #
retrograde anterograde
retrograde anterograde
re trograde anterograde
retrograde anterograde
retrograde
retrograde
retrograde anterograde
retrograde an terograde
retrograde anterograde
Tracer
2 % fluoro-gold 2.5% HRP
5% WGA-HRP 5% HRP
2% fiuoro-gold 5% WGA-HRP 15%
2% fluoro-gold 5% WGA-HRP 15% HRP
5% WGA-HRP
2% fluoro-gold 5% WGA-EIRP
3% fluoro-gold 5% WGA-HRP
10 % biotin dextran 2% fluoro-gold
Total Volume In jected
6OnL 10
80 I l . 5w 270 nL 7& 37 CLL 360 nL 8W 44 CrL 360 nL
180 nL 180 ni,
70 nL 10 w 73 nL 3w
240 nL 90 nL
Page 39
24 Surgical preparation for bladder distension
Ketamine (25 mg/@ im.) was used for induction of anesthesia Halothane (14%)
in 1:l nitmgen in O2 at a total flow rate of 2 Umin was useci during the surgicai period The
trachea, carotid artery* and femoral vein were cannalated for amficial respiration, blood
pressure measurement and fluid and dmg administration, respectively. A continuous i.v.
drip (approximately 30 dropdmh) of 5% dextrose in Eünger's solution was maintained
during surgical and experimental periods. Through a rnidline abdominal incision. a
pediairic feeding tube was placed in the dorne of the bladder and secureci with a purse-
string suture. Nitrous oxide and halothane were discontinued at the end of the surgery and
anesthesia was maintained with chloralose (50 mgkg, iv.). Supplemental doses of
chloralose were given as required based on withdrawai responses to toe pad or interdigital
skin pinch.
BIadder distension
The bladder was i d ' with sterile saline using a motor-dnven syringe (Sage
Instruments, mode1 355) at a rate of 15 hourly diuresis units (1 hourly d i m i s unit = 1.1
ml/kg/hr). Bladder pressure was rnonitored on a side arm of the filling line with a pressure
transducer (Micron Instruments) comected to a Gould chart recorder. An abrupt increase in
bladder pressure during bladder nIling was taken to indicate a micturition contraction
because voiduig of fluid through the urethra may not occur rehbly under chloralose
anesthesia (Rudy et al. 1991). The blad&r was drained before the peak of the micaintion
contraction or during any sustained in- in bladder pressure contraction D avoid over-
distension of the bladder. Over a period of two houn. the bladder was repeatedly distended
in 5 animals, or continuousIy drained in one control cat (case W75 1). This control cat did
not receive any injections of either ~trograde or antemgrade tracers.
Page 40
25 Perfusion and tissue collection
In the studies in which fluoro-gold was used as the retrograde tracer and the bladder
was distendeci, the animals were transcardially perfused with 1.0 L of a chilled pre-rllise
solution of NaNO, (0.1% sodium nitrite in phosphate buffered saline, pH 7.4). and h e d
with 1.5-2.0 L of 4% paraformaldehyde in phosphate buffer (pH 7.4). For the studies in
which 5% WGA-HRP was used for the PAG injection, the animals were transcardially
perfûsed with NaNo,, and fmed with 1.00% padormaldehyde with 1 .Z% glutaraldehyde
in phosphate buffer (pH 7.4).
After Iaminectomy, the spinal cord specimens were coilected as blocks: TlGT11,
L5-L6, L7-S 1, and S2Co. In 2 experiments, the spinai cord specimens were coilected as
individual segments (L7-S3). The midbrain was excised following craniotomy. The dorsal
mot ganglia and bIad&r were also coilected. The tissues were post-fixed in 4%
paraformddehyde at 4 OC for 2-3 days, and then transferred into a 20% sucrose solution in
phosphate buffer, pH 7.4. for another 1-2 days for cyroprotection.
Sectioning the PAG and spinal cord
'Ine midbmin and spinal cord segments were marked with knife cut on the left
ventral quadrant and placed on a fro~en stage. Frozen sections of 40 pm thickness were cut
with a sliding sledge microtome into four series. One of the four series was collected in
serial order whüe the other three series were collected in a non-serial order in phosphate
buffer saline solutions. The sections collected in a non-serial manner were then processed
for immunohistochemistry by the batch method.
For the fluoro-gold injections, the PAG and spinal cord serial sections were floated
out onto slides, coverslipped, and viewed under an Olympus microscope with ultraviolet
illumination Viewing of c-fos protein, biotin dextran and WGA-HRP transport Ui the
Page 41
26 spinal cord or their injection sites in the PAG, requkd M e r processing (see next
section). Reacted sections were then fke floated onto gelatin subbed slides and allowed to
air dry for 24 hrs. The sections were dehydrated by subsequent rimes through an
increasing alcohol series, foilowed by two rimes in xylene. The slides were coverslipped
using EnteIlan coverslipping medium.
Immunohistochemistry
c-Fos reaction. To localite the neurms in the spinal cord which were activaîed by
bladder distension, one series of the spinal cord sections used a polyclond sheep anti-Fos
prirnary antibody (Cambridge Research Biochemicals, 1:1000) directed against the c-fos
protein. Sections were M e c processeci and stained according to the avidin-biotin
technique using biotinylated rabbit anti-sheep secondary antibody (Dimensions Lab Inc.,
1 : 1000 dilution). The chromagen, diaminobenzadine @AB)-nickel or DAB-brown, was
used to react with 0.34 hydrogen peroxide to stain and allow visii=iIi7ation of the antibody
bound-nuclear Fos protein product The control expriment (#75 1) and three other
experiments used DAB-brown for c-fos visuahation while two experiments used the
DAB-nickel method. Light microscopie analysis of the slides was perfonned. (See
Appendix for further details)
Retmgmde and a n t e r u g r . îracer imm~7~0histochern~cal reactiomr One series of the
spinal cord sections (antemgrade and retrograde tracing studies) was reacted in the same
manner as c-fos, but using a TMB reaction.
The WGA-HEW Iabelling was viewed under polarized Iight using an Olympus
microscope. Light microscopy was used to visualize the retrograde transport of biotin-
dextmn. For retrograde tracer injections, the PAG was also reacted to determine the
Page 42
27 location of the injection site and the de- of spreadldiffusion of the tracer. (See Appendix
for M e r detatls.
Data handling
Fos-immunoreactive ce& were counted in designated zones of the different spinal
cord segments (L7-S3). The cord was divided into 5 major divisions: medial dorsal hom;
laterai dorsal horn; dorsai commissure; sacral parasympathetic nucleus; ventral hom. These
zones were created by setting the horizontal divison at the apex of the dorsal white matter
and at the ventral border of the central canal, These horizontal divisions were then bisected
with vertical lines to create the quadrants in the dorsal hom. Areas below the central canal
were considered to be the ventral hom zone. The number of c-fos positive ce& in a zone or
spioal segment were expressed as a percent of the total number of fos immunoreactive cells
in an experiment. The percentages for the five bladder distension experiments were then
averageû.
For the tracer studies, the sections. the injection sites and the labelled t e h a i s or
neurons were plotted using an Olympus microscope, Scope Ploüer 1.51 (Scott Pronych.
Dalhousie University) software and a Macintosh cornputer. In the retrograde tracing
experiments, WGA-HRP labelled c e b were counted Erom the 20 sections with the most
labelhg in each spinal cord segment
Neurophysiological Experiments
Surgical preparation
Twenty adult cats (2.9-5.7 kg) of either sex were anesthetized with ketamine
(25mglkg. i n ) . The tmchea was then cmulated for &cial respiration and delivery of
Page 43
28
1-596 halothane in 1:l nitrous oxide in 9 (total flow rate of Wmia). The right femoral
vein and left cmtid artery were cannulated for the adminiseation of 5% dextrose in
Ringer's solution and dmg and for blood pressure monitoring, respectively. A catheter for
recordhg bladder pressure was inserted into the urînary bladder tbrough an incision in the
urethra. 1-2 cm below the bladder ne&. The pressures were measured using pressure
transducers (Micron Instruments Inc.) connected to a chan recorder (Gould Inc.).
Bladder-directed branches of the right and left PWs were exposed through a
midline abdominal incision and fieed of comective tissues. Silver foil electrodes, 2x10 mm
with a 2x2 mm area of the electrode bared, were used to stimulate the PLNs. The
electrodes were secured onto the uretbta with 5-0 polyester suture (Ethibond Ethicon
Su- Ltd). The nerves were placed onto the bard regions on the foil ekctrodes which
were then fol&d over the nerves. The electrodes and the nerves were insulated from the
surrounding tissues with PlastibaseB (Squibb Canada Inc.), or KwikCast (World
Precision Instruments). The abdomen was then closed in two layers.
With the cat prone, the sensory (SPUD) and motor branches (mPUD) of the
pudendal nerve, the superficial p e ~ e a l (SFP) nerve, and the caudal cutaneous femoral
(CCF) nerve were dissected on the left side. When possible, the motor branch of the
pudendal nerve was separated into anal and urethral branches. Inmuiectomy was carried
out nom LAS2 to expose L7 and more caudal spinal cord segments and their spinal roots.
After the surgical preparations, the halothane anesthesia was replaced by chloralose
(50 mglkg. i.v), and supplemental doses of chloralose (each 10-20 mg&, i.v.) were given
a s required based on a withdrawal response to pinching of the interdigital skin or toe pads.
Gallarnine triethiodide (20 mg initially and supplemented as quired, iv.) was used to
irnmobilize the animal during the recordhg session. Assessrnent of the anesthetic level
Page 44
29 under these conditions was based on cardiovascular response to pinching interdigital skin
or toe pads. Body temperature was maintahed at 35-38 OC with a heating pad and lamp.
With the completion of the lamiiiectomy. the cat was transferred from the surgical
table into a stereotaxic hune (David Kopf Instruments, Mode1 1404) and a spinal unit on a
vibration isolation table (Technical Manufkchiring Company. M i m g series 63-500).
Bilateral pneumothorax was performed to demase respiratory movements. SPUD. m m .
SFP. and CCF were mounted on bipolar platinum iridium hook electrodes submerged in a
pool of heated mineral oil fonned by tyhg skin fiaps.
The dura was excised and refiected nom the spinal cord. and the cord was covered
by a heated mineral oil pool. Using a cartilage that nias across L741 junction as a
landmark, the nght S 1 and S2 dorsal mots were identified and traced to their spinal
segments. A silver bail rexording electrode was placed near the SI-S2 junction to record the
cord dorsum p0tentia.L
Access to the brainstem was achieved by craniotomy. In most experiments part of
the cortex and the bony tentorium were removed so that surface landmarks could be used
for more accurate placement of the electrodes. Dexamethasone (2 ml, Lv.) was given to
reduce swelling and thus distortion of brain structures.
Mapping of the PAG - Extracellular Field Potentials
Stimulation of peripheral nerves
The PLNs were stimulated by a single pulse or a train of 3 pulses of 0.5 msec
width at 1000 EZz, with an intensity 2-5 t h e s the threshold canent required to elicit a cord
dorsum potentid or a pelvic-pudendal reflex In the absence of a reflex or cord dorsum
recording, a maximum intensity of approximately 2-3 mA was used The central cut ends
of the other peripheral nerves (SPUD, SFP, and CCF) were stimulated with single pulses
Page 45
30 of 0.2 msec wideth at an intensity of 5 times the threshold ciirrent r e q d to eiicit a cord
dorsum potential
Extracellular field potential recordings in the PAG
Single, monopolar tungsten microeIectrodes with 60 pm-tip diameters and
resistances of 9-12 MOhms were used to search the PAG h m A6.0 to P2.5. Distances
between transverse planes ranged b m 05-1 mm and in each track recordings were made
at depths h m 2-8 mm below the surface of the midbrain. In three experiments
mstrocaudal sûarching was restricted to APO-Pl5 to permit more data to be obtained in
searching in the medioIateral plane. Here, distances between tracks ranged nom 0.5 to 1.0
mm. The signals were amplified (gain=loOOX), fitered (bandpass=l Hz-3kH.z). and
displayed on an oscilloscope. Cord dorsum and PAG field p0tentia.k were digitized at
sampling rates ranging from 2-10 kHz (MacLab Scope, AD Instruments Inc.). The
digitized sigaais were averaged over 16 sweeps and the averages stored on a Macintosh
cornputer.
PAG field potential data handling
Maximum field potential locations for each of the nerves were detemiined and
mapped onto standard outlines of the cat midbrain b m a stereotaxic atlas of the cat brain
(Snider and Niemer, 1961). For the PLN stimulation, our hypothesis assumes that there is
a maximum PLN field potentid which could be elicited in the PAG. Therefore, maximum
responses fuaher than lmm from the border of the PAG were disregardeci and the next
largest field potential located in or within 1 mm of the PAG was considered to be the
maximum.
Page 46
Mapping of the Spinal Cord - Single Units
Stimulation of the PAG
Prior to PAG stimulation, the bladder was distended with 5-1 0 mL of saline. Two
strategis were used to determine the appropriate placement of the PAG stimulating
electrode. ûne strategy involveci eliciting bladder contractions. Sites in the caudal PAG
(AO.IP1.5) were shulated with constant current pulses of 3 mA, at 50 Hz, in search of
an optimum site for evoking PAG-evoked bladder contraction. The other strategy used
PLN-evoked field potential recordings in the PAG to detamine the optimum PAG site for
antidromic studies.
For most experiments, a single stimulating tungsten microelectrode (60 p - t i p
diameter, 9- 12 MOhms) was placeci in the midbrain at a location which when stundated
elicited a rise in bladder pressure (measured by the bladder transducer and chart recorder).
Insulation was removed for a distance of 1130 pn h m the electrode tip to increase the
suface area and reduce the resistance of the electrode. Several arrangements of indifferent
electrodes were tried, but nonnaliy, the indifferent ektrode was placed either on muscle or
cerebellum. In two experiments two stirnulating electrodes were used, separated by a
distance of 1.8 mm h m the indifferent electrode. In this case, one stimulating electmde
penetrated the left PAG while the other penetrated the right, with the indifferent elecaode at
m i m e .
Once a location in the PAG was found that could elicit a large bladder response, the
stïmulating electrode was kept at that PAG location. For the single unit searches in the
spinal cord, the PAG was stimulated with a train of 4-5 0.5 msec pulses of 3 mA inteI1Sity
at 333 Hz. Trains were delivered at an intertrain interval of 2 sec.
Page 47
32 Single unit recordings in the spinal cord
Using a micloelectrode pulier (Narishige Scientinc Instrument Laboratory, PE-2).
recording glas microelectrodes were drawn fiom 1.2 mm X 0.68 mm (outer and b e r
diameter, respectively), 4" hlamented glass capillaries (A-M Systems Inc.). They were then
med with a solution of 4M NaCL A microelectrode manipulator (Tramertex, her ican
Medicd Systems) mounted on a custorn-built Lundberg arch was used to move the
recording electrode to search the spinal wrd for activated cells.
The spinal cord was seanihed in the caudal to mstral direction h m the SYS3
junction to SUS2 junction in steps of 150pn. At each rostmcaudal plane, the
microeiectrode was moved fmm the midline in lateral steps of 150-200 jim untü the dorsal
mot entry zone was reached The search was stopped in a track when a depth of 2ûûû-2700
pm was reached Extracellular si@ were amplified (gain= 1000X). nItered (band pas=
300 H z - 3 W ) and displayed on an oscilloscope. The signals were also digitized at a
sampling rate of 21 kHz and displayed and stored on a Macintosh cornputer (SpiLe 2.2;
140 1 plus, Cambridge Electronics Design).
Peripheral nerve stimulation for spinal cord unit recordings
Units antidromically or orthodromically activated by PAG stimulation were tested
for peripheral nerve inputs. The PLN was stimulateci with a train of 2-3 0.3 msec wide
pulses at 333 Hz and an intensity of 5 tirnes the threshold current for producing a cord
dorsum potential or a maxiniun of 3 mA The other peripheral nerves were stimulateci with
single 0.2 msec wide pulses at an intensity of 5 times the threshold current for producing a
cord dorsum potential Units responsive to PAG and peripheral nerve inputs were
subjected to a collision test
Page 48
Criteria used to confirm antidromic activation
(1) Unit discharge at constant latency h m PAG stimulus.
(2) Stimulus-locked following of a bief train of stimuli in the PAG at a rate exceeding 333
Hz (usuaIly lm Hz).
(3) Collision of orthodromie and antidromic action potentials during a critical intemal,
Histology
Recording and stimulation sites in the PAG were marked by passing 5û-100 pi of
cunent through the elecwde for 2-45 mia In the PAG field potenbal experiments. the
animals were perfused transcardially with a solution of NaNe, followed by 4%
paraformaldehyde in phosphate b s e r (0.1 M, pH 7.4). The fixeci brainstem was removed
and post-hxed in 4% parafonnaidehyde and transferred into 20% sucrose for
cyroprotectioa F i micrometer sections were cut with a cryotome and stained with
thionin. The location of the electrolytic lesions were determinecl according to a cat brain
atlas (Snider and Niemer, 196 1).
Spinal cord recording locations were marked by cuthg off and leaving the tip of
the glas deetrode in the spinal cord The segment of the spinal cord containing the
electrode tip was removed and placed in 10% fornalin The cord was freeze sectioned with
a cryotome. F i micrometer sections were stained with 2% thionin and the micmelectrode
tracks were identifed. In these experiments the brainstem was pemised through the carotid
artery with 350 mL of 0.9% saline followed by 350 mL of 10% forniaiin. Brauistem
processing was as desCnbed above.
Page 49
34
Statistical Test
The Mann-Whitney Rank Sum test was used to compare median values of central
conduction velocity determined in the two types of neurophysiological studies.
Page 50
RESULTS
Neuroanatomicai Experiments
PAG retrograde tracer injections
The objective of this part of our wodc was to use retmgradely transported tracers to
fïnd neurons in the spinal cord with direct projections to the PAG. Of the three tracers used
(Table 1). only WGA-HRP injections into the PAG d t e d in retmgradely labelled
neurons in the lumbosacral spinal cord at the h v a i time examineci, some of the survivd
t h e s were too long. Figure 2 depicts the achial site of WGA-HRP injection and degree of
spread for the tracer for cases W93 and #8 10, with survival times of 4 and 3 days,
respectively. The main injection site for case #8lO was AI%, while W93 injection was at a
more rostral position of A2.5-A3.0.
Injection at more caudal regions of the PAG resulted in more retrograde1y labelled
cells in the spinal cord (#810). even though twice the volume of WGA-HRP was used in
case #793 (Table 1 and Figure 2). Comparing the two PAG injection experiments, there
were more spinal neurons projecting to caudal PAG since the total number of retrogradel y
labelled spinal neurons was 127 with an injection at AW, versus 75 with an injection at
A2.5-A3.0 (Table 2). Labelhg for #793 was M t e d to segments S2 and S3 with no
retrograde labelling in more rostral regions of the cord. For case #8 10, remgrade labelhg
occurred in segments L7 to S3 (Figure 3A and 3C). By combining the data nom the two
cases, 50% of the labelling occurred in S2, while 23% occumd in SI. 17% in S3. and
104b in L7. The total number of retrogradely labelled neurons in S2 was estimated to range
from 48 to 293.
Page 51
36 Labelling in the spinal cord was prwiominantly contralaterat In both cases there
w a e cluste15 of retrograde1y labelled nemm at the sacral parasympathetic nucleus and the
central canal at the S2 level (Figure 3A and X). Labelled œlls were found in the lateral
parts of lamina V and W in the S2 and S3 segments. This pattern of labelhg was more
apparent in the S2 segments. In L7 and S1 segments, l e s labelhg was seen in the sacral
parasympathetic nucleus but in S1 there was a cluster of PAG projecting neurons around
the central canal and in lamina W extending ventrolaterally. There were a few labelled
neurons along the lateral collateral pathway of S 1. There was some labelling in the ventrai
hom in the S2 and S3 segments, but especially in S3 (#793) along the lateral band of the
sacral parasympathetic nucleus neuroas.
Figure 4 contains camera lucida drawings of retrogradely labelled neurons of
various shapes and sizes, seen at different locations and in different orientations. One type
of neuron displayed a M o r m (or bipolar) soma with two main dendrites (or one dendrite
and an axon) onented in opposite directions. This type of neuron was most often observed
in lamina X Another type of neuron that was retrogradely labelled had a triangular soma
with two apical and two basal fibers extending from the soma (one of which is an axon).
There were also multipolar, round or polygonal somata with dendrites radiating in all
directions. The diameter of the somas for these labelled neurons ranged from 20-80 p.
The number of dendrites extending h m these retrograde cells varied h m 0-4. The larger
neurons, with multiple dendntic extensions, were most often seen around the central canaL
These dendrites tended to extend horizontally towards the sacral parasympathetic nucleus or
across the d o d commissure. The dendrites of neurons found in the lateral coilateral
pathway or lateral band tended to extend in the direction of the sacral parasympathetic
nucleus in a vertical manner.
Page 52
Primary Ptrerent terminai distribution in spinal cord.
HRP injected into the bladder wall labelleci afferent t e m h i s at the dorsal mot
entry, lamina 1, Lissauer's tract, along the laterai collateral pathway, and the lateral band in
the S2 segment (Figure 3B). There was no labelling along the medial collaferal pathway
leading to the dorsal commissine ( i a m i ~ X). Only ipsilateral & h g of the bladder
afferent terminals was observed Bladder afferent terminal distribution overlapped with
regions containhg neurons retrogradely labelled h m the PAG dong the lateral coilaterai
pathway and the sacral parasympathetic nucleus
c-Fos induction by bladder distension.
Having attempted to localize sacral spinal neurons projecting to the PAG. an
activity-dependent marker, c-fos protein, was used to detemine whether these neurons
might be related to bladder function. c-Fos. an immediate early gene, may be induced in
neurons by various chemical and mechanical stimuli (Weinberg, 1985; Hunt et al, 1987).
Bladder distension can activate the expression of this gene (Birder and deGroat, 1992).
In the absence of bladder distension (n=l), c-€os was expressed in the spinal cord
in L7-S3 @5gure 5) with liale reaction product discemible in S2 and S3. C-Fos protein
immunoreactivity was detected in the dorsal commissm (lamina X) and dorsolateral s a d
parasympathetic nucleus. Bladder distension (n=5) induced c-fos expression on both sides
of the spinal cord (Figure 5'). Labelled neurons were located along the lateral collateral
pathway, near the dorsal mot entry zone and along the lateral edge of the dorsal hom, and
in lamina V, W and X (Figure 5). The c-fos-positive cells fouad in L7, S 1 and S3 were
more concentrateci around the cenaal canal ( i a m i ~ X). In the experiments in which the
bladder was repeatedly distende4 the majority of the c-fos immunoreactive neurons was
found in S2 (Table 3).
Page 53
Figure 2: Retrograde Tracer Injection Sites in the PAG. Schematic drawing of the site of
5% WGA-HRP injection and degree of spread for the tracer for cases #8 10 (A) and #793
(B). IIlustrations are actual tracings of the cat midbrain sections. Out of the 3 AP levels
injecteci with the tracer? the main injection site for case #8 10 was AEQ while for #793
injection. it was at A2.ÉA3.0.
Page 54
Table 2: HRP Cell Counts. Total c d counts for remgradely labelled ce& found in 20 sections with the most labelhg for cases #8 10 and W93.
A
case #810 # Ce&
L7 S1 S2 S3 total#
20 47 48 12 127
15.70% 37.00% 37.80% 9.40% 100%
case #793
L7
O
O
# CelIs
S1 S2 S3 total #
O 52 23 75
O 69.30% 30.70% 100%
Page 55
Figure 3: Distribution of Spinal Cord Neurons Projecting to the PAG and Bladder AfTerent
Temllnals in the Sacral Cord Retrogradely labelled celis in tbe lumbosacral spinal cord
after injection of 5% WGA-HRP into the PAG in cases #8 10 (A) and #793 (C). WGA-
HRP injected into the bladder waiI ipsilaterally labelled afferent terrninals at the dorsal mot
entry, lamina 1, Lissauer's tract, dong the laterd collateral pathway in the S2 segment (B).
Page 56
Figure 4: Iliustrations of Spino-PAG Tract Cells. Camera lucida drawings of S2 spinal
cord neurons retrogradely labelled following injection of 5% WGA-HRP into the PAG.
Each neuron is numbered and ttiese numbers correspond to the numberings in the
representative spinal wrd section. The bar lines in the section indicates the position, and
orientation of the axond and deadritic extensions for each neuron.
Page 59
44 Figure 5: Bladder Distension Increased c-Fos Immunoreactivity in S2.
Fos-immunoreactive cells in representative sections h m L7 (AT B. C), SI @, ET F), S2
(G, H). and S3 (I, J) in experiment #785, with repeated bladder distension over 2 hrs (A,
B. D, E. G, I) and a conttol experiment in which the blad&r was drained for 2 hrs (C, F.
H, J). Panels B and E are rnagnified photos of the boxed areas in paneis A and D. Arrow in
panel E indicates c-fos in DC; c-fos cluster. are indicated in panel G by: srnail arrow. LDH
(also referred to as lateral collateral pathway); medium size amwow, SPN; large amw, DC;
smail arrowhead, MDH; large arrowhead, LVH (also referred to as lateral lamina WI) Sale
bars = 500 p.
Page 60
Table 3: Sum of c-Fos Immunoreactive CeIls in Each Spinal Cord Segment
Blad&r distension for 2 hrs increased c-fos immunoreactivity in S2-as indicated by the sum
of c-fos positive cells (n=5) in each of the spinal cord segments; control (n=l).
Spinal Cord Segmental Total # c-fos positive cells Total # c-fos positive celis
Levels Control B Iadder Distension
L7 1545 149
S 1 1742 2520
S 2 1332 12115
S 3 905 41 10
Page 61
46 The -test induction of the fos protein by bladder distension also occurred in S2, with a
srnaller increase occurring in S3 and a decrease in L7 and S 1 (Figure 6). Bladder distension
induced c-fos expression in ail five zones of S2 but the greatest induction occurred in the
DC and the sacral parasympathetic nucleus (Table 4 and Figure 7). In L7. bladder
distension suppressed the expression of the c-fos gene in ai l five zones. A decrease in Fos-
immmoreactivity o c c d in 4 zones in the S 1 segment. An induction of c-fos was
observed in the medial dorsal horn of S l with bladder distension. In S3, c-fos was induced
in the medial dorsal hom. dorsal commissure and sacral parasympathetic nucleus but
suppressed in the lateral d o r d horn and lateral ventral hom (Figure 7).
Figures 3 4 B. and C. and Figure 543, show that the distribution of c e h
wtrograddy labelied f?om the PAG overlaps with the Fos-immunoreactive cells in S2.
These spinal neurons are also located within the vicinity of the HRP 1abeUed afferent
terminais. W e were unable to obtain more definitive evidence that retrogradely Iabelled ceh
were related to bladder fiinctions (Le. were Fos-immunoreactive in bladder distension
experiments) due to technical ciifficul tics
Page 62
Spinal Cord Segments
Figure 6: Distribution of c-Fos Immunoreactive Cells in the Lumbosacral Cord The bar
graph illustrates the degree of c-fos protein induction in different spinal cord segments. The
greatest induction of c-fos by bladder distension occurred in S2, with a smder increase in
S3, and a decrease in L7 and S 1, The number of c-fos immunoreactive neurons in each
segment was expressed as a percent of the total of c-fos neurons in L7-S3. and this
percentage was averaged for 5 experiments. Bars indicate the standard deviation
Page 63
Table 4: Sum of c-Fos Immunoreactive Ceils in Different Zones of Each Spinal Cord
Segment The major increase in c-fos imrnunoreactivity. following bladder distension for 2
hrs, occurred in the zones DC and SPN of S2. as iadicated by the sum of c-fos positive
cek (n=5) in each segmental zones.
Spinal Experimental MDH
Levels Group
Conml
bladder
distension
Control
bladder
distension
Control
bladder
distension
Control
bladder
distension
LDH
529
32
467
479
38 1
2043
465
327
SPN
447
43
508
610
478
381 1
237
1660
LVH
8
O
35
1
66
429
4
217
Page 64
Spinal Cord Regions in each Segments
Figure 7: The Dorsal Commissure in S2 is the Major Site of c-Fos Indi
The bar graph depicts the area of greatest c-fos induction in each spinal cord segment
Bladder distension induced c-fos expression in all five zones of S2 but the greatest
induction occurred in the DC and the SPN, The number of c-fos neurons in a zone was
expressed as a percent of the sum of c-fos neurons in the 5 zones, and then averaged
(n=5). The control groups are L7-CTRL, SI-CTRL, S2-CTRL, S3-CIRL and the
experimental groups are L7, SI, S2, S3. MDH, medial dorsal hom; LD& lateral dorsal
horn; DC, dorsai commissure: SPN, sacml parasympathetic nucleus; L W , lateral ventrai
hom. Bars represent the standard deviation for the bladder distension group.
Page 65
Neuroph ysiological Experïrnent
PAG extracellular field potentials
Multiphasic field potentials could be elicited in the PAG by stimulation of the PLN
in 13 of 20 experiments. Data from a representative experiment are shown in Figure 8. The
fastest component of the field potential illustrated in Figure 8B was not consistently present
in other field potentials elicited by PLN stimulation. Central Iatencies for the field potentials
(component 2) recordeci in the PAG were measmd as the difference between PAG and the
cord dorsum potentials evoked by different peripheral nerves. The median and range of the
centrai latencies of the fint 4 components of the field potential figure 9) for each
peripheral newe stimulated are presented in Table SA The median latency for contralateral
PLN stimulation was similar to the ipsiiateral PLN stimulation (T'able 6A), but faster than
SPUD and SFP. The median central conduction velocity caldated in the field potential
studies was detemllned to be not significantly dinerent fiom the median conduction
velocity calculated in the spinal unit studies using the MaunWhitney Rank Sum Test
Assuming no signiricant contribution h m synaptic delays, all median conduction
velocities Lie in the range of A6 fibers. Although the maxUnum velocities correspond to
those of large diameter myelinated (A) fibers. We did not &termine whether the different
components were due to pathways with different conduction velocities or due to differently
routed pathways.
The PLNevoked field potentials diminished perceptibly with distance in the
dorsoventral 8B). anteroposterior (Figure 10A) and mediolateral (Figure 10B)
dimensions. The field potential recordings for SPUD and SFP display similar
characteristics to those for PLN stimulation. They were multiphasic (Figure 9) and the s k
Page 66
Figure 8: PAG Field Potentials Evoked by PLN Stimulation. An example experiment in
which PLN stimdation-evoked potentials recorded (B) d o m a track in PAG at P 1 .O (A).
Each trace averages responses to 16 consecutive single stimuli. Arrows indicate diflferent
components of the multiphasic response. The maximum for the second component was
found in the dorsal PAG (as indicated by 'rnax'. in panel A). CAq, cerebral aqueduc~ SC,
superior co~cdus; FL, fasciculus longitudinalis medialis; LDT, laterodod tegmental
nucleus; W. nucleus raphe magnus.
Page 67
Figure 9: Multiple Components in PAG Field Potential Recordings Representative PAG
field potential recordings foIlowing PLN, SPUD, and SFP stimulations demonstrating
rnultiphasic repenses. Arrows indicate different components present. The arrows are
numbered and these numbers correspond to the component numbers in table 4 (median CV
estimations).
Page 68
Table SA: Median Latencies for Different Components of tk MuItiphaSc Eeld Potential
Respooses Penpherd nerve stimulation produces a multiphasc respome (differeflt
components with ciiffirent latencies) in the PAG. Latencies for the different components
correspond to the numbered m w s in figure 9.
L-PL median 7.8 msec
min. 2.3 msec
max. 9.5 msec
SPUD median 1 1 .O msec
min- 7.6 msec
max. 15.0 msec
SFP median 12.0 msec
min. 2.9 rnsec
max- 23.9 msec
Cpt 2
12.6 msec
7.4 msec
20.9 msec
19.5 msec
10.3 msec
33.3 msec
18.3 msec
10 msec
28.2 msec
Cpt 3
18.4 msec
9-6 msec
36.6 msec
22.9 msec
12.9 msec
42.9 msec
25.6 msec
17.3 msec
47.0 msec
Cpt 4
58.6 msec
42.1 msec
7 1 .O msec
80.8 msec
60.0 msec
123.9 msec
54.1 msec
33.1 msec
74.0 msec
Page 69
Table SB: Median Conduction Veiocities for Different Compooents of the Muitiphasic field
Potentiai Responses. Penphed nerve stimulation produces a rnultiphasic response
(different components with different CVs) in the PAG. C V s for the different components
correspond to the numbered arrows in figure 9.
Cpt 1 Cpt 2 Cpt 3 Cpt 4
L-PL median 45.4 dsec 33.7 m/sec 24.1 misec 8.4 dsec
min. 37.3 mlsec 16.9 m/sec 9.7 m/sec 8.4 m/sec
max. 153.9 m k c 47.8 d s e c 36.9 d s e c 5.0 d s e ç
SPUD median 32.8 mkec 20.5 &sec 20.4 dsec 5.1 misec
min. 23.6 dsec 10.6 m/sec 8.3 dsec 2.9 m/sec
max. 46.6 m k c 34.4 m/sec 27.4 dsec 5.9 d s e c
SFP median 36.9 rn/sec 22.7 mfsec 17.9 dsec 8.3 dsec
min. 14.8 misec 12.6 &sec 7.5 dsec 4.8 m/sec
max. 122.1 m/sec 35.4 rn/sec 20.5 dsec 10.7 d s e c
Page 70
Table 6A: Median Latacies, for the 2nd component of the field potentid response in the
PAG for ipsilateral and contralateral PLN stimulations.
Median Latency Range (msec)
(msec)
contralateral PLN 10.0
ipsilateral PLN 9.0
SPUD 19.5
SFP 18.3
Page 71
Table 6B: Median Conduction Velocities, estimated for the 2nd component of the k l d
potential response in the PAG for ipsilateral and contratateral PLN stimulations.
Nerves Median CV (m/sec) Range (misec)
contralateral PLN 38.8
ipsilateral PLN 35.1
Page 72
Figure 10: Size of PLN-EvoW Field Potentials Varies with PAG Recording Site.
Representative traces demomtrating that the field potential decreases as the PAG recording
electroàe is moved in the anterior-posterior (A) and medio-lateral (B) direction.
Page 73
58 of the field potentials incffased as the recording electrode appmhed a population of
n e m m in the PAG responsîve to SPUD (Figure 1 1A) and SFP 11B) stimulation.
At the location in which the maximum &Id potential was evoked following the
stimulation of the L-PLN, SFP and SPUD stimulation did not evoke a maximum field
potential ( Figure 12A) (les than 25% of maximum). In this figure, at sites at which
maximum field potentials were recoakd for SPUD and SFP, L-PLN stimulation elicited a
submaxirnal response (50% of maximum) (Figure 12B).
In individual experiments, the PAG was searched over some AP distance. An
example of the results h m one experiment in which w k s were made at 3 AP levels are
illusaated in Figure 13. A maximum response was &tected in the domlateral region of the
PAG (APO) ipsilateral to the PLN stimulation site. Smaller responses were detected on the
contralateral side of the PAG.
Figure 14 illustrates the distribution of maximum PAG field potentials found in 13
experiments following PLN. SPUD, and SFP stimulation. Most of the maximum field
potentials elicited by PLN stimulation, were found at more caudal levels in the PAG
(-P1.0). Although most of these activated neurons are concentrated in the dorsal region.
there were also bladder responsive neurons in the ventrolateral region of the PAG.
Stimulation of SPUD and SFP nerves elicited field potentials within the vïcinity of
those evoked by PLN stimulation in the PAG but the maximal did not occur at the sites of
PLN maxima (Figures 11 and 12). The distribution of SFP and SPUD sensitive neurons in
the PAG ranged from P2.0 to A2.5 for SPUD and P2.O to A3 J for the SFP (Figure 14).
There is a cluster of points for SPUD sensitive neurons in the P0.5 - P1.0 level at an
intemediate region of the PAG. The distribution of newons responsive to each nerve
stimulation are not the same (Figure 14). This is compatible with a 'private' pathway for
bladder input.
Page 74
A SPUD B SFP
Figure 1 1: Peripheral Nerve Stirnimulation Evoke SimiIar Field Potentiai Respoases.
Representative field potential recordings illustrating characteristics for SPUD- and SFP-
evoked responses. They are multiphasic and the response increased as the recording
electrode approached a population of neurons in the PAG respollsive to SPUD (A) and
SFP (B) stimulation.
Page 75
L-PL
SFP 1
SPUD 1:
L-PL SFP SPUD
Figure 12: Different Areas of the PAG Are Responsive to Different Peripheral Nerve
Inputs- Representative traces of field potential recordings in caudal PAG Uustrating a site
which elicited a maxirnum for PLN stimulation not responding to SPUD or SFP
stimulation (A). At sites which maximum SPUD- and Sm-evoked field potentials were
de- L-PLN stimulation elicited a submaximal response (B).
Page 76
Figure 13: Pattern of Field Potential Responses to R-PW Stimulation. Schematic drawing
of the distribution and tracks of R-PLN-evoked field potentials in the PAG (A1.0-APO)
greater than 50% of the maximal recording in experirnent 960730. Both sides of the PAG
were searcheci in 3 AP levels. The maximum field potential (large filleci chle) in the
dorsdateral PAG, is ipsilateral to the stimulation site. Responses 7699% (medium) and
5 1-758 (small) of the maxima surrounds the maxima and diminishes with distance-
Page 77
Figure 14: Distribution of Maximum Field Potential Recordings in the PAG.
The distribution of field potential maximas differ for each nerve stimulation. The different
symbols distinguishes the different nerves while their location. on the right or left side of
the PAG, indicates contralateral or ipsilateral stimulation, respectively. Dots. PLN;
squares. SPUD; opencircle, Sm.
Page 78
63 The question of ipsilated versus contraiateral transmission of bladdes impulses
was adâressed in this snidy. There may be a slight ipsilateral dominance in bladder
trammision (Figure 15A). Athough inputs h m SPUD appear to dernonstrate a
contralateral Pansmission (Figure 15B) while SFP inputs seem to demonstrate an ipsilateral
dominance (Figure 1 SC), there were only 3 cases in which both sides of the PAG searched
for SPUD and SFP. Therefore, these resuits are inconclusive.
Spinal Cord Units
The placement of the stimulating elecaode in PAG was detemineci by: (1) the size
of the bladder contraction response to stimulation of the PAG; andor (2) the size of the
field potential recordings in the PAG foliowing stimulation of the PLNs.
PAG stimulation in the region where bladder responses were produced also often
produced an increase in blood pressure of 25 to 75 rnrnHg. However, there were
experiments in which PAG stimulation produced a strong blood pressure response but not
bladder. The site of PAG stimulation producing a pressor response was usually very
distinctive as evidenœd by a gradualiy increasing response as the electrode was moved
d o m a track (Figure 16).
PAG stimulation also produced a rise in intravesical pressure which appeared to be
dependent on bladder volume. although this point was not systematically studied. The
biadder response ranged fkom an increase of 1.0 to 18 cm &O in 15 experiments. Figure
16 illustrates a representative experiment in which PAG stimulation produœd bladder
pressure responses which increased from 3 cm %O at 3.0 mm below the sUTfaçe of the
midbrain to 12 cm %O at 6.0 mm below the surface. It is not known if stimulation of more
v e n d areas would have elicited larger bladder contractions, since ttiese regions were not
searcheci. Bladder activity elicited by PAG stimulation was always small compared to those
elicited by stimulation of the PLNs at 10 Hz (Figure 16).
Page 79
Figure 15: IpsÜateral versus ContraIateral Transmission. Histograms illustrating the
ipsilateral and contralateral transmissions of bladder information. PLN (n=13) (A) and SFP
(n=3) (C) information are conveyed in an ipsilateral manner while SPUD (n=3) (B) is
contralateral.
Page 80
-3 -4 -5 -6 rev. -6
Figure 16: Effects of PAG Stimulation. PAG stimulation (A) at 50 Hz produced a large
bladder (B) and blood pressure (C) response, unWce L-PLN stimulation at 10 Hz which
produced a larger bladder but not blood pressure response. Bladder response to PAG
stimulation increased as the recording elecaode was moved in the dorsoventral direction
Page 81
66 In two experiments, the PAG stimulating sites were detennined by
searching for the W e s t field potential responses in the PAG that codd be produœd with
PLN stimulation This response was con€iied by stimulating at this PAG site and
examining the size of the bladder response. It was found that at PAG sites which showed a
field potential in response to PLN stimulation, PAG stimulation produced blad&r
con tractions.
Spinal neurons antidromically activated from the PAG
In 13 cats, 17 uni& were found in the lumbosacrai spinal cord which were
considered to be antidromicaIly activated h m the PAG (TabIe 7) based on: (1) consistent
latency of response; (2) reponse foilowed a high fquency of PAG stimulation (333-
1000Hz). The distribution of these uni& is iUustrated in figure 17. They were located
conttalateral to the PAG stimulation site. Some appeared to be located in the white matter
(5/17) based on micrometer readings. There is a possibiüty that these five discharges are
due to axonal activation and not soma1 discharges since this was not tested at the time of the
recording. Antidromically activated neurons were primarily located in the dorsal hom with
the exception of two neurons in the v e n d hom and 3 in the intemediate area. Of the 17
neurow antidromically activated from the PAG in this study, 4 (23.5%) received inputs
horn the peripheral nerves we had prepared for stimulation (Figure 17). They were mainly
located in S2 and responded to SPUD only (lamina VIi), SPUD/SFP/CCF m i n a V) and
SFP/CCF (iamina V) (Figure 17). ûnly one (1.9%) neuron. found in lamina W of rosaal
S3. received an input from PLN (Figure 18). It is not hown if t h - neuron would have
been responsive to other periphed nerve stimulations since it was lost before they could be
tested.
In 3 of 4 cases confirmation that the peripheral input reached the it~ltidrornically-
activated neuron was obtained by using a collision test
Page 82
Table 7: Counts of PAG Antidromically and Orthodromidy Activated Neurons in the
Sacral Cord The majority of the newons aictivated (antidrornicaily or orthodromically) by
the PAG did not respond to the test nerve stimulations.
# antidrornic % of totai # orthodromic % of total
cells #antidromic cells # orthodromie
ce1 1s cells
total # 17
PAG only 13
PAG & PLN 1
al1 nerves 3
no PLN
al1 nerves
(incLPLN)
Page 83
Figure 17a: Disaibution of Lumbosacral Neurons Activated by PAG Stimulation PAG
stimulation at APO represented by a dot (A) antidromically (B) and synaptically (C)
activated neurons in S2 and S3 @) contralateral to the PAG stimulation sites. Stars indicate
PAG activated neuroas.
Page 84
Figure 17b: Distribution of Lumbosanal Nemm Activated by PAG Stimulation PAG
stimulation sites (dots) at PL0 (A) antidromicaily activated neurons in S2 (B & C) and S3
(D) contralateral to the PAG stimulation sites. PAG stimulation also synaptically activated
neurons in L743 @-H). Asterisks indicate PAG activated neurons; stafi represent PAG
activated nemm receiving peripheral input.
Page 85
Figure 18: Sacral Spinal Neuron Which Projects to the PAG Conveys PLN Input. Neuron
in S3 (star in F), was antidrornically activated by PAG stimulation at P1.0 (dot in E) (333
Hz), and it responded with a latency of 41 msec (C). Stimulation of the PAG (50 Hz)
produced bladder (A) and blood (B) pressure responses. L-PLN stimulation activated this
neuron at a latency of 12 msec @). Amws indicate start of stimulation for the PLN or
PAG. Calibration bar in @) refers to panels C and D.
Page 86
Figure 19 Uustrates a representative experiment An S2 dorsal hom projecting neuron was
identified by antidromic activation at a constant latency of 5 msec h m the con-
caudal PAG (P1.0) (Figure 19F). The neuron had a latency of 5.5 msec (Figure 19D) and
followed a train of 3 stimuli in the PAG at 100 Ek @gure 19C). This neuron also
responded to SFP (Figure 19D), SPUD and CCF stimulation. Figure 19E demonstrates
collision of SFP stimulation evoked and antidromic PAG stimulation evoked discharges at
an interstimulus delay of 8 msec. Collision of SPUP and CCF- stimulation evoked
discharges with the PAG-stimulation evolrwi discharge was also confirmed at an
interstimulus interval of 8 msec. These d t s indicate that the spinal neuron projecting
directly to the PAG rewived afferent input fmm all uiree peripheral nerves. However, it did
not have an input fiom PLN.
The median latency for neurons that were antidromically activated by PAG
stimulation was 11.5 msec (n=17) with a range of 5-60 rnsec. Using a conduction distance
of 361 mm ( h m the cord dorsum electrode to the PAG electrode), the median conduction
velocity was estimateci to be 3 1.4 dsec. Only one conduction distance was measured in
these spinal unit searches but the average conduction distance for the field potential studies
was 355 mm (n=l 1). Using the latency from the projecting neuron which also responded
to L-PLN stimulation (iarnina VII), the conduction velocity was caîcuiated to be 8.8 &sec.
The conduction velocity for the neuron responding to only SPUD (lamina VII) was 12.4
m/sec. The antidmrnicaJly activated neurons receiving convergent peripheral input from
SPUDISFPKCF (lamina V) and SFEVCCF (lamllia VI) had conduction velocities of 72.2
m/sec and 40.1 mkc, respectively.
PAG stimulation also activated spinal neurons synaptically. These neurons were
found more frequently (1W177) than those which were antidromically activated (171177)
(Table 7). The characteristics of synaptic activation were: (a) variable latency for single unit
nriag, (b) burst firing, (c) failure to foilow a train of 4 stimuli at 333-1000 Hz
Page 87
Figure 19: Spinal Neuron, Receiving SFP Input, Projects to PAG. A neuron in S2 (G) was
antidromically activated by PAG stimulation at P1.0 (F) with a latency of 5 msec @).
Stimulation of the PAG produced bladder (A) and blood (B) pressure responses. The
neuron also responded to SFP nerve stimulation with a latency of 3 msec. The unit
followed PAG stimulation at 100 Hz (C) and responded to both SFP and PAG stimulation
at an interstimulus interval of 10 mec @) but not 8 msec (E). Small arrows indicate SFP
stimulation; medium size arrows indicaie PAG stimulation. Large m w indicates where the
PAG stimulus-evoked discharge was expected to appear. Calibration bars in (E) refer to
panels C, D, and E.
Page 88
73
An example of such a neuron is presented in Figure 20. In addition to the PAG. this neuron
was activated by stimulation of GPL but not &PL. SPUD and SFP inputs were not
exarnined Only 61160 (3.8%) of these cells received an input from PL, SFP and SPUD
(Table 7). Spinai neurons that were synaptically activated from the PAG were found mostly
in S2. distributeci from lamina V to the ventral horn and ventral white matter (Figure 17),
with a median latency of 14.0 msec (n=120) and a range of 6-34 msec. The conduction
velocity was estimated to be 25.8 mlsec.
In experiments in which orthodromically or aatidromically activateà neurons did not
respond to PLN inputs, there were indications that the PLNs were functioning. Indication
of functional PLNs include PLN responses in other backs. bladder contraction following
PLN stimulation, and the presence of PLN cord dorsum potentials or reflex responses.
Spinal neurons which were antidromically activated h m the PAG but did not respond to
P M . SPUD, SFP or CCF, could be receiving input from other pelvic viscera (bowel. sex
organs) which were not studied.
Convergence of merent input
During the search for spinal neurons antidromidy activated nom the PAG, 27
neurons were encountered that were activated by FLN stimulation but not by PAG
stimulation. M y one of these, found in the ventral hom of S2, was a f b t order
intemeuron receiving PLN input There were twice as many higher order (37) intemewons
as fiirst or&r (16) intemeurons. The criteria used to distïnguish between fmt and higher
order intemeurons was the latency of the response recorded in the spinal cord. If the
latency for the response was shorter than 3 msec, then the neuron was considered to be a
6rst order intemeuron. Some first order intememm for SPUDlSFPfCCF input also acted
a s higher order intemeurons for PLN input (i.e. a neuron responds faster for one input
Page 89
74 compared to another). The distribution of the higher order and nrst order intemewons were
similar (Qwe 2 1).
Effets of PAG stimulation on spinal newon activity
Although it was not searched for coosistently, interactions between PLN and PAG
stimulations were observed in 5 spinal nemns. These interactions were observed to be
inhibitory (Kgure 2 2 4 and facilitatory (Figure 22.B). Co-stimulation of the PAG and G
PLN was required to elicit a consistent response h m a sacral spinal neuron (Figure 22A
and 22B). A late component of a response elicited by R-PLN stimulation was depresseci
with CO-stimulation of the PAG (Figure 22C and 22D).
Page 90
Figure 20: PAG Orthodrornically Activateci Spinal Neuron
Neuron in S2 (dot in D) which was synaptidy activated by ipsilateral PAG stimulation
(iarge dot in C). Responses of the sacral spinal neuron to LPAG (A) and GPL (B)
stimulation are illustrated.
Page 91
Figure 21: Distribution of Spinal Fust and Higher ûrder htemeurons in Sacral Spinal
Cord Panels show distribution in sacrai segment SI-S3 of fmt (x) and higher order (small
dots) interneurons responding to stimulation of left SPUD, SFP, andor CCF. Ceils which
are fmt order intemeuron for SPUD, SFP or CCF and receive PLN input are indicated by
squares. Those which receive higher order inputs including P M , are indicated by open
circles. One ceil was a finit order intemeuron for PLN and responded to ody PLN
stimulation (large dot).
Page 92
Figure 22: PAG and PLN Interactions. PLN stimulation did not elicit a response (A) in an
S2 neuron (dot in E) but CO-stimulation of PL and PAG stimulation at APû (star in F) did
elicit a response (large arrow in B). In the same experiment, another S2 neuron (dot in G)
responded to PLN stimulation (C), but CO-stimulation with PAG (star) d t e d in inhibition
of a component of PLN response (D). Large m w Ui D indicates where the unit should
have appeared. Smaller m w s indicate the start of the stimulation.
Page 93
DISCUSSION
The act of mictuntion is compriseci of a storage ( e g ) and an expulsion phase.
These phases require opposite functions nom the bladder and the u r e k The bladder acts
as a reservoir to store urine and also as a pump to expel it during voiciing. During urine
expulsion, the urethra opens, dilates and becornes cornpliant to allow urine flow. It is
closed and contracted during storage. The neural components controlling the complex
events of rnicnirition involve highly complex central and peripheral, afferent (sensory) and
effereat (motor) autonomie pathways which are integrated and coordinated by cephalic
control centers, spinal cord nuclei, and peripheml gangiia
M o u s mo&k of bladder control postulated that bladder information was camied
via the spinal cord to Barrington's nucleus in the dorsolateral pontine tegrnentum. This was
recognized as the center for bladder conml as lesions of this area abolished micturition.
Recently, it has k e n proposed that the periqueductal gray (PAG) region of the brainstem
may be involved in michirition Our hypothesis is that the PAG plays a role in the
mictwition reflex pathway.
The aims in this research project were: (1) to l o c h the bladder responsive celis in
the PAG and spinal cord; (2) to characterize this ascending pathway.
There were two approaches which were taken to locaiize, iden-, and characterize
bladder sensitive neurons projecting from the sacml spinal cord to the PAG. One approach
involved nemanatomical tracing studies while the other entailed neurophysiologicai
studies.
Retrograde eracing techniques were applied to establish a direct projection from the
sacral cord to the PAG. This part of the neuroanatornical study was abandoned upon the
Page 94
79 publication of an article by Vmderhorst et al <19%) which provided retrograde and
anterograde evidence of direct pmjections fmm the sacral spinal cord to the PAG.
Anterograde tracïng and c-fos induction studies were used to reved the conveyance of
bladder-related information in this ascending pathway.
Electrophysiological techniques were used to test the hypothesis that there are
bladder sensitive cek in the PAG. The PAG was searched for field potentials evoked by
PLN stimulation. To M e r characteize the ascending projeçtion h m the sacral spinal
cord to the PAG, projecting neurons were identified by antidromic activation from the
PAG, and tested for PLN and other identified peripheral nerve inputs (SPUD. SFP, and
CCF) to detennine the specif~city of this pathway.
Neuroanatomical Experiments
The mode1 for the micturition reflex pathway is quickiy changing to include the
periaqueductai gray region Our interest in this pathway however. lies with the neurons
projecting from the spinal cord to the PAG. Neuroanatornical studies by Blok et al (1995)
have shown that thete are very few nemm in the spinal cord with direct projections to
Barrington's nucleus. Retrograde labelhg revealed direct spinal cord projections u, the
PAG (Vanderhorst et al, 1996). In th& shidy. injections into the intemediate area of the
PAG at PO.5, resdted in predominately condateral retrograde labelhg of neurons in
caudal S2. These resuits correspond with ou. retrograde fmdings. Of nine retrograde
experiments. only two using WGA-HRP resulted in labelhg in the sacral cord The other
WGA-HRP retrograde experiments used a longer sufvival time than 4 days and thus the
tracer could have been metaboliml prior to andysis.
Page 95
80 In the present study, contralateral retrogradely labelleci ceh were rnainly observed
in the SZS3 segments of the spinal cor& These d t s are also compatible with our
bladder tracing data
The objective of our retrograde study was not localkation of an area of the PAG
which receives projection fiom the d The objective was to establish a direct projection
h m the spinal cord to the PAG, therefore the tracer volume injected hto the PAG was not
small or discrete. Now that it has k e n established that there are spinal projections to the
PAG, it would be more usehl to identify the specific region of tbe PAG that is receiving
these projections by making smaller injections. A possible concem with our retrograde
study is the degree of tracer diffusion at the site of injection. It should be noted that
although the most caudal area of the PAG shown in figure 2 was at the level of Pl 5, the
tracer could have ciifhiseci into more caudal areas. These caudal areas may consist of
structures which have ken shown to regdate micturition. For example the tracer codd
have reached Barrington's nucleus, the locus coedus, or parabrachial nucleus. Therefore,
it is possible that the retrogradely labelied n e m m in the spinal cord are projecting to these
sites and not the PAG. But as mentioned earlier, studies have already shown a paucity of
projections h m the lumbosacral cord to the Barrington's nucleus of the cat (Blok et al,
1995). With regards to the locus coeruius, it has been shown by Aston-Jones et al (1986)
that previous reports of dorsal horn projections to the locus coenrlus (Cedarbaum and
Aghajanian, 1978) were inaccurate and were instead found to projet to the Iaterally
adjacent parabrachial nucleus or to the pontine central gray media1 to locus coerulus. There
are contlicting resdts for studies examining the d e of the locus coeruius in micnuition.
Whüe Sasa and Yoshimura (1994) and Yoshimura et al (199ûa,b) have provided evidence
for locus coenrlus involvement in micturition, Satoh et a1 (1978) have show that bilateral
destruction of this area did not result in urinary disorders (Satoh et al, 1978).
Page 96
81 Studies have shown spinai projections to the parabrachial nucleus (Blomqvist et ai,
1989). This nucleus is known to mceive information €rom the bladder and to project to
Barrington's nucleus (Valentino et ai, 1994). Furthemore, elecaical and chernical
stimulation of the parabrachial nucleus in the rat resulted in bladder contraction (Lumb and
Morrision, 1987). This evidence implies that the parabrachial nucleus may also have a d e
in the micturition pathway. However, its d e may be more reiated to viscerd nociception
since retrogradely labelled spinoparabrachial tract c e k expressed c-fos protein foliowing
formaiin irritation of the bladder @ing et ai, 1994). It shodd aiso be noted that although
Barrington's nucleus does receive projections from the parabrachial nucleus, it receives
more extensive projections h m the PAG (Wentino et al, 1994). niese findings indicate
that although the btainstern bladder control circuitry may consist of the ventrolateral PAG
and Bamington's nucleus, it may also involve other brainstem structures.
Vanderhorst et al (1996) made discrete tracer injections into various divisions of the
PAG, limiting the spread of the tracer into other regions of the brainstem which may be
involved in the micturition reflex pathway. Tbeir injection sites which did not include the
locus coeruius or the parabrachial nucleus, resdted in retrograde labelling in the spinal
cord.
Based on our findings, the spinal cord areas with neurons projecting to the PAG
consist of the lateral band, the sacral parasympathetic nucleus, and regions around the
central canal (including the dorsal commissure). These projections are bilateral but
predominately contralateral. In an antemgrade tracing study where HRP was administered
udaterally to the PLN, labeiied dorsal mot ganglion neurons were found only ipsilaterally
(Nadelhaft and Booth, 1984). Some of the bladder afferent inputs which are conveyed to
one side of the sacral spinal cord have access to a population of neurons projecting to both
sides of the PAG, but predominantly contralateral, Sacral preganglionic nemm are also
labelied ipdaterally h m the P W (Nadelhaft and Booth, 1984) but they have long axon
Page 97
82 collaterals projecting to the opposite side of the cord (Morgan et al, 1991). nie latter
kding implies that the crosshg of bladder information occiirs at the spinal cord level in the
efferent pathway. However, the question of communication between opposite sides of the
cord and brainstem at the afferent level =mains unclear.
In the present saidy. retrogradely labeUed neurons pmjecting to the PAG were
found in lamina X (dorsal commissure). They were bipolar neurons widi dendrites
extending across the dorsal commissure. This indicates that the crossing of the micairition
pathway could occur at the level of the sacrai cord before the information is conveyed to the
PAG. By contrast, NadeIhaft and Vera (1995) ushg pseudorabies virus tracing, found
cells with dendrites extending into the dorsal commksure, but not crossing to the other side
of the spinal cord. Nevertheles, there is evidence for afferent fibers crossing in the sacral
spinal cord. Although Ueyama et al (1984) found that tenninals of SPUD afferents could
be found ipsilaterauy in lamina 1 of the sacral segment and graciüus nucleus, bilateral
labehg of terminais could &O be visrtali;rf!d in lamina lII and IV and in the dorsal
commissural grey matter. Matsushita and Tanami (1983) have also reported afferents
cmssing the midline in dorsal and central commissures of the sacral cord in the cat
In our anterograde study, afhent fiber terminai labelhg occurred ipsilateral to the
injection site in the bladder wd. This corresponds to other antemgrade studies in which
HRP was applied to PLNs (Nadelhafl and Booth, 1984; deGroat et al, 1981) and in
retrograde studies (Downie et al, 1984). Studies using pseudorabies virus also showcd
hbehg in ipsilateral intemediolaterd area before it is observed in other areas of the corci,
such as the dorsal gray commissure (NadeIliaft and Vera, 1995). With time pseudorabies
Wus travelled to the brainstem and eventually labelled the contralateral intermediolateral
regions of the spinal cord.
Retmgradely labeiled celis were observed in the ventral hom, along the lateral edge
of caudal S2 and S3 spinal cord sections. Inis area in the ventral horn corresponds to the
Page 98
83 lateral band region of the sacral parasympathetic nucleus (de Groat, 1981 ; Morgan et al,
1991,1993). A similar clustering of neurons were c-fos immunoreactive in our bladder
distension study. Retrogradely IabeIled tells in this area were also observed at the S2 and
S3 level in the study by Vanderhorst et al (19%). They mentioned that these snall clusters
were fonned at distinct rostrocaudal intexvals of 220-300 um.
Although the ventdateral lamina W region does not appear to receive bladcier
afferent tenninaIS in our study, the neurons in this region were synaptically activated by
bladder distension (figure 5G). Furthermore, anatomical studies in which sacrai
preganglionic neurons were labelled by retrograde HRP traciag methods revealed that the
sacral parasympathetic nucleus CO& of three divisions: a dorsal band of neurons located
in lamina V and VI, a lateral band of nemm located in lamina W, and an interband region
which lies between the dorsal and lateral band regions (deGroat et ai. 198 1; Morgan et al.,
1979; Nadelhafi et ai., 1980,1986; Mawe et al, 1986). Electrophysiological and axonal
tracing studies demonstrateci that neurons in the dorsal band innemate the large intestine
while those in the lateral band innenate the bladder (deGmat and Ryall, 1969; Morgan et
al, 198 1). This implies that in our study, neurons in the ventdateral region of lamina W
which projects to the PAG may be involved in bladder-s-c activites. But in our
anterograde tracing study, bladder afferent tenninals were not labelled in this region
(figure2C). In contrast Nadehafi et al (1980) revealed antemgrade HRP labelled axons
extending ventrally into lateral lamina W. However, the absence of aEerent terminal
labelling in this region in our studies may be attributed to s m d tracer volume used to
achieve labelling in the more distal fields such as the lateral lamina W.
The interband region of the sacral parasympathetic nucleus consists of intemeurons
and spinal tract neurons projecting to at least L2 and Tl3 (deGroat and Ryall, 1969;
Morgan et al, 1981). These internewons are well-positioned to receive viscerd aî3erent
Page 99
84 cohterals (figole 23) and in tum may be involveci in intersegmental and supraspinal
The specifïcity of bladder aerent terminal labelhg in anterograde studies would
depend on where dong the PLN the tracer was applied since the PLN separates distally into
distinct branches which can be traced to different effector organs (Langley and Anderson,
1895,1896). Since the tracer was injected into the bladder wall in our study, the tenniaals
that were labelled could be attributed to bladder afferents terminating in S2, near the dorsal
mot entry. dong the lateral coilateral pathway towards the sacral parasympathetic nucleus
and more medially. nie distribution of the labelled bladder afZerent tenninals corresponds
with the location of spinal tract neufons in the second sacrai segment of the spinal cord that
were labelled remgradely foilowing the unilateral application of HRP to cut axons in the
dorsdateral fimiculus at the fmt lumbar level (deGroat et al, 1981).
The pattern of afferent terminal labelling in the present snidy corresponds closely
with other anterograde studies (deGroat et ai, 1981) with the exception that in our study,
the medial coiiaterai pathway extendhg towards the dorsal commissure was not ~ ~ ~ ~ ~ e d .
This was unexpected since it has been dernonstrated that the projections to the medial
collateral pathway and dorsal commissure are rehtively weak for colon afferents not
bladder (deGroat, 1986).
Our anterograde tracing study also did not resdt in labelling of preganglionic
neurons and their processes in contrast to other studies in which HRP was applied to the
cur PLN (Morgan et al, 1981; Nadelhaft and Booth, 1984). It is possible that a longer
survival time is required to achieve bladder afferent labeiiing in the medial collateral
pathway and lateral band 1~egions. There may be a difference in transport times for different
fibers. Another expianation for these ciifferences in results is that an insufficient volume of
the tracer was injected to £dl the ce11 bodies and dendrites of preganglionic neurons, or the
fibers projecting to the medial collateral pathway and lateral band regions. In the case of
Page 100
85 non-labeiled preganglionic neurons, it is also possible that the tracer was not picked up by
the efFerent fibers with tracer injection into the bladder wall.
Our studies show that the areas which receive bladder aEerent tenninals overhp
with the regions sending projections to the PAG. Therefore, bladder af fe~nts reaching the
sacral cord via the PLNs. tenninate near the dorsal root entry, dong the Iaieral coliateral
path and in the sacral parasympathetic nucleus and in the vicinity of neurons projecting to
the PAG.
c-Fos protein induction by bladder distension
Although the above hd ings provide M e r support for the involvement of the
PAG in the mictuntion refiex pathway in tenns of neuroanatomical connections, it does not
provide funct iod support for our hypothesis. Tracing studies do not provide information
about the specificity of the ascending pathway for bladder transmission.
The functional signifcance of the S e r m connectivity was studied using the
presence of c-fos protein as a macker for neuronal activity. The expression of the protein
product of this immediate eariy gene was inducible in urethane anestheshed rats by bladder
distension (Birder and deGroat, 1992). In the present study, we found that c-fos protein
expression was inducible in chlordose anesthesized cats by bladder distension. The S2
spinal cord segment contained the major@ of the n e m n s that were immunoreactive for c-
fos protein. The localization of c-fos protein in cells in S2 reflects the larger number of
urinary tract afferents projecting to this segment, as demonstrated by our antemgrade
tracing studies.
It should be noted that negative results with c-fos technique are mcult to interpret.
Increased c-fos expression implies that murons have received synaptic input. However, the
Page 101
86 failure to detect c-fos expression does not necessarily mean that neurons have not been
synaptically activated. It is possible that some n e m m do not express c-fos uuder any
conditions, or they may produce leveis of c-fos protein below the threshold for àetection
with immunocyochemical techniques. There are even certain types of synaptic activation
which rnay in- c-fos expression. whereas other types rnay nor For example, Birder
and deGroat (1992) reported the failure of sphincter motoneurons to express c-fos in
response to a constant infusion of saline into the bladder. Synaptic activation of the
sphincter motoneurons was demonstrated by extenial u r e W sphincter contraction during
voiding. In accordance with the fïndings of Birder and m a t , sphincter motoneurons. in
our c-fos induction study, did not display imrnunoreactivity for the c-fos protein. Other
reports have also indicated that c-fos levels are below threshold for detection in
motoneurons (Jenkins and Hunt, 1991).
Following bladder distension. c-fos protein immunoreactivity was detected in five
major areas of the S2-S3 spinal segments: medial dorsal horn, lateral dorsal hom, dorsal
commissure, sacral parasympathetic nucleus, and lateral ventral hom zones which (with the
exception of lateral ventral hom) correlates with findings in the rat (Birder and ffiroat,
1992). These areas reœive PUD and PLN afferents innemathg the urethra (Thor et al,
1989) and bladder (Nadelhaft and Booth, 1984), respectively. Birder and deGroat (1992)
reporteci h m their selective denervation experiments that pudendal nerve afferents
conveying somatic information activated ceh prirnarily in the dorsal CO-ure and
medial dorsal horn zones. wheceas PLN afferents carrying information from the viscera
activated cells in the dorsal commissure and sacral parasympathetic nucleus regions.
Therefote they suggested that the dorsal commissure region may be important for
viscerosomatic interaction since it exhibited the highest induction of c-fos expression by
PLN and SPUD stimulation. The medial dorsal hom and sacral parasympathetic nucleus
regions exhibited greater specifïcity and rnay be important for processing somatic and
Page 102
87 viscerd inputs, respectively. They also reported that the dorsal commissure region was
important for p m g nociceptive information. Therefore, in our study, bladder
distellsion r e s u l ~ g in an increased expression of c-fos in the dorsal commissure region of
the S2 and S3 segments could be attribut& to greater viscerosomatic interactions or painfid
stimulation of the bladder* Our hding of an increase in c-fos positive œlls in media1 dorsal
hom in the S1 and S2 segments indicates greater PUD proceshg by the neurons in this
area with bladder distension A greater viscerai proceshg occurred in the sacral
parasympathetic nucleus region of S2 with bladder distension, a s indicated by the increase
in neuronal activity (increase in c-Fos protein expression) in this area.
Although, there was an uicrease in c-fos expression in the medial dorsal horn,
lateral dorsal horn, sacral parasympathetic nucleus and dorsal commissure regions of S2
following bladder distension, the regions of &reatest induction was the dorsal commissure
followed by the sacral parasympathetic nudeus. This suggests that bladder distensions or
reflex bladder contractions preferentially activate ce& in the dorsal commissure and sacral
parasympathetic nucleus. These findings are in agreement with data h m Birder and
deGroat (1992).
Data from our studies demonstrate that the location of the c-fos positive neurons
overlaps the sacral cord region where WGA-HRP retrograde1y labelled neurons projecting
to the PAG were found This Unplies that these synaptically activated c-fos positive
neurons may be involved in an ascending projection systern to the PAG. This postulate is
supported by other preliminary reports (Birder et al, 1990) demonstrating that a percentage
of the c-fos positive ceh after bladder imitation were spinal tract neurons pmjecting to the
hypothalamus or to the region of the pontine micturition center and others were
pregangiionic neurons sending axons to the pelvic ganglia (deGroat et al, 1992). Ding et al
(1994) were able to induce c-fos-iike protein in neuronal ce11 bodies within the sacral
parasympathetic nucleus by chemicaiiy irritating the bladder* They found that more than
Page 103
88 5W6 of these fos immunoreactive neurons were retrogradely iabeUed with fluoregold
injected into the parabrachial nucleus. This indicates that some non-preganglionic nemns
within the sacral parasympathetic nucleus may trammit noxious information h m the
bladder to the parabrachial nucleus.
The neuroanatomical hdings, provide presunptive evidence that the connections
h n the bladder to the sacral cord and h m the sacral cord to the PAG rnay be conveying
bladder afferent information to the PAG. Neurophysiological studies were undertaken to
provide M e r support for the functional significance of these connections and to
characterize this ascending pathway in terms of the specificity of the information it
conveys.
NeurophysioIogicaI Experiments
PAG extracellular field potentials
We were able to evoke field potentials in the PAG by stimulating the PLN. This
response suggests that there may be neurons in the PAG reœiving bladder information.
The distribution of maximum field potentials (figure 14) indicates that the region which is
most responsive to PLN stimulation is around P0.5 - P1.O. This is consistent with
neuroanatornical data showing that the PAG at the P1.0 level is the major target of
Iumbosacral afZerents (Blomqvist and Craig, 1991; Vanderhorst et al, 1996; Yezierski,
1988; Blok et al, 1995).
Our neurophysiological resuits revealed maximum field potentials for PLN
stimulation in two regions of the PAG, the dorsolaterai and ventrolateral regions. Within
the PAG, spinal projections terminate in lateral PAG at PO5 (Vanderhorst et al 1996) in
cats. Liu (1983) found that tracer injections into the ventdateral PAG of the rat resdted in
the most neuronal labelling at the lumbosaaal region, as compared to dorsolateral PAG
Page 104
89 injection. Furthemore, Blomqvist and Craig (1991) found that injection înto the lumbar
cord resuited in terminal labehg in lateral PAG at P1.0 but labelling in two areas in caudal
PAG (P1.5-P2.0) in the cat: dorsolateral and ventrolateral PAG.
HEW trachg shidy in cats revealed projections to Barrington's nucleus fmm lateml
regions of the PAG (Blok et al, 1995). Cameron et al (1995), however, demonstrated
greater Phaseolus vulgaris-leucoagglutinin (PHA-L) labelling in Barrington's center of the
rat following tracer injection into dorsoiateral PAG than ventrolateral PAG.
Based on these controversial hdïngs, it is evident that the circuitry within the PAG
is very complex and unclear with respect to the micturition pathway. The different species
used in different laboratones may add to the complexity of the circuitry involved in the
mictwïtion pathway. Furthemore. not all of the projections have to be involved in the
mic turition pathway.
The neuroaatomical studies mentioned above do not provide fiinctional evidence
for the axonal projections h m the sacral cord to the PAG and from the different regions of
the PAG to Barrington's nucleus. Neurophysiological studies in rat have demonstrated an
optimum site in the dorsal PAG for recording short latency PLNevoked field potentials
(Noto et al. 1991). The same study also revealed an optimum stimulation site in the ventral
PAG to elicit PLN discharges. This suggests rhat there are two regions in the PAG
involved in transmitting bladder information. which corresponds with our field potential
data (figure 14). The role of the nemns located in the dorsolateral PAG io the micturition
pathway is unclear. But it is known that the ventrolateral PAG is important in supporthg
reflex rnictuition since preliminary studies (S. Matsuura, G.V. Allen, J. W. Downie
unpublished) show that cobalt blockade of synaptic transmission at sites in the ventrolateral
PAG interrupts micturition in the urethane anesthetized rat
In a study on spinal neurons projecting to unloiown supraspinal targets, ascending
axons responsive to PLN stimulation also responded to pudendal and hypogastric nerve
Page 106
91 However, the PAG covers a fairly large area, so it is possible that lesioned areas did not
cover the location of the newons involved in micturition
In rhis study, stimulation of more caudal ~egions of the PAG not ody elicited
bladder contraction. but also antidrornically activated 17 sacral spinal cord nemns. This
may seem to be a low yield but one explanation for the low yield in antidromically activated
neurons is that we were not stimulating in the right area of the PAG rezeiving bladder
transmission from the s d cord. The PAG sites which when stimulated elicit a large
bladder response may not correspond to the site to which spinal projections tenninate. Even
the location of the maximum PW-evoked field potentials in the PAG may not correspond
to the termination site of the ascending pathway in the PAG if the projection teminates onto
a mail nucleus or if the termiaals of the pmjecting n e m m spread diffusely in the PAG.
Since retrograde studies have s h o w sacral neurons pmjecting to lateral PAG in cats
(Vanderhorst et al, 1996), stimulation at this site may result in more identiocation of PAG
antidrornicaliy activated neurons conveying PLN input
It is possible that the activity of the projecting neurons are suppressed by the
anesthesia used during our search for single in the sacral spinal cord. However, on-
going studies in decerebrate cats (J.W. Downie and HJ. Du, unpublished) have not
produced a higher yield in antidromicaliy activated neurons respoasive to PLN stunulation,
Although a yield of 17 antidrornically activated newns in the sacral cord seems
low, it corresponds with our fiidings in retrograde tracing experiments and tracing and
neurophysiological results from other laboratories. Vanderhorst et al (1996) only found an
absolute count of 115 retrogradely labelleci neurons in S2 fmm an injection of WGA-HRP
into the ventdateral PAG (P1.0). In our retrograde study, we estimated a minumum of 48
and a maximum of 293 labelleci ce& in S2. Bearing in mind that only a srnail area of the
PAG was stimuiated in each experiment in our spinal cord unit studies, the yield of 17
antidrornically activated neurons sounds reasonable. FiIrthemore, Yezierski and Schwarz
Page 107
92 (1986) only found 13 antidrornicaIly activated neurons in L7-S 1 foilowing stimulation of
13 sites within the caudal PAG using an array of 2- 4 stimulating electrodes.
'Ik median conduction velocity for the pmjecting neurons found in this study (3 1.4
mfsec) was similar to the conduction velocity estimated for spinomesencephalic rract œlls in
the lumbosacral cord antidromically activated h m the PAG (Pl 5 ) in lamina 1, V-VIII
(34.9 f 2 1.1 dsec) (Yezierski and Schwartz, 1986). Yezierski and Schwartz (1986)
mentioned that variations in estimated conduction velocities may be amibuted to cells
projecting to different midbrain levels; cells in different hinctional classes; ceils in different
spinai hmhae. Spinomesencephalic projecting cells originating in the superficial spinal
Iaminae (1 and II) have been reported to have slow conduction velocities (mean: 14.1 f 5.7)
while cek in lamina III-N and VII-VIII conduct faster (mean: 56.3 f 20.8 mlsec). The
range they reported for cells found throughout lamina 1-MII that were activated by
stimulation of the PAG at AKI was 8.8-102 mkc. In our study, the only antidromically
activated œll that received PLN input had a conduction velocity of 8.8 dsec, which would
f a within the large range reported by Yezierski and Schwarz (1986), but this neuron was
Iocated in lamina VlL
The estimated median conduction velocity for the ascending axons to the PAG
conveying bladder information in our spinal cord antidromic unit study (3 1.4 m/sec) was
not signincantiy different fiom the one estimated in the field potential recordings (33.7
m/sec). The latency for the second component of the multiphasic field potential response
was used to calculate these conduction velocities. This latency was used because it was the
fastest component which was consistently present
According to our centrai conduction veIocity estimations. the most common fibers
used by these projectïng neurons to convey bladder information to the PAG are Adelta
fibers. McMahon and M o d n (1982a) estirnated that the conduction velocity for fibers
Page 108
93 conveying PL input between U and the brainstem to be about 30 dsec, which
corresponded with the conduction velocities detedeci for other autonomie spino-bulbar
pathways (Coote Bi Dowmnan. 1966.20-30 dsec for car& and rend nerves). Their
calcdations supports the median conduction velocity estirnated in die present study.
DeGroat (1975) reported a conduction velocity of 1û-11 dsec for the ascending fibers to
Barrington's nucleus (using a latency of 3 0 msec from PLN e v o M field potentials in
the rostral pontine areas (Bmhgton's nucleus) and a conduction distance of 400 mm). The
latency of PLN-evoked potentials recorded in Barrington's nucleus @eGn,at, 1975; Lalley
et al. 1972) is longer than the latency recorded in the PAG in our shidy. A finding
compatible with the afferent infornation king relayed through the PAG before it miches
Barrington's nucleus.
We sought to determine whether the population of sacral spinal ceUs projecting to
the PAG contained a subgroup that responded exclusively to PLN input It was reasoned
that such a population would be most k l y to represent the afferent limb of the micturition
reflex mode1 proposed by Blok and Holstege (1994). However, it was recopized that an
earlier study failed to find ascending nemm with purely PLN input (McMahon and
Morrison, 1982a). This led to a hypothesis in which the specificity in the micturition reflex
was detennined by PL afferent gating of the spinal cord output rather than by bansmission
of bladder-specinc afferent idormation to the brainstem (McMahon and Monison. 1982a).
One dficulty with the latter study is that the supraspinal target of the asçending neurons
was not defined.
In the present study, information from three peripheral nerves converged onto a
sacral spinal neuron located in the dorsal hom and traveled in the same axon to a
ventrolateral region of the PAG (P 1 .O) (figure 19). Although it is not known how eady
dong the transmission path the convergence of input occurs. the central latency of the
mponse to the peripheral nerve stimulation indicates that chis projecting neuron was not a
Page 109
94 first order internewon for any of the three nerve inputs. Different peripheral neme inputs
can travel and terminate ont0 specific neurons in the sacrai cord where they may interact
with other intemettrons in the cord before the information is conveyed to the PAG.
Neurons appear to act as fïrst order interneucon for one input but serve as a higher order
internemon for another type of input (figue 22). Of the 4 antidromidy activated sacral
neurons, only one was a k s t order intemeuron (for SFP input). This neuron &O served as
a higher order internewon for SPUD and CCF mmmision.
Although our data demonstrate a convergence of different peripheral nerve inputs
ont0 spinal-PAG projecting ceUs, the inputs that converged onto these projecting nemm
did not include the PLN. nierefore these d t s do not negate the possibility of a pnvate
pathway for bladder input to the PAG. It does, however, provide evidence for a non-
specific pathway for SPüD, SFP and CCF input to the PAG. The fact that there was an
antidromically activated neuron that responded only to SPUD input suggests that there may
be a sWc and non-smc pathway for the aansmission of SPUD inputs to the PAG.
These data would fit with the data h m the field potential study in which maximum FPS
were found for SPUD in a wide range of AP levels (A2.5-P2.O). and h r e was a cluster of
maximas in the lateral (intemediate) region of the PAG at P0.5-P1.0.
The pudendal nerve conveys afferent information from other structures besides the
urethra. It conveys information £kom somatic structures such as the skin of penis, clitoris
and perineum (Vanderhorst et al, 1996). striated muscles of the pelvic floor (Sato et al,
197 8) and the urerhral and anal mucosa (Kwihara et al, 1980). Input fmm visceral
structures such as the vagina, and part of the uterine cervix are a h conveyed by these
nerves (Vanderhorst et al. 1996). It is possible that different types of input conveyed in the
PUDN separate at the levei of the spinal cord and travel in different paths to enninate at
diffexent regions of the PAG responsible for diffe~ent functions. while other PUD inputs
converge with other peripherai nerve inputs ont0 spinal projecting neurons.
Page 110
95 The PAG has already been shown to be involved in a variety of functions such as
lordosis (Sakuma and Pfaff, 1979). cardiovascular regdation (Lovick, 1993). vocalization
(Zhang et al, 1994) and pain modulation (Bandlet et al, 1991; Carrive and Bander, 1991).
Vanderhorst et al (1996) observed that in cats, neurons remgradely filied by KRP injection
to the lateral PAG were located in regions that overlapped with pelvic and pudendal afferent
terminais in the sacral cord. Also in cats, Blok and Holstege (1994) demonstrateci that the
lateral part of the PAG containecf neufons projecting to Barrington's nucleus. 'Ibis is the
region that, in our work, contains a clustering of maximum field potentials for SPUD
stimulation (figure 14). Therefore, neurons in the lateral PAG which receive projections
h m the sacral cord (in an area of PUD afferent tenninations) could convey PUD
information to Bamington's nucIew, more specifically the L-region, where neurons
projecting to Onuf 's nucleus are located (Holstege et al, 1986). Puciendal aEerent
information related to rnicturition may be conveyed in a private pathway to the lateral region
of the PAG. The other inputs carried by the pudendal nerve rnay terminate in other regions
of the PAG related to lordosis, or pain modulation. This is supported by our fmding that
only 4 of the l? ceus we found projecting directly to the PAG received inputs fiom our test
nerves which implies that the major part of this pathway is not directly related a lower
urinary tract function.
The fmding that only 1 out of 4 nemns which was antidrornicdy activated from
the PAG was found to receive a PLN input suggests that there are very few direct
projections to the PAG conveying bladder-specific information. Even so, this is also
evidence for ci-t projections to the PAG h m the sacral cord conveying PLN input. It
was unfortunate that this antidromicaily activated neuron which received PLN inputs
(figure 18F) was lost before other peripheral nerve inputs could be tested to answer the
question of selectivity in the ascending pathway for michnition.
Page 111
96 Although the yield of antidromically activated sacrai spinal cord neufons was low.
there were numerous neurons in which were synapticaily activated nom the PAG. These
nemns which were orthodromicaily activated wese located in the ventral horn and white
matter of the sacrai spinai cord, contralateral to the site of stimulation in the lateral and
ventrolateral PAG. Even though this could represent a direct PAG to spinal cord pathway
(Mouton and Holstege, 1994). previous work suggest that PAG stimulation could also
cause activation of other descending pathways fkom, for example, raphe nuclei (Chandler et
al. 1994) or ventral medulla (WiIlis, 1988) (generally inhibitory) or even from Barrington's
nucleus (Wrely to be excitatory).
Yezierski and Schwan (1986) found that PAG stimulation inhibited not only
background activity or responses evoked by low or high intensity cutaneous stimuli , but
also responses to the stimulation of deep smicnires such as joint andlor muscle. Because
the PAG regions receiving sacral spinal cord input can exert inhibitory or facilitatory effecu
on s d neurons receiving PLN inputs (figure 22). this projection may constitute an
important Serent limb of a negative or positive feedback loop involved in the control of
spinal neurons involveci in michintion.
The regions of the PAG receiving input fiam the sacral cord have been shown to be
involved in the descending control of spinal neurons (Liebeskind et al, 1W3; Oliveras et al,
1976). actions of morphine (Yaksh et al, 1976). sexual (Hansen and Gummesson, 1982)
and locomotor behaviours (Skulrety. 1963). This area is also h o w n to influence or elicit
motional responses such as fear and anxiey, defensive behaviours and vocalization
(Behebani, 1995). Spinal-PAG projecting cells provide aEerent inputs to an area important
in the motivational e t i v e dimensions of pain. The perception of pain is closely
associated with bladder dysfunction. In the absence of micturition, nIling of the bladder
produces a feeling of urgency and severe discornfort, leading to pain. Therefore. pain input
Page 112
97 (due to bladder overdistension) could be mediated by afferent fifibers in the PLN. along with
other inputs such as bladder fùllness. and conscious desire to void
Conclusion
To conclude and speculate on the significance of our fïndings, inputs related to
bladder fimction enter the sacral spinal cord at S2. to terminate around the dorsal mot entry.
Lissauer's tract, and along the lateral coilateral pathway. The information is then conveyed
to spinal interneurons located mainiy in the sacral parasympathetic nucleus. Although the
micturition pathway has been shown to be polysynaptic. the &gree of synapsing that
occurs in the cord (or in the brainstem) is not known. What is known is that eventually.
bladder-related information is conveyed by spinal projecting neurons (lamina V-VU) to the
ventdateral PAG (P1.0). This conclusion is supported by our fudings of retrugradely
labelleci neurons from WGA-KRP injections into the PAG within the vicinity of c-fos
induced expression by bladder distension. PLN stimulation activating neumns in the
dorsolateral and ventrolateral PAG and PAG stimulation resulting in an increase Ui bladder
pressure imply a role for the PAG in bladder functions. The latency for these evoked
responses in the PAG was shorter than those reported in Barrington's nucleus which is
consistent with bladder inputs king conveyed to the PAG before they are transmitted to
Banington's nucleus. These fhdings support our hypothesis that the PAG is involved in
the basic micturition reflex. But the question of selectivîty in input transmitted in the
ascending limb of the michintion pathway remains imsolved.
Significance
The proceshg of bladder afferent activity for micturition at the level of the spinal cord and
brainstem is not weU understood. However. this is important in the developement of our
understanding of the organization of this physiological process. This pmject should
Page 113
98 provide a stmt in the unravehg of the various roles of ascending bIad&r affixent
infoxmation and addressing the question of convergence and redundancy in these
pathways. The devance of this infornation cm be seen in sensory dysfunction and spinal
cord injury where blad&r control is disturbed at the se11sory level. An understanding of the
bladder afferent pathway may aid in the development of pharmacologial interventions to
manage bladder dysfunction.
Page 114
Appendix A
DAB (brown) immunohistochemistry protocol for Fos protein and *biotin
dextran detection.
1. Place h e d spinal cord specimens in sumse bufXer (20% in O. 1 phospate buffer)
ovemight
2. Section using k z i n g mirotorne and collect sections in 0.05 M phosphate buffesahe
(PBS).
3. Wash 3 times in PB S (1 minute per wash).
4. Make 2% rabbit senun (dilute 200 pL aliquot of rabbit senun into 10 mL of 1% Triton-
X in PBS).
5. Make blocking senun (dilute 200 pL aliquot of rabbit senun into 4.5 mL of 2 4 rabbit
se- and add 100 mg bovine serum albumin).
6. Pellet sections and incubate for 90 minutes in 1 mL of blocking serum.
7. Wash 3 t h e s in PBS.
8. Pellet sections and add 1 mL of 2% rabbit senun plus primary antibody 1: 1000.
Incubate sections overnight at room temperature on shaker.
9. Wash 3 times in PBS and peliet
10. Incubate for 1-2 hours in 1 mL of BioRas 1: 1000 (Biotuiylated Rabbit Anti-Sheep to
2% rabbit senun),
11. Make up ABC solution 30 minutes pnor to use (see notes below).
12. Wash sections 3 times and pellet
Page 115
13. Incubate for 1-2 hours in 1 mL of ABCelite 1 : l W (Avidin-Biotin-peroxidase
Corn piex).
14. Wash sections 3 times in PBS and pellet
15. Weigh out a srnail amount of DAB (diaminobenzidine). Wear gloves.
16, Dissolve the DAB in a smaU amount of distilled water.
17. Add appropriate volume of 0.1 M phosphate buffer such that the DAB is at a
concentration of O 5 mg/mL.
18. Make up fie& hydrogen peroxide (dilute 1 mL of 30% hydrogen peroxide into 100 mL
of distilled water).
19. Rem sections in DAB until the sections start to turn brown (approximately 20 minutes)
by adding 5 mL of 0.5 mg/mL of DAB solution and 150 of diluted hydrogen
peroxide to the sections.
20. Pour the DAB solution into the waste DAB bottle and wash the sections 3 tirnes in
PBS.
2 1. Store sections in the fridge.
* for biotin dextran, the proceshg of sections start at step #Il.
Page 116
Appendix B
Nickel enhanced DAB (black) immunohistocbemistry Protocol.
Repeat steps 1-16 in Appendbc A for standard DAB (brown) immunohistochemistry
except al l stages must be done in TBS not PBS or PB.
Make up 0.6% ammonium-nickel sulfate solution (0.6 g (NH4)2Ni(Sû4)2 in 100 mL
of TBS. pH 8.0).
Add 0.6% ammonium-nickel solution to DAB such that DABM is at a concentration of
0.16 mg/mL.
Make up nesh düute hyârogen peroxide (dilute 1 mL of 3û% hydrogen peroxide into
100 mL of distilled water).
React sections in DABm solution untü they start to tum gray (approximately 3
minutes) by adduig 5 mL of the 0.16 mg/mL D M i solution and 80 pL of the
hydrogen peroxide to the sections.
Pour the DABINi solution into the waste DAB bottle and wash sections 3 times in TBS.
Store sections in the fndge.
Page 117
Appendu C
Histochemi*stry for the deteetion of WGA-HRP or HRP.
Make Solution A: dissolve 200 mg sodium nitrofemcyaaide into 185 mL of distilled
water and then add 10 mL of acetate buf5er (pH 3.3). StV for approximateIy 5 minutes.
Make Solution B: dissolve 15 mg 33 '5 5'-tetramethyl bemidine (TMB) into 7.5 mL
of absolute ethanol by gentiy heating (37-40 OC).
Mùr solutions A and B (Solution C) for approxirnately 15-20 seconds and add to
sections.
Incubate for 20 minutes in the da*
Repare fresh 0.3% hydrogen peroxide and add 2 mL of 0.3% hydmgen peroxide per
100 mL of Solution C to the sections.
Incubate for 20 minutes in the da&
Wash sections 3 times with post reaction solution (dilute 50 mL of acetate bdfer, pH
3.3. in 950 mL distilled water) and store sections in this solution in the fndge.
Page 118
REFERENCES
N-Chaer, E., Lawand, N.. Westlund, K. and Willis, W. Pelvic visced input into the
nucleus gracüis is largely mediated by the polysynaptic dorsal column pathway. J.
Neurophysiol. 76:2675-2690 (1996).
Andrew, J. and Nathan, P. Lesions of the anterior frontal lobes and disturbances of
micturition and defaecation. Brain. 87:233-262 (1964).
Aston-Jones, G., E ~ i s , M. Pienbone, V., Nickeil, W. and Shipley, M. The brain nucleus
Iocus coenileus: Restricted afferent control of a broad efferent network. Science. 234734-
737 (1986).
Awad, S., Bruce, A. Carro-Ciampi, G., Downie. J.. Lin, M. and Marks. G. Distribution
of alpha- and beta-adrenoceptors h human urinary bladder. Br. J. Pharmacol. H):525-529
(1974).
Bahns, E.. Ernsberger, U., Janig, W. and NeIke, A. Functional characteristics of lum bar
visced afferent fibers h m the urinary bladder and the urethra in the cat Wugers Arch.
4û7:510-518 (1986)
Bahns, E., Halsband, U. and Janig, W. Response of sacral visceral afferents from the
lower urinary tract, colon and anus to mechanical stimulation Ptlugers Arch. 410:296-303
(1987).
103
Page 119
104 Bandler. R and Carrive, R Integrated defeace reaction elicited by excitatory amino acid
microinjection in the midbrain periaqueductal grey region of the umestraùied cat. Brain
Ra. 439:95- 106 (1988).
Bandler, R, Carrive, P. and Zhang, S. Integration of somatic and autonornic reactioos
within the midbrain periaqueductal grey: Viscerotopic, sornatotopic and hinctional
organization. Prog. Brain Res. 87:269-305 (1991).
Bandler, R and Depaulis, A Elicitation of intraspecinc defence reactions in the rat h m
midbrain periaqueductal grey by microinjection of kainic acid, withourt neurotoxic effects.
Neurosci. Lem 88:291-296 (1988).
Bandler, R., McCuiloch. T. and Dreher, B. Merents to a midbrain periaqueductal grey
region involved in the 'defence reaction' in the cat as revealed by horseradish peroxidase. L
The telencephalon. B*iin Res. 33O:lW- 1 19 (1985).
Barrington, F. The nervous mechanism of micturition. Q u a J. Exp. Physiol. 8:33-71
(19 15).
Barrington, F. The relation of the bd-brain to micturition. Brain. 4:23-53 (1921).
Barrington, F. The effect of lesion of the hind- and mid-brain on micturition in the cat.
Quart. J. Exp. Physiol. 15:81-102 (1925).
Behbehani, M. Functional chafacterististics of the midbrain periaqueductal gray. Rog.
Neurobiol. 46575-605 (1995).
Page 120
Beitz, A The midbrain periaqueàuctal gray in the rat L Nuclear volume, cell number,
demity, orientation, and regional subdivisions. J. Comp. Neurol. 237:445-459 (1985).
Birder, L. and deGroat, W. Increased c-fos expression in spinal nemm after imitation of
the Iower urinary tract. J. Neurosci. 12:4878-4889 (1992).
Birder, L.. Roppolo, I.. Iadarola, M. and deGroat. W. C-fos as a mader for subsets of
visceral second order newons in the rat lambosacral spinal cord. Soc. NeuroSn Abstr.
16:703 (1990).
Blaivas, J. The neurophysiology of micbintion: a clhical study of 550 patients. I. UroL
127:948-963 (1982).
Blaivas, J., Sinha, H., Zayed, A and Labib. K. Demisorextemal sphincter dyssynergia.
J. Urol. 125:542-557 (198 1).
Blok, B., De Weerd, H. and Holstege, G. Ultrastructural evidence for a paucity of
projections h m the lumbosacral cord to the pontine michvition center or M-region in the
cat: a new concept for the organization of the rnicturition reflex with the periaqueductal gray
as central rday. J. Comp. Neurol. 359:300-309 (1995).
Blok, B. and Holstege, G. Direct projections fkom the periaqueductal gray to the pontine
micturition center (M-region). An antemgrade and retrograde tracing study in the cat.
Netuosci. Lett. 166:93-96 (1994).
Page 121
106 Blok, B. and Holstege, G. Ultrastruchnal evidence for a direct pathway fkom the pontine
mictuntion œnter to the parasympathetic preganglionic motonewons of the bladder of the
cat Neurosci. Lett 222:195-198 (1997).
Blomqvist, A. and Craig, A Organization of spinal and trigeminal input to the PAG. In:
The Midbrain Periaqueductal Gray Matter. Depauiis, A. and Bandler, R (eds) Plenum
Press, New York pp345-363 (1991).
Blomqvist, Ma and Berkely. Spinal input to the parabrachial nucleus in the cat. Brain Res.
480:29-36 (1989).
Bradley, W. and Teague, C. Cerebellar control of the urinary bladder. Exp. Neurol.
23:399-411 (1969).
Brown, P. and Fuch, J. Somatotopic representation of hindlimb skin in the cat dorsal horn.
J. Neurophysiol. 38: 1-9 (1975).
Cameron, A, Khan, I., Westlund, K. and Willis, W. The efferent projections of the
periaqueductal gray in the rat: a Phaseolus vulgaris-Leucoagglutinin study. IL Descending
projections. J. Comp. Neurol. 35 l:585-60 1 (1995).
Carrive, P. and Bandler, R Viscerotopic organization of neuroas subsewing hypotensive
reactions w i t b the midbrain periaqueductal grey: A cornkitive functional and anatomid
study. Brain Res. 541:206-215 (1991).
Page 122
Cedarbaum. J. and Aghajanian. G. Afferent projections to the rat locus coeruleus as
detennined by a retrograde tracing technique. J. Comp. NeuroL 178: 1-16 (1978).
Chai, T. and Steen. W. Neurophysiulogy of micturition and continence. Urol. North
A m e h 23:22 1-236 (1996).
Chambers, W. Elecnical stimulation of the cerebellum in the cat. Am. J. Anat. 8055-93
(1969).
Chandler. M., Oh, U., Hobbs, S. and Foreman, R Responses of feline raphespinal
neurons to urinary bladder distension. J. Auton. New. Sys. 47:213-224 (1994).
Cooteman, H. and Downman, C. Central pathways of some autonomie reflex discharges
J. Physiol. 1 8 W 14-729 (1966).
Craig. k and Mense, S. The distribution of afferent fibers from the gastrocnemius-soleus
muscle in the dorsal horn of the cat, as revealed by transport of HRP. Neurosci. Leu.
41:233-238 (1983).
deGroat, W. Nervous control of the uruiary bladder of the cat Brah Res. 87:201-2 11
(1975).
deGroat, W. Spinal cord projections and neuropeptides in viscerd afEerent neurons. pmg.
Brain Eh. 67:166-187 (1986).
Page 123
108 &mat, W. Birder, L. Morgan, C. Roppolo, J. Yoshiyama., M. Suzuk T. Recent
developments in the neuropharrnacology and neurophysiology of the lower urinary tract
Neurourol. Urodynamics. 1 1 :269-272 (1992).
deGmat, W. and Kawatani, M. Neural c o n t d of the urinary bladder: possible relationship
between peptidergic inhibitory mechanisms and detnisor instability. Neurourol. Urodyn
4:285- (1985).
deGroat, W. Nadelhaft, 1. Milne, R Booth. A. Morgan, C. Thor, K. mganbation of the
sacral parasympathetic reflex pathways to the utinaqr bladder and large intestine. J. Auton.
New. Syst 3: 135- 160 (1981).
deGroat, W. and Ryall. W. Reflexes to sacrai parasympathetic neurones concemed *th
michintion in the cat. J. Physiol. 200:87-108 (1969).
Demy-Brown, D. and Robertson, E. On the physiology of m i c t d o n . Brain. 56: 149
(1933).
Depaulis, A, Keay, K. and Bander, R Longitudinal neuronal orgaaization of defensive
reactions in the midbrain periaqueductal gray region of the rat Exp. Brain Res. 90:307-3 18
(1992).
Ding, Y., Qin, B., Li, J. and Minino, N. Induction of c-fos-like protein in the
spinoparabrachial tract-neurons locating within the sacral parasympathetic nucleus in the
rat Brain Res. 659:283-286 (1994).
Page 124
Ding, Y. Takada, M. Tokuno, H. Minino, N. Direct projections h m the dorsal pontine
tepentu. to pudenda1 motonewons innervatiog the extemal urethral sphincter muscle in
the rat. J. Comp. NemL 357:318-330 (1995).
Downie, J., Champion, J. and Nance, D. A quantitative analysis of the aEerent and
extrinsic efferent innervation of specinc region of the bladder and urethra in the cat Brain
Res. Bull. 12:735-740 (1984).
Edvardsen, P. and Setekleiv, J. Distribution of adrenergic receptors in the urinary bladder
of cats, rabbits and guinea-pigs. Acta Pharrnac. Toxîc. 26:437-445 (1968).
Edvardsen. P. and Ursin, T. Micturition thresholds in cats with amygdala Iesions. Exp.
Neurol, 2 l:49S-SO 1 (1968).
Elbadawi. A Functional anatomy of the organs of micturition. UroL North America
23: l77-2O9 (1996).
Fedirchuk, B. and Shefchyk, S. Effects of electncal stimulation of the thoracic spinal cord
on bladder and ex te rd urethral sphincter activity in the decerebrate cat Exp. Brah Res.
84:635-642 (199 1).
Fields. H., Partridge, L. and Winter, D. Somatic and viscerd receptive field properties of
fibers in ventrai quadrant white matter of the cat spinal corci. J. Neurophysiol. 33:827
(1 970).
Page 125
Fietcher, T. and Bradley. W. Merent nerve endings in the urinary bladder of the cat Am.
J. Anat 128: 147- 158 (1970).
Gjone, R Excitatory and inhibitory bladder response to stimulation of 'limbic",
diencephalic and mesencephalic structures in the cat Acta PhysioL Scand. 66~91-102
(1966).
Griffith, D., Holstege, G., Dalm, E. and de Wail, H. Controls and coordination of
bladder and urethrai function in the braînstem of the cat Neurourol. Urodyn. 9:63-82
(1990).
Hansen, S. and Gummesson, B. Participation of the lateral midbrain tegrnentum in the
neuroendocrine control of sexud behaviour and lactation in the rat. Brain Res. 25 l:3 19-
325 (1982).
Habler, H., Janig, W. and Koltzenburg, M. Activation of unmyehateci afferent fibers by
mechanical stimuli and inflammation of the urinary bladder in the cat. J. PhysioL 425545-
562 (1990).
Hida, T. and Shirnizu, N. The interrelation between the laterodorsal tegmental area and
Iumbosacral segments of rats as studies by HRP rnethods. Arch. Histol. Jap. 45:495-504
(1982).
Holstege, G. Neuronal organization of micturition. Neurourol. Urodyn. 11:273-277
(1992).
Page 126
Hoistege, G., GRniths, D., De Wail, H. and Dalm, E. Anatomid and physiological
observations on supraspinal control of bladder and urethral sphincter muscles in the cat J.
Comp. NeuroL 250:449-46 1 (1986).
Honda, C. Viscerd and somatic aEerent convergence onto nemm near the central canal in
the sacral spinal cord of the cat J. Neurophysiol. 53: 1059-1078 (1985).
Hunt, S., Pini, A and Evan, G. Induction of c-fos-like protein in spinal cord neurons
following sensory stimulation. Nature. 328:632-634 (1987).
Janig, W. and Koltzenburg, M. On the frmction of spinal primary afferent fibers supplying
colon and urinary bladder. 3. Auton. Nerv. Syst 30:S89-S96 (1990).
Jenkins, R. and Hunt, S. Long-tenu increase in the levels of c-jun mRNA and jun protein-
like immunoreactivity in motor and sensory neurons following axon damage. Neurosci.
Lett. l29:lO7-llO (1991).
Jurgens, U. and Pratt, R Role of the periaqueductal grey in vocal expression of ernotion.
Brain Res. 167:367-378 (1979).
Kabat, H., Magoun, H. and Ranson, S. Reaction of bladder to stimulation of points in the
forebrain and midbrain. J. Comp. Neurol. 63:211 (1936).
Kawatani, M., Houston, M., Rutigliano, M., Erdman, S. and deGroat, W. Colocalization
of neuropeptides in afferent pathways to the urinary blad&r and colon: Demonstration with
Page 127
112 double color immuoohistochemistry in combination with axonal tracing techniques* Soc.
Neurosci. Abstr, 1 1: 145 (1985).
Kerr, F. and Alexander, S. Descendhg autonomie pathways in the spinal cord. Arch.
Neurol. 10:249-261 (1964).
Kim- Y., Ukai. Y., Kirnura, K., Sugaya, K. and Nishizawa, O. hhibitory infiuence
h m the nucleus reticularis pontis oralis on the mictuxition reflex induced by electrical
stimulation of the pontine mictuntion center in cats. Neurosci. Lett. l9S:2 14-2 16 (1995).
Knise, M. and dffimat, W. Micturition reflexes in ciecerebrate and spinalized neonatal rats.
Am. J. Physiol. 258:R1508-RI511 (1992).
Kum, M. The nervous control micturition. Physiol. Rev. 45:425-494 (1965).
K m , M. and Yamamoto, H. Fiber connections of the pontine detnisor nucleus
(Barrington). J. Comp. Neurol. 123: 161- 186 (1964).
Kuzuhara, S., Kanayawa, 1. and Nkanaishi, T. Topographical localisation of the Onufs
nuclear neurones innervating the rectal and vesical saiated muscle sphincters: a retrograde
fluorescent double labehg in cat and dog. Neurosci. Len 16:125-130 (1980).
Langley, J. and Anderson, & The innervation of the pelvic and adjoining viscem Part II.
The bladder. J. Physiol. (Lond). 19: 85- 139 (1 895).
Page 128
113 Langley, J. and Anderson. K The innervation of the pelvic and adjoining viscera, Part
W. Anatomical observations. J. Physiol. (Lond). 20:372-406 (1896).
Lalley. P., deGroat, W. and McLain, P. Activation of the sacral paresympathetic pathway
to the urinary bladder by brain stem stimulation. Fed Rc. 31:386 (1972).
Liebeskind, J., Gdbaud. G.. Besson, J. and Oliveras. J. Anaigesia from electrical
stimulation of the periaqueductal gray matter in the cat behaviod obsexvations and
inhibitory effects on spinal cord intemeurons. Bmin Res. 50:441-446 (1973).
Levin, R and Weîn, A Distribution and function of adrenergic receptors in the urinary
bladder of the rabbit Molecu. Phannac. 16:441-448 (1979).
Loewy, A. Saper. C. and Baker. R Descendhg projections from the pontine micturition
center. Brain Res. 172533-538 (1979).
Lovick, T. Integrad activity of cardiovascular and pain regdatory systems: role in
adaptive behavioural responses. Rog. Neurobiol. 4O:63 1-644 (1993).
Lui. R Laminar ongins of spinal projection neurons to the periaqueductal gray of the rat
Brain Res. 264: 118-122 (1983).
Lumb, B. and MonWn, J. An excitatory influence of dorsolateral pontine structures on
urinary bladder motility in the rat. Brain Res. 435363-365 (1987).
Page 129
Ma, W., Blomqvist, A and Berkley. W. Spino-diencephalic relays through the
parabrachial nucleus in the cat Brain Res. 480:37-50 (1989).
Magoun, H., Atlas. D., IngersoU, E. and Ranson. S. Associateci facial, vocal and
respiratory componentî of emotiond expression: An experimental study. J. Neurol.
Psychopath. 17:241-255 (1937).
Mallory, B.. Steers, W. and deGmat, W. Electrophysiological study of micairition refiexes
in the rat. Am. J. Physiol. 257:R410-R421 (1989).
Martner, J. Influences on the defecation and rnicauion reflexes by the cerebeUar fastigiai
nucleus. Acta Physiol. Scand. 9495- 104 (1975).
Matsushita, M- and Tanami, T. Contralateral teminations of primary afferent axons in the
sacml and caudal segments of the cat, as studied by anterograde transport of horseradish
peroxidase. J. Comp. Neurol. 220:206-218 (1983).
Mawe, G., Bresnahan, J. and Beattie, M. Rimary afferent projections from dorsal and
ventral roots to autonomie preganglionic newons in the cat sacral spinal cord: Light and
electron microscopic observations. Brain Res. 290: 152- 157 (1 984).
Mawe, G., Bresnahan, J. and Beattie, M. A light and electron microscopic analysis of the
sacral parasympathetic nucleus after labelling af5erent and efferent elements with HRP. J.
Comp. Neurol. 250:33-57 (1986)
Page 130
115
Mayer, D. Wolfe, T. Akil, H. Carder, B. LiebesLiad. J. Analgesia from elecaical
stimulation in the midbrain of the rat Science. 174: 1351-1354 (1971).
McMahon, S. and Morrison, J. Factors that determine the excitability of parasympathetic
reflexes to the cat bladder. J. Physiol. 322:35-43 (1982b).
McMahon, S. and Morrison, J. Inhibitory interactions between colonic and vesical
afferents in the rnicturition refiex of the cat J. Physiol. 322:45-52 (1982~).
McMahon, S. and Momson. J. Spinal neurones with long projections activated from the
abdominal viscera of the cat. I. Physiol. 322: 1-20 (1982a).
McMahon, S. and S p W e , K. Brainstem influences on the parasympathetic supply to the
urinary bladder of the cat Brain Res. 234237-249 (1982).
MeIzack, R, Stolter, W. and Livingston, W. Effects of discrete brainstem lesion in cats on
perception of noxious stimulation. J. Neurophysiol. 21 :353 (1958).
Morgan, C., deGroat, W., FeUBns, L. and Piang, S. Axon collaterah indicate broad
intraspinal mle for sacral pregangiionic neurons. Roc. Nati. Acad. Sci USA 88:6888-
6892 (199 1).
Morgan, C., deGroat, W., Felkins, L. and Zhang, S. IntraceUu1a.r injection of neurobiotin
or horseradish peroxidase reveak separate types of preganglionic neurons in the sacral
parasympathetic nucleus of the cat J. Comp. Neurol. 331:161-182 (1993).
Page 131
Morgan, C., Nadelhaft, L and m a t , W. Localbation of bladder preganglionic neurons
within the sacral parasympathetic nucleus of the cat NeUrosci. L e t 14: 189-194 (1979).
Morgan, C., Nadelhaft, 1. and deGroat, W. The distribution of visceral primary afferents
fkom the PLN within Lissauer's tract and the spinal gray matter and its relationship to sacral
parasympathetic nucleus. J. Comp. NeuroL 201 :4 15-440 (198 1).
Morrison, J. Neural connections between the lower urinary tract and the spinal cord. In:
The Physiology of the Lower Urinary Tract Torrens, M. and Momson, J. (eds) Spring-
Veriag, Berlin Heidelberg (1 987).
Mouton, L. and Holstege, G. The periaqueductal gray in the cat projects to iamina VIII and
the medial part of lamina W throughout the length of the spinal cord Exp. Brain Res.
101 :253-264 (1994).
Nadelhaft, L and Booth, A. The location and morphology of preganglionic neurons and the
distribution of visceral afferents from the rat pelvic nerve: A horseradish perxoidase study.
J. Comp. Neurol. 226:238-245 (1984).
Nadelhaft, I., Morgan, C. and deGroat, W. Localkation of the sacral autonomie nucleus in
the spinal cord of the cat by the horseradish peroxidase technique. J. Comp. NeuroL
193:265-28 1 (1980).
Page 132
117 Nadelha L and Vera, P. Central nervous system neurons infected by pseudorabies vins
injected into the rat urinary bladder foiiowing dateral transection of the pelvic nerve. J.
Comp. Neurol. 359:443-456 (1995).
NadeIhaft, 1. and Vera, P. Neurons in the rat brai . and spinal cord labelled a k r
pseudorabies Wus injected into the extemal urethral sphincter. J. Comp. NeuroL 375502-
517 (1996).
Nadelbaft, L, Vera, P. Card, J. and Miselis, R Central nervous system neurons labeiied
following the injection of pseudorabies virus into the rat urinary bladder. Neurosci. Lett
143~27 1-274 (1992).
Nathan, P. Awareness of bladder nIling with divided sensory tract: J. Neurol. Neurosurg.
Psychiat 19:lOl-105 (1956).
Nathan, P. and Smith, M. The centripetal pathway from the bladder and urethra within the
spinal cord. J. Neurol. Neurosurg. Psychiat. 14:262-280 (195 1).
Nergardh, A. and Boreus, L. Autonomie receptor fiinction in the lower urinary tract of man
and car Scand. J. Urol. Nephrol. 6:32-36 (1972).
Nishizawa, O., Ebina, K., Sugaya, K., Noto, H., Satoh, K., Kohama, T., Harada, T.
and Tsuchida, S. Effect of cerebellectomy on reflex micturition in the decerebrate dog as
detennined by urodynamic evaluation. UroL Int 44: 152-156 (1989).
Page 133
Noto, H., Roppolo. I.. Steers, W. and deGroat, W. Excitatory and inhibitory influences
on bladder activity by e1ectfica.i stimulation in the pontine micturition center in r;u. Brain
Ra. 492:99- 1 15 (1989).
Noto, Et, Roppolo. W., Steers, W. and de Groat, W. Electrophysiologicai analysis of the
ascending and descendhg components of the micaintion reflex pathway in the rat Brain
Ra. 549:95- 105 (199 1).
Oliveras, J., Besson, B., Guilbaud, G. and Liebeskind, J. Behavioural and
electrophypsiologicd evidence of pain inhibition nom midbrain stimulation in the cat Exp.
Brain Res, 20:32-44 (1976).
Otake, K. and Nakamura, Y. Single neurons in Bamington's nucleus projecting to both the
paraventricular thalamic nucleus and the spinal cord by way of axon collaterals: a double
labelling study in rat: Neurosci. Len 20997-100 (19%).
Reynolds, D. Surgery in the rat during electricd analgesia by focal brain stimulation.
Science. 164:444-445 (1969)
Rizvi, T., En&, M., Aston-Jones, G., Jiang, M., Liu, W., Behbehani, M. and Shipley,
M. Preop tic projections to Barrington's nucleus and the pericoenilear region: architecture
and tenninai organization. J. Comp. Neurol. 347: 1-24 (1994).
Page 134
119
Rwi, T., Ennis, M., Behbehani, M. and Shipley, M. Connections between the central
nucieus of the amygdala and the midbrain periaqueductal gray: Topography and Teciprocity.
J. Comp. NeuroL 303:121-131 (1991).
Rizvi, T., Ennis M. and Shipley, M. Reciprocal connections between the medial preoptic
area and the midbrain periaqueductal gray in rat: A WGA-HRP and PHA-L study. J.
Comp. Neurol. 3 15: 1-15 (1992).
Roppolo, J., Nadehaft. L and deGroat, W. The organization of pudendal motoneurons
and primary aEerent projections in the spinal cord of the rhesus monkey revealed by
horseradish peroxidase. J. Comp. Neurol. 234475-488 (1985).
Rudy , D., Downie, J., and Mc Andrew, J. Alphachioralose alters autonomie reflex
function of the lower urioary tract Am. J. Physiol. 26 1:R1560- 1567 (199 1).
Ruggiero, D. and Reis, D. Neurons containhg phenylethanolamine-N-methyltransfera~e:
A cornponent of the baroreceptor diex? IIX Epinephrine in the Central Nervous System. J.
Stolk, D. U'Prichard and K. Fuxe (eds). Oxford Press (1987).
Salcuma, Y. and Pfaff, D. Mesencephalic mechanisms for integration of fernale
reproductive behaviour in the rat Am. J. Physiol. 237:R278-R284 (1979).
Sasa, M. and Yoshimura, N. Locus coenileus noradrenergic newons as a micturition
center. Microsc. Res. Tech. 29:226-30 (1994).
Page 135
120 Sam, M., Minino, N. and Konishi, A Localization of motonemm innervating perineal
muscles: An HRP shidy in cat Brain Res. 140: 149-154 (1978).
Satoh, K., Shimizu, N., Tohyama, M and Maexia, T. Locaiization of the micturition reflex
center at domlateral pontiae tegmentum of the rat Neurosci. Leu. 8:27-33 (1978).
Shefchyk, S. The effects of lumbosacral deafferentation on pontine micturition center-
evoked voiding in the decerebrate cat. Neurosci Len 99: 175- 180 (1989).
Simerly, R and Swanson, L. Projection of the medial preoptic nucleus: Phaseolus vulgaris
leucoagglutioanterogra& tract-tracing study in the rat. J. Comp. NeuroL 270:209-242
(1988).
Skultety, F. Relation of periaqueductal gray matter D stomach and bladder motility.
Neurology. 9: 190- 198 (1959).
Skultety , F. S tirnulation of periaqueductal gray and hypothalamus. Arch. Neml. 8:608-
620 (1963).
Slack, B. and Downie, J. Phannacological analysis of the responses of the feline urethra to
autonomie nerve stimulation. J. Auton. Nerv. Syst. 8: 141- 160 (1983).
Snider, R and Niemer, W. A stereotaxic atlas of the cat brain. The University of Chicago
Press (1961).
Page 136
Sugaya, K. Matsuyama, K., Takahisaki, K. and Mori, S. Electrical and chernical
stimulations of the pontine mictuntion center. Neurosci. Lett 80: 197-201 (1987).
Tan, J. and Holstege, G. Anatomid evidence that the pontine laterd tegmental field
projects to lamina 1 of the caudal spinal trigeminal nudeus and spinal cord and to the
Edinger- Westphal nucleus in the cat Neurosci. Lea 643 17-322 (1 986).
Tang, P. and Ruch, T. Localization of brainstem and diencephalic areas controbg the
micturition reflex. J. comp. Neurol. 106:213-245 (1956).
Thor, K., Morgan, C., Nadelhaft, 1.. Houston, M. and dffiroat, W. Organi7iition of
afferent and efferent pathways in the pudendal nerve of the female cat 3. Comp. Neml.
288:263-279 (1989).
Torrens, M. Suprapontine intluences on the mictuntion reflex. Scand. J. Urol. Nephrol.
29:suppl. 175: 11- 13 (1995).
Uemura, E., Fletcher, T. and Bradley, W. Distribution of lumbar afferent axons in muscle
coat of cat urinary bladder. Am. J. Anat 139:389-398 (1974).
Uemura, E., Fletcher, T. and Bradley, W. Distribution of lumbar and sacral afferent axons
in submucosa of cat urinary bladder. Anat Rec. 183579-588 (1975).
Uemura, E., Fletcher, T., Dirks, V. and Bradley, W. Distribution of sacral afferent axons
in cat urùiary bladder. Am. J. Anat. 136:305-314 (1973).
Page 137
Ueyama, T., Minino, N., Nomura, S., Konishi, A., Itoh, K. and Arakawa, H. Central
distribution of afferent and efferent components of the pudendg neme in the ca t J, Comp.
Neurol. 222:38-46 (1984).
Ueyama, T., Minino, N., Takahashi, O., Nomura, S., Arakawa, EL and Marsushima R.
Central distribution of efferent and afferent components of the pudencial nerve in macaque
monkeys. J. Comp. Neurol. 232548-556 (1985).
Valentino, R, Page, M., Luppi, P., Zhu, Y., Van Bockstade, E. and Aston-Jones, G.
Evidence for widespread afferents to Barrington's nucleus, a brainstem region rich in
corticotropin-releasing hormone neurons. Neuroscience. 62: 125- 143 (1994).
Vanderhorst, V., Mouton, L., Blok, B. and Holstege, O. Distinct cell groups in the
lumbosacral cord of the cat project to different areas in the perkqueductai gray. J. Comp.
Neurol. 376:36 1-385 (1996).
Vaughan, C. and Satchell, P. Urine storage mechanisms. Prog. Neurobiol. 46:2 15-237
(1995).
Vizzard, M., Erickson, V., Card, J., Roppolo, J. and deGroaf W. Tramneuronal labelhg
of neurons in the adult rat brainstem and spinal cord after injection of pseudorabies virus
into the urethra J. Comp. Neurol. 355:629-640 (1995).
Weinberg, R The action of oncogenes in the cytoplasm and nucleus. Science. 230:770-
776 (1985).
Page 138
Willette, R.. Morrison, S., Sapru, EL and Reis, D. Stimulation of opiate receptors in the
dorsal pontine tegmentum inhibits re£iex contraction of the urïnary bladder. J. Phamiad.
Exp. Ther. 2444û3-4û9 (1988).
Willis, W. Anatomy and physiology of descending conml of nociceptive responses of
dorsal hom neurons: a comptehensive review. h g . Br& Res. 77:l-29 (1988).
Winter, D. Receptor characteristics and conduction velocities in bladder afferents. J.
Psychiatr. Res. 8:225-235 (1971).
Yaksh, T., Yeung, J. and Rudy, T. Systematic examination in the rat of brain sites
sensitive to the direct application of morphine: obsewations of differential effects within the
periaqueductal gray. Brain Res. 1 14%- 104 (1976).
Yezierski, R Spinomesencephalic tract: projections h m the lumbosacral spinal cord of the
rat, cat, and monkey. J. Comp. Neurol. 267: 13 1- 146 (1988).
Yezierski, R and Schwartz, R. Response and receptive-field properties of
spinomesencephalic tract cells in the cat. J. Neuroph ysiol. 5576-96 (1986).
Yoshimura, N.. Sasa, M.. Yoshida, 0. and Takaon, S. Alpha-adrenergic receptor-
mediated excitation h m the locus coedeus of the sacral parasympathetic preganglionic
neuron. Life Sci. 47:789-797 (1990a).
Page 139
124
Yoshimura, N., Sasa, M., Yoshida, 0. and Takaori, S. Mediation of the rnicnintion reflex
by centrai norepinephrhe h m the locus coeruius in the cat. J. UroL 143:840-843
(1990b).
Zhang, S., Davis, P., Bander, R and Carrive, P. Brainstem inkgration of v o c ~ t i o n :
Role of the midbrain periaqueductal gray. J. NeurophysioL72: 1337- 1356 (1994).
Page 140
l MAGE NALUATION TEST TARGET (QA-3)
APPLIED IWGE . lnc 1653 Easî Main Street - -. - - Rochester, NY 14609 USA - -- - - Phone: 71 61482-0300 -- -- - - Fax: 71 6i2üû-5989