IS. ... :1

312.!

l 9 vii.-.
. z u :x
itfﬂmnffftl:

. «to 1:. .1 ....

9.
g...
I. ‘

:4. .

a.
a
t

1:: -
hp. '-
¢}e"

 

 

L... L32»:
24:4... r
.V

 

 

MICHIGAN STAT

M l/Illll’lllllNW

 

 

 

 

lllllllflllill/ll

_) 1 3 1293 01420 3206

M . r

 

 

 

 

Ar,
l’t'

This is to certify that the

dissertation entitled

Mechanosensitive Reflex Input from Distal Colon
to Neurons in Pelvic Plexus Ganglia

presented by

Jingyang Lin

has been accepted towards fulﬁllment
of the requirements for

DQCIQEBI degree in _Eh¥siolng¥

Major professor E

Date “"9““

MS U is an Afﬁrmative Action/Equal Opportunity Institution 0-1277 1

 

 

 

LIBRARY"
Mlchlgan State!
Universlty I

 

 

PLACE IN RETURN BOX to remove thb checkout from your record.
TO AVOID FINES Mum on or baton dot. duo.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSUIOAnNﬁmEJve‘ ' 'j ' n

 

MECHANOSENSITIV E REFLEX INPUT FROM DISTAL COLON TO
NEURONS IN PELVIC PLEXUS GANGLIA

By

Jingyang Lin

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Physiology

1995

 

ABSTRACT

MECHANOSENSIT IVE REFLEX INPUT FROM DISTAL COLON TO
NEURONS IN PELVIC PLEXUS GANGLIA

By

Jingyang Lin

Neurons in pelvic plexus ganglia (PG) give rise to postganglionic ﬁbers that
innervate vascular and visceral smooth muscle of colon, anal-canal, urethra-urinary
bladder (UUB) and reproductive organs. Whether these neurons receive

L r‘ J' ‘ ‘ sensory input from distal colon and mediate colonic and other
visceral reﬂexes is not known.

The present studies were performed in vitro using a guinea-pig colon-PG
preparation which consisted of PG, pelvic nerves, hypogasu’ic nerves, colonic-rectal
nerves (CRN) and urethra-urinary bladder nerves (UUBN) with or without an attached
segment of colon or colon and UUB. Intracellular recordings and intraluminal pressure
recordings were used to study electrophysiological properties and synaptic transmission
of PG neurons and PG-mediated reﬂex contractile responses of colon and UUB,
respectively.

In isolated PG, neurons were quiescent. Electrical stimulation of the attached
nerve trunks elicited nicotinic fast excitatory post-synaptic potentials (f—EPSPs) and/or
action potentials (APs) in 70-80% of neurons tested. The inﬂammatory mediator, human
recombinant interleukin-ll} (hrIL-lB) acting on IL-1 receptors, altered the membrane

potential, membrane input resistance and blocked f-EPSPs in some PG neurons. Blockade

 

Jingyang Lin

of f-EPSPs by hrIL-lB was due to presynaptic inhibition of acetylcholine release.

When PG were attached to a colon segment via CRN, 27% of neurons exhibited
continuous f-EPSPs and APs which were reversibly blocked by hexarnethonium and
abolished by sectioning the CRN. Passive colon distension increased the frequency of f-
EPSPs and APs in active neurons and initiated f-EPSPs and/or APs in quiescent neurons
(24%). These results suggest that neurons in PG receive cholinergic mechanosensory
input from the colon. Colon application of the inﬂammatory mediator, bradykinin,
sensitized the mechanosensory input. Distension of one colon segment or electrical
stimulation of colonic nerves evoked contraction of a second colon segment or of the
urinary bladder. Electrical stimulation of UUBN evoked colon contractions, suggesting
that neurons in PG mediate excitatory reﬂexes to the colon and urinary bladder. The
reﬂexes were capsaicin-insensitive, indicating that primary afferents were not involved.

This dissertation showed that neurons in PG receive mechanosensory inputs from
distal colon and mediate visceral organ reﬂexes that are modulated by the inﬂammatory
mediators, bradykinin and hrIL-IB. The PG-mediated reﬂexes may be signiﬁcant in

modulating defecation and micturition reﬂexes physiologically and pathologically.

 

ACKNOWLEDGMENTS

I would like to express my appreciation to all the members of my guidance
committee, Dr. Thomas Adams, Dr. William Atchison, Dr. James Galligan, Dr. Donald
Jump and Dr. Sharleen Sakai, for their generous advises and supports. Their help,
especially the help from Dr. Galligan and Dr. Jump, has made it possible for me to reach
the goal of doctoral degree. I would also like to express my special appreciation to Dr.
Jacob Krier who provide me the opportunity to work in his laboratory and provide the
major guidance on my research. I wish he could see the progress that he had helped me
to make today.

My appreciation also goes to my colleague and dear friend, Dr. Kalina Venkova
who kindly provided many valuable suggestions and comments on my dissertation work
and manuscript writing.

Finally, I would like to express my special appreciation to my daughter Lillian
Zhang, my husband, Dr. Hong Zhang and my father, Dr. Zhongxian Lin for their love and
supports. Their attitude toward knowledge and science have made the major inﬂuence on

my achievement today.

 

TABLE OF CONTENTS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

LIST OF TARIFQ ix
LIST OF FIGURES x
LIST OF ABBREVIATIONS xiv
SUMMARY OF SPECIFIC AIMS 1
GENERAL HISTORY OF AUTONOMIC NERVOUS SYSTEM 7
The autonomic nervous system 9
Autonomic ganglia 4
Innervation of the gastrointestinal tract 6
Parasympathetic innervation 6
Sympathetic innervation 7
Autonomic reﬂex of the gastrointestinal system 8
The defecation reﬂexes 9
CHAPTER 1.
MORPHOLOGY, ACTIVE AND PASSIVE PROPERTIES OF NEURONS IN
PELVIC PLEXUS GANGLIA 14
I’ L‘. .‘l 4' 14
Anatomy of pelvic plexus ganglia of male guinea-pig ............................. 14
Immunohistochemistry of male guinea-pig pelvic plexus ganglia ........... 17
Electronmicroscopy studies of male guinea-pig pelvic plexus ganglia... 1 7
Electrophysiological and pharmacological studies of pelvic
plexus ganglia 18
I 1 5L .1- 19
Electrophysiological procedures 19
Horseradish peroxidase (HRP) injection ‘ ’ '1 ' l9
Passive properties~ 70
Active properties 70
Sodium and calcium components of actionr ‘ " ’ 7i
Rpmlte 7 I
Morphological properties of neurons in pelvic plexus ganglia ................ 21

Passive and active electrophysiological properties and sodium,

 

calcium components of the action potentials of neurons in

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

guinea-pig pelvic plexus ganglia 77
Phasic and tonic properties of pelvic plexus ganglion neurons ............... 22
Sodium and calcium components of actionr ‘ " ' 76
I" . 24
CHAPTER 2.
SYNAPTIC TRANSMISSION IN PELVIC PLEXUS GANGLION NEURONS ........ 40
I A... J A.’ 40
I l ‘L J 41
Electrophysiological procedures 44
Chronic surgical procedures 45
Reclllte 45
Synaptic potentials and pharmacological studies 45
Pattern of synaptic , ‘ " ' 47
Convergent synaptic inputs from central and peripheral nerve trunks
to a single neuron in pelvic plexus ganglia 50
Origin of synaptic potentials evoked by electrical stimulation of
colonic-rectal nerves and urethra-urinary bladder nerves ................... 50
Conduction velocities of preganglionic ﬁbers in central nerve
trunks and ﬁbers in peripheral nerve trunks 57
Antidromic " , 57
I" ' 6'41
CHAPTER 3.
NEURONS IN PELVIC PLEXUS GANGLIA RECEIVE MECHANOSENSORY
INPUT FROM DISTAL COLON 68
' *« J .- 5::
Concept of neurons in prevertebral ganglia receiving
mechanosensory input ﬁom peripheral orgnnc 68
Properties of visceral mechanoreceptors in distal colon-rectum ........... 69
Classiﬁcation of mechanoreceptive primary aﬂerents in
colon-rertum 70
I I AL A 70
Measurement of intraluminal colonic pressure ...................................... 71
Electrophysiological procedures 71
Rpcnltc 72
I" ' 90
CHAPTER 4.

NEURONS IN PELVIC PLEXUS GANGLIA MEDIATE REFLEX
CONTRACTIONS BETWEEN SEGMENTS OF COLON AND BETWEEN

COLON AND URINARY BLADDER 94
1 ... .r a: 94

Sacral parasympathetic reﬂex, " , 94

vi

 

 

Peripheral reﬂexes mediated by prevertebral ganglia .......................... 98

 

 

 

 

 

 

 

 

 

 

 

 

Eﬂects of capsaicin on primary aﬂerentﬁberv 99
Neuropeptides in primary sensory neurnm~ 101
‘1 AL .I 103
Measurement of intraluminal colonic pressures and urinary
bladder pressure 103
Experiments to study reﬂexes between segments of distal colon and
between colon and urinary bladder 104
R PC] 1th 1 07
Effects of acute capsaicin ‘— ‘r ‘ 120
I" ' 122
CHAPTER 5.
INTERACTION OF INFLAMMATORY MEDIATORS WITH PELVIC
PARASYMPATHETIC NERVOUS SYSTFM 126
' ‘ J " 126
Inﬂammation of the gastrointestinal system 126
Actions of bradykinin on visceral aﬂerentﬁbers .................................... 129
Inﬂammatory mediator interleukin-I B 133
l l ‘L J 1 3 8

 

Experiments designed to study the eﬂects of braaykinin on
mechanosensitive input ﬁom distal colon to neurons in pelvic

 

 

 

 

 

 

 

 

 

 

 

 

plexus ganglia 138
Measurement of intraluminal colonic pressure 139
Electrophysiological procedures 139
Experiments designed to test the eﬂects of human recombinant

interleukin-1B on neurons in pelvic plexus gang/in 140

R 99] 1th l 42
Eﬂects of bradykinin on colon contraction and mechanoreceptor

mediated synaptic input to neurons in pelvic plexus ganglia ............... 142
Eﬁ’ects of human recombinant interleukin-1 B on membrane potential

and membrane input I ' ‘ 147
Depolarimtinn 147
Hyperpolarimtinn 147
Eﬂects of human recombinant interleukin-1B on synaptic

:. ' ' 154
Orthodromic actionr ‘ “ ’ 154
Eﬂects of human recombinant interleukin-1 B on acetylcholine

potentials mediated by nicotinic receptor 159
Specificity of the eﬂects human recombinant interleukin-1B .................. 160

I" ' 160
The eﬂect of bradykinin on mechanosensory input(s) to neurons in
pelvic plexus ganglia 160

 

The eﬁ'ects of human recombinant interleukin-1B on synaptic
transmission and on the active and passive properties of

vii

 

neurons in pelvic plexus ganglia

 

SUMMARY

 

SIGNIFICANCE OF THE STUDY

 

BIBLIOGRAPHY

 

viii

169

175

179

180

 

LIST OF TABLES

Table 1. Summary of morphological properties of neurons in pelvic plexus ganglia
of male guinea-rig" 75

Table 2. Electrophysiological properties of phasic and tonic neurons in pelvic
plexus ganglia of male guinea-pig" 7.8

 

Table 3. Distribution of single and multiple synaptic potentials of PG neurons in
animals with an intact nervous system and after degeneration of
preganglionic ﬁbers in hypogastric and pelvic nerves 55

 

Table 4. Electrophysiological properties of neurons in normal, HGN
transected and decentralized neurons in pelvic plexus ganglia of male

guinea-2‘2" 56

 

Table 5. Summary of mechanosensitive input(s) to neurons in pelvic plexus ganglia..89

 

LIST OF FIGURES
Figure 1. Parasympathetic and sympathetic innervation of the gastrointestinal tract....8

Figure 2. Pelvic-hypogastric plexus 15

 

Figure 3. Camera lucida drawings of horseradish peroxidase (HRP)-injected
neurons in pelvic plexus ganglia of male guinea-pig" 72

 

Figure 4. Responses to direct depolarizing current injection of neurons in
guinea-pig pelvic plexus ganglia 77

 

Figure 5. Sodium and calcium components of the action potential and spike
dependence of the afterspike hyperpolarization in pelvic plexus ganglion

 

 

 

 

 

 

30
Figure 6. Effects of tetraethylammonium, tetrodotoxin, low Ca2+ high Mg2+ and
co-conotoxin on action potentials recorded from pelvic plexus ganglion
37
Figure 7. Diagrammatic sketch of lumbar sympathetic and sacral
parasympathetic pathways to neurons in pelvic plexus ganglia and to distal
colon-rectum of male guinea-pig" 42
Figure 8. Synaptic responses of neurons in pelvic plexus ganglia during stimulation
of urethra-urinary bladder nerves, colonic-rectal nerves, pelvic nerves
and hypogastric nerves in normal Krebs solution and in Krebs-
hexamethonium ‘ ‘° 46
Figure 9. Synaptic inputs (a) and single, multiple synaptic inputs (b) to neurons in
pelvic plexus ganglia 48
Figure 10. Convergence of synaptic input(s) to a neuron in pelvic plexus ganglia
from central and peripheral nerve minke 51

Figure 11. Synaptic input to pelvic plexus ganglion neurons in animals with an
intact nervous system and alter degeneration of preganglionic ﬁbers
in hypogastric nerves and pelvic nerves 53

 

 

Figure 12. Calculated conduction velocities of preganglionic ﬁbers which
provide synaptic inputs to neurons in pelvic plexus ganglia 58

Figure 13. Antidromic action potential recorded from one neuron in pelvic
plexus ganglia 59

Figure 14. Antidromic response to neurons in pelvic plexus ganglia (a) and
calculated conduction velocities of nerve ﬁbers which provide antidromic
inputs to neurons in pelvic plexus ganglia (h) 61

 

Figure 15. Excitatory synaptic inputs to three neurons in pelvic plexus ganglia ......... 73

Figure 16. Effect of distension of distal colon on synaptic input to a neuron in
pelvic plexus ganglia 75

Figure 17. Effect of distal colon distension on initiating synaptic responses to a
quiescent neuron in pelvic plexus ganglia 77

 

Figure 18. Effect of hexarnethonium on synaptic inputs from distal colon to a
neuron in pelvic plexus ganglia 78

 

Figure 19. Effects of sectioning colonic-rectal nerves on synaptic inputs to a
neuron in pelvic plexus ganglia R7

 

Figure 20. Plot of relation between instantaneous action potential frequency and
action potential interval in neurons of pelvic plexus ganglia 84

 

 

Figure 21. Stimulus-response curves for distal colon ' r‘ s 86

Figure 22. Stimulus-response curves for two types of distal colon
' ‘ \ 512
Figure 23. Diagrammatic sketch of sacral parasympathetic central reﬂex
pathway to colon (A) and diagrammatic sketch of sacral
parasympathetic reﬂex pathway to colon and urinary bladder via pelvic
plexus ganglia (B) 96

 

Figure 24. Diagrammatic sketch of in vitro preparations for studying reﬂexes
between two segments of colon (A) and between the colon and urinary
bladder (B) 105

 

Figure 25. Contractile responses of distal colon segment during passive
distension of proximal colon segment and electrical stimulation of

A

colonic nerve ﬁbers attached to proximal colon g 108

xi

 

Figure 26. Contractile responses of urinary bladder during passive distension of
colon-rectum and electrical stimulation of colonic-rectal nerves ..................... 110

Figure 27. Contractile responses of colon-rectum during electrical stimulation of
urethra-urinary bladder nerves 1 12

 

Figure 28. Effects of electrical stimulation of colonic nerve ﬁbers originating
from proximal colon segment on contractile responses recorded from
distal colon c, ‘ 114

Figure 29. Effects of electrical stimulation of urethra-urinary bladder nerves
on contractile responses recorded from distal colon segment .......................... 116

Figure 30. Effects of electrical stimulation of colonic-rectal nerves on
contractile responses recorded from urinary bladder 118

 

Figure 31. Effects of capsaicin on urethra-urinary bladder nerves-evoked
contractile responses of guinea-pig distal colon segment when superfused
on pelvic plexus ganglia only, in vitrn 121

 

Figure 32. Effects of bradykinin on spontaneous colon contractions and
electrical activities of one neuron in pelvic plexus ganglia 143

 

Figure 33. Effect of colon-rectum distension on synaptic input(s) of a neuron in
pelvic plexus ganglia before (A), during (B, C) and after (D)
bradykinin superfused on colon-“mm" 145

 

Figure 34. Effects of human recombinant interleukin-1 B on resting membrane
potential, membrane input resistance and membrane current of two
neurons in pelvic plexus ganglia (depolarization effects) 148

 

Figure 35. Effects of human recombinant interleukin-1 B on resting membrane
‘ " ‘ 150
Figure 36. Effects of human recombinant interleukin-1 B on resting membrane
potential, membrane input resistance and membrane current of two
neurons in pelvic plexus ganglia (hyperpolarization effects) ........................... 152

Figure 37. Human recombinant interleukin-1 B induced outward currents at
different holding potentials (A) and voltage sensitivity of these currents
(B) recorded from a neuron in pelvic plexus ganglia 155

 

Figure 38. Effects of tetrodotoxin and m-conotoxin on action potentials evoked
by direct depolarizing current ' j " 156

 

 

Figure 39. Effect of human recombinant interleukin-1 B on amplitude of fast
excitatory postsynaptic , ‘ " ' 158

 

Figure 40. Effects of human recombinant interleukin-1 B on fast excitatory
postsynaptic potentials of one neuron in pelvic plexus ganglia ........................ 161

Figure 41. Effect of human recombinant interleukin-1B on fast acetylcholine
potentials of one neuron in pelvic plexus ganglia 163

Figure 42. Eﬁects of interleukin-1 receptor antagonist on pressure application of
human interleukin-1 B-induced membrane hyperpolarization in one neuron....165

Figure 43. Effects of human interleukin-1 B on synaptic potentials of one neuron
in pelvic plexus ganglia in the absence and presence of interleukin-1
receptor ‘ a ' ‘ 167

 

xiii

 

LIST OF ABBREVIATIONS

A: Angstrom unit,1x10’8cm.

Ach: acetylcholine.

4-AP: 4-arninopyridine

APs: action potentials.

ASH: aﬁerspike hyperpolarization
B: urinary bladder.

BK: bradykinin.

BSA: bovine serum albumin

C: coagulating gland.

C6: hexamethonium.

Ca2+: calcium ion.

CCK: cholecystokinin

CG: celiac ganglia.

CGRP: calcitonin gene related peptide.
Cl": chloride ion.

CM: circular muscle.

CN: colonic nerves.

CNS: central nervous system.

C02: carbon dioxide.

COL-RECN: colonic-rectal nerves.
cyclic GMP: guanosine 3’, 5’-cyclic monophosphate.
DAB: 3, 3’-diaminobenzidine.

xiv

 

DAG: diacylglycerol.

DRG: dorsal root ganglia.

EF S: electrical ﬁeld stimulation.

ET: endothelin.

f-EPSPs: fast excitatory postsynaptic potentials.
GABA: y-aminobutyric acid. '

H20: water.

HGN: hypogastric nerves.

hrIL-l B: human recombinant interleukin-1 B.
HRP: horseradish peroxidase.

HTM: high threshold mechanoreceptors.
IBD: inﬂammatory bowel disease.

IL-l: interleukin-l

IL-lra: interleukin-1 receptor antagonist.
IMG: inferior mesenteric ganglia.

K+z potassium ion.

KC]: potassium chloride.

KS: Krebs solution.

L: lumbar root.

LEM: leucocyte endogenous mediator.
LM: longitudinal muscle.

LTM: low threshold mechanoreceptors.
MCF: mononuclear cell factor.

Mg”: magnesium ion.

MP: myenteric plexus.

mV: millivolts.

nA: nanoArnperes.

 

Na+z sodium ion.

NAN C: nonadrenergic, noncholinergic neurons.
NE: norepinephrine

NKA: neurokinin A.

NKB: neurokinin B.

NOS: nitric oxide synthase.

NP: nitroprusside.

NPY: neuropeptide Y

02: oxygen.

P: prostate gland.

PELN: pelvic nerves.

PG: pelvic plexus ganglia.

PKC: protein kinase C.

rhTNF: recombinant human tumor necrosis factor.
S: sacral nerve.

SCG: superior cervical ganglia.

SMG: superior mesenteric ganglia.

SP: substance P.

SV: seminal vesicles.

TEA: tetraethylammonium.

Tks: tachykinins.

TNIB: tetramethyl benzidine dihydrochloride.
T'I'X: tetrodotoxin.

UUBN: urethra-urinary bladder nerves.

VD: vas deferens.

VIC: vasoactive intestinal contractor.

VIP: vasoactive intestinal polypeptide.

 

   
   
   
    
  
    
   

SUMMARY or SPECIFIC AIMS
Th overall ob}ecti\cs o!

' vitrn gumtwmq
l: \L tmuttL 41:“: MT

1. To determine the mu. '. my he _-_ - - Mu: -~

in; ..4 pmpcrtte‘ and the
m Ofﬁcdtm pulenlinl . n5 ncur-‘ns 1:! ml ~ it 1’.

“LL-.4

D. To "Italy last excitatory

it...

;}i..j,)(h 2.21.- m 1mm m print; plexus ganglion

I“. To determine whether neurons .. 7.34. alt-\u gsag‘wn mum.

Ill
“ .“.'\.
b

‘__ u l- *\,‘*J

..=.“. 213-6 -L’-5 Nag,“ w" 35—55%?!“9'5-21

-.:.v-. :'—\.' H: .1335" ‘ "_ l 3,, .-~ 5/» ".75;-

écaww ‘c‘t

‘-u I . c .1 *iv: -tat-nn are to telctmmc 1hr rule 0-”

 

SUMMARY OF SPECIFIC AINIS
The overall objectives of the present dissertation are to determine the role of
pelvic plexus ganglion neurons in the regulation of contractile activity of colon smooth
muscle. The experiments proposed will utilize a newly-designed in vitro guinea-pig
distal~ colon-pelvic ganglia preparation. The speciﬁc aims are:
I. To determine the morphology, the active and passive properties and the
component of action potentials of neurons in pelvic plexus ganglia.
II. To study fast excitatory synaptic transmission in pelvic plexus ganglion
neurons.
III. To determine whether neurons in pelvic plexus ganglia receive
mechanosensory input(s) from the distal colon.
IV. To test the hypothesis that neurons in pelvic plexus ganglia mediate reﬂex
contractions between segments of the colon and between the colon and urinary bladder.
V. To study the interaction of inﬂammatory mediators with pelvic
parasympathetic system.
i. To study the effect of bradykinin on mechanosensory input(s) to neurons in
pelvic plexus ganglia.
ii. To determine the effects of human recombinant interleukin-113 on synaptic

transmission and on the active and passive properties of neurons in pelvic plexus ganglia.

 

GENERAL HISTORY OF THE
AUTONOMIC NERVOUS SYSTEM
The autonomic nervous system

If any thing ail a man,

so that he does not perform his functions,

If he have a pain in his bowels

-for that is the seat of sympathy-

he forthwith sets about reforming his world.
Henry D. Thoreau
The Writings of Henry David Thoreau.
Boston: Houghton Mifﬂin, 1906.

The literary allusion of the bowels as the seat of sympathy, stated by the
American novelist Thoreau, underscores that innate sense each of us subconsciously
acknowledges of the reactivity of the gastrointestinal tract to rigors of life. Called
"sympathy", our reactions to our environment entail a complex nervous system capable
of responding acutely and instantaneously to particular circumstances that challenge us.
The search for this nervous system and the desire to understand its physiology have
constantly occupied that attention of mankind. The "sympathy" nervous system now is
called the autonomic nervous system. It is also called the visceral, vegetative, or
involuntary nervous system (Langley, 1921).

In the periphery, the autonomic nervous system consists of nerves, ganglia and
plexuses that innervate the heart, blood vessels, glands, other visceral organs and smooth
muscles. The autonomic nervous system is widely distributed throughout the body and

regulates functions, which occur without conscious control.

 

3

Aﬁerent ﬁbers from visceral structures are the ﬁrst link in the reﬂex arcs of the
autonomic nervous system. With certain exceptions, such as peripheral reﬂexes
mediated by the autonomic ganglia and local axon reﬂexes, most visceral reﬂexes are
mediated through the central nervous system (CNS). The afferent ﬁbers are, for the most
part, nonmyelinated and are carried into the cerebrospinal axis by the vagal, pelvic and
splanchnic and other autonomic nerves. The cell bodies of visceral afferent ﬁbers lie in
the dorsal root ganglia of the spinal nerves and in the sensory ganglia of certain cranial
nerves, such as the nodose ganglion of the vagus. Autonomic afferent nerves mediate
visceral sensation (including pain and referred pain) in vasomotor, respiratory and
viscerosomatic reﬂexes and in regulation of interrelated visceral activities. Neuroactive
peptides including substance P, vasoactive intestinal polypeptide (VIP), cholecystokinin
(CCK), somatostatin and calcitonin gene related peptide (CGRP) may play a role in
afferent neurotransmission.

The efferent nerves of the autonomic nervous system supply all innervated
structures of the body except skeletal muscle, which is served by somatic nerves. The
efferent part of the autonomic nervous system consists of two large divisions: (1) the
sympathetic or thoracolumbar outﬂow and (2) the parasympathetic or craniosacral
outﬂow. The principal neurotransmitter of all preganglionic autonomic ﬁbers and all
postganglionic parasympathetic ﬁbers is acetylcholine (Ach). The adrenergic ﬁbers
comprise the majority of the postganglionic sympathetic ﬁbers and the principal
neurotransmitter is norepinephrine (NE). However, ATP, a purine nucleotide and
neuropeptide Y (NPY), a neuropeptide are also found and can be released as
cotransmitters from postganglionic sympathetic terminals.

The sympathetic system is distributed to effectors throughout the body, whereas
the distribution of the parasympathetic system is much more limited. Preganglionic
sympathetic ﬁbers may traverse a considerable distance of the sympathetic chain and

pass through several ganglia before they ﬁnally synapse with postganglionic neurons. In

 

4

contrast, the parasympathetic system has its terminal ganglia very near to or within the
organs innervated and thus can be more circumscribed in its inﬂuences.

The integrating action of the autonomic nervous system is of vital importance for
the well-being of the organism. In general, the autonomic nervous system regulates the
activities of structures that are not under voluntary control and that, as a rule, function
below the level of consciousness. Thus, respiration, circulation, digestion, body
temperature, metabolism, sweating and the secretion of certain endocrine glands are

regulated, in part or entirely, by the autonomic nervous system (Skok, 1973).

Autonomic Ganglia

As early as the second century, the Greek physician Galen (cA. a. d. 129-199) had
identiﬁed nervous elements extrinsic to the gastrointestinal tract that exerted this
particular inﬂuence “sympathy” on the abdominal viscera. Galen's account of a visceral
nervous system described ﬁne ﬁlaments coming off the branches of what is now
identiﬁed as paravertebral sympathetic chains. Several groups of “swellings” were also
observed in the abdominal cavity that are now called autonomic ganglia. The autonomic
ganglia are aggregates of nerve cells with glial cells and connective tissue surrounding
them. The extracellular space around the ganglion cells is protected by a blood-ganglia
barrier that exists in the major ganglionic capillary system. The ganglia form either
separate bodies connected by nerves with each other, with the central nervous system and
with peripheral organs (extramural ganglia), or they may form plexuses located within
the walls of visceral organs (intramural ganglia). All ganglia receive preganglionic ﬁbers
from the central nervous system (CNS) and send postganglionic ﬁbers to effectors such
as smooth muscles and glands. The sympathetic ganglia receive preganglionic ﬁbers
only ﬁom the thoracic and lumbar segments of the spinal cord while the parasympathetic
ganglia receive preganglionic ﬁbers from the brain stem and from sacral segments of the
spinal cord (Szurszewski and King, 1989).

 

5

The cell bodies of sympathetic preganglionic ﬁbers are located in the lateral
column of the spinal cord in the nucleus intermediolateralis, which spreads from the ﬁrst
thoracic segment to the third lumbar segment. The cell bodies of parasympathetic
preganglionic ﬁbers are located in the nuclei of III, IV, IX and X cranial nerves and in
sacral segments of the spinal cord (Szurszewski and King, 1989).

Three types of pathways pass through the ganglia. The most common pathway is
“centrifugal pathway”, present in all autonomic ganglia of vertebrates and invertebrates.
In autonomic ganglia of this pathway, preganglionic ﬁbers transmit impulses to many
neurons located either in the same ganglia or in several ganglia, and then the impulses
spread from postganglionic ﬁbers to the peripheral organs. The second type of pathway
is the “peripheral reﬂex pathway”, present in the intramural ganglia of the
gastrointestinal tract, in the inferior mesenteric ganglia and in the celiac-superior
mesenteric plexus. The impulses arising in the visceral receptors, whose cell bodies are
located in the visceral organs, reach the ganglia through afferent ﬁbers and may be
transmitted synaptically to ganglion neurons sending their axons to peripheral organs.
There is a large number of these neurons which receive both preganglionic and peripheral
reﬂex innervation, and integration between these pathways occurs in the ganglia. The
third type of pathway is “centripetal pathway”. Traditionally, this pathway consists of
mostly afferent ﬁbers of central origin crossing the ganglia without making synaptic
contacts within it. However, in the late 1970’s, the presence of synaptic contact of this
pathway with neurons in the ganglia has been suggested in the inferior mesenteric
ganglia (Skok, 1973).

Two lines can be deﬁnitely traced in the development of autonomic ganglia from
lower forms of vertebrates to higher forms. While unipolar neurons are the only type in
amphibian sympathetic ganglia, multipolar neurons are the typical sympathetic ganglion
neurons in birds and mammals. The number of nerve cells in the ganglia increases ﬁorn

lower forms of vertebrates to higher forms of vertebrates and the increase in number of

 

6

neurons is greater than the increase in number of preganglionic ﬁbers. These processes
are presumably related to more complicated organimtion of the ganglia in higher

vertebrates (Skok, 1973).

Innervation of the Gastrointestinal Tract

Gastrointestinal functions are controlled by a hierarchy of hurnoral, neuronal and
myogenic mechanisms. The neural control of the gastrointestinal tract is provided by the
parasympathetic and the sympathetic divisions of autonomic nervous system and enteric
nervous system. The subdivision is based on the location of the ganglia and connections
with the central nervous system. The parasympathetic and sympathetic nerves supply the
extrinsic innervation of the gut. The enteric nervous system comprises the intrinsic
innervation of the gut.

Parasympathetic innervation The preganglionic neuronal cell bodies of the
parasympathetic division are located in the medulla of the brain and in the sacral region
of the spinal cord. Peripheral ganglia of the parasympathetic division are positioned
within or at the boundaries of the organ system they innervate (Fig. 1a). The
preganglionic parasympathetic neurons are cholinergic and their axons terminate on
neurons in the intramural plexuses and/or neurons in parasympathetic ganglia. The

rpostganglionic parasympathetic neurons and enteric neurons provide the ﬁnal common
pathway to the muscle for either long reﬂexes via the parasympathetic nerves and enteric
neurons or short reﬂexes within the plexus via enteric neurons, respectively (Langley,
1921).

The efferent parasympathetic innervation of the upper gastrointestinal tract is
provided by the nucleus ambiguous and the dorsal motor nucleus of the vagus in the
medulla oblongata. The vagus nerve provides parasympathetic innervation to the upper
gmtrointesu'nal tract up to the mid-transverse colon. The vagus nerves branch soon after

entering the abdomen, with major trunks supplying the stomach and liver. The nerves

7

destined for more distal regions travel predominantly in the celiac branch of the vagus
nerves and reach the intestines via the paravascular nerve bundles. Only 10% of the
abdominal vagal ﬁbers are efferent. The remaining 90% are afferent ﬁbers whose cells
bodies lie in the nodose and jugular ganglia and transmit sensory information to the brain
(Grundy, 1989). In the lower areas of the gut, the parasympathetic outﬂow originates
from the anterior columns of the sacral spinal cord segments 2, 3 and 4. In most species,
these nerves join with sympathetic and afferent ﬁbers to form the pelvic plexus in which
the two groups of nerve ﬁbers cannot be separated. The postganglionic ﬁbers originating
from pelvic plexus ganglia innervate the anorectal area, the descending colon and the leﬂ
half of the transverse colon.

Sympathetic innervation The cell bodies of preganglionic neurons of the
sympathetic division are located in the interrnediolateral cell columns of the spinal cord
between the 5th thoracic and 3rd (or 4th in man) lumbar segments. Peripheral ganglia of
the sympathetic division are organized in chains of paravertebral ganglia positioned along
either side of the spinal cord and prevertebral (celiac, superior, and inferior mesenteric)
ganglia within the abdomen (Fig. 1b). The preganglionic neurons in these pathways are
cholinergic and they form nicotinic synapses on post-ganglionic neurons, which are
adrenergic. Preganglionic ﬁbers to the prevertebral ganglia pass through the
paravertebral sympathetic chain without synapsing. They form the thoracic and lumbar
splanchnic nerves which terminate in the prevertebral ganglia. Bundles of postganglionic
sympathetic nerves and some preganglionic parasympathetic ﬁbers from the celiac branch
of the vagus nerve form paravascular nerves which accompany the arteries to the effector
organ. The levels of regions of the bowel determine the particular ganglia that supply
them. Most efferent adrenergic ﬁbers innervating the gastrointestinal tract terminate on
the vasculature and on enteric neurons (Norberg, 1964). A few adrenergic ﬁbers extend
into the mucosal villi and around the mucosal glands. Except for sphincteric regions in

some species, direct innervation of the smooth muscle is sparse or

 

;
l
l
:

   

 
   
 

 

 

(‘3’ Mama :2: ~.
7%" ée;~’a'-=tw~.v~tpwmv AM “$350,? ‘.

Figure l. a: Parasympathetic innervation of the gastrointestinal tract. The ﬁgure shows
only the preganglionic efferent supply. The postganglionic parasympathetic neurons (not
shown) are located intramurally or close to the innervated organ. b: Sympathetic
innervation of the gastrointestinal tract. Preganglionic neurons are represented by solid
lines; post-ganglionic sympathetic neurons are shown as broken lines. SCG=superior
cervical ganglia; CG=celiac ganglia; SMG=superior mesenteric ganglia; IMG=inferior

mesenteric ganglia. (Aﬁer Jan Tack, 1992)

 

 

 

" ii i

 

 

 

 

00m TV

 

 

 

9,

is

w>¢wz 0.)..wc
‘

IIII»

a

I

I

I

I

a!

.U I
I

I

V I
20.523 II
3:30 I
I
I
I
I
I“.

 

\

w>¢u2 m30<>

 

10

absent (Jacobowitz, 1965; Gillespie and Maxwell, 1971). Sympathetic nerves bundles
also contain large numbers of afferent ﬁbers. Sensory information travels to the
prevertebral ganglia and to the CNS in ﬁbers with cell bodies in either the gut wall or in

dorsal root ganglia of the spinal cord.

Autonomic Reﬂex of the Gastrointestinal System

An extensive sensory innervation of the gastrointestinal tract is evident from a
consideration of the numerous reﬂexes that coordinate digestive function. Secretion,
motility, blood ﬂow, and also possibly absorption throughout the length of the
gastrointestinal tract are modiﬁed according to the needs of digestion, as determined by
the volume and nature of luminal contents. Gastric emptying, for example, is regulated to
provide the controlled delivery of chyme to the intestines and depends on excitatory and
inhibitory mechanisms activated by the physical and chemical composition of
gastrointestinal contents (Cooke, 1975). Hormones as well as neural reﬂexes are
responsible for this high degree of control. Neural reﬂexes can occur at several different
levels within a complex hierarchy of regulatory mechanism. Reﬂexes can occur entirely
within the intramural plexuses, an independent nervous system which contain all the
reﬂex elements, including receptor mechanisms, afferents, efferents and the station for
integrating the information (Woods, 1981). Other type of reﬂex with sensory neuron cell
bodies in the enteric nervous system, send afferent projections to the prevertebral ganglia
and make connections with postganglionic sympathetic ganglionic neurons (celiac
ganglia and inferior mesenteric ganglia) (Crowcroft, Holman and Szurszewski, 1971;
Weems and Szurszewski, 1977; Kreulen and Szurszewski, 1979a; Kreulen, Muir and
Szurszewski, 1983) and parasympathetic ganglionic neurons (pelvic ganglia) (Lin and
Krier, 1993). The efferents from these autonomic ganglion neurons project back to the
wall of the gastrointestinal tract. The establishment of this reﬂex pathway has

consequently changed our view of the prevertebral ganglia from that of mere relay

 

11

stations on the sympathetic and parasympathetic pathways to integrative centers
controlling the abdominal viscera. The decentralized gut is therefore capable of complex
reactions mediated through peripheral reﬂexes (Kreulen and Szurszewski, 1979; King and
Szurszewski, 1984; Lin and Krier, 1994), which accounts for the often minor and
transient consequences of extrinsic denervation, because the peripheral controls are able
to adapt to the loss of the extrinsic supply (Grundy, 1981). Finally, the third type of
reﬂexes with sensory neuron cell bodies in external ganglia and connect the abdominal
viscera with the CNS. Neural activity in these sensory ﬁbers rarely reaches
consciousness; it is more involved in regulatory processes. However, sensations do arise
ﬁ'om the gastrointestinal tract. Under normal circumstances, these sensations are
generally vague and poorly localized, but can become intense under pathological
conditions. The afferent ﬁbers largely follow the same route to the central nervous
system as the efferent ﬁbers leaving it; thus they can be divided broadly into
parasympathetic and sympathetic afferent ﬁbers. The cell bodies of vagal afferents lie in
the superior and inferior (jugular and nodose) vagal ganglia and input to the brain stem.
Afferent ﬁbers in the splanchnic and pelvic nerves have their cell bodies in the dorsal root
ganglia of the spinal cord and input to the cord through the dorsal roots (Grundy and
Scratcherd, 1989).

The Defecation Reﬂexes

Defecation occurs by coordinated contractions and relaxations of smooth and
skeletal muscles of the colon, rectum, anal sphincters and pelvic floor. It is a complex act
involving the integration of somatic and autonomic reﬂex mechanisms and is dependent
on the participation of neural populations located within the wall of the colon and rectum
(enteric nervous system) and neuronal populations located at various levels of the

neuroaxis (including the spinal cord, brain stem, telencephalon and diencephalon).

 

12

Ordinarily, defecation is initiated by defecation reﬂexes. One of these reﬂexes is
an intrinsic reﬂex mediated by the local enteric nervous system. This reﬂex is fortiﬁed
by another type of defecation reﬂex, a parasympathetic defecation reﬂex that involves the
sacral segments of the spinal cord. However, based upon the studies that will be
described in the present dissertation, reﬂexes mediated by parasympathetic ganglia,
pelvic plexus ganglia, may also be part of the parasympathetic defecation reﬂex. Despite
the defecation reﬂexes, other effects are also necessary before actual defection occurs.

It is generally considered that the primary stimulus of defecation is distension of
the rectum. Receptors located in the wall of the rectum are activated, resulting in reﬂex
relaxation of the internal anal sphincter and simultaneous contractions of the external anal
sphincter. Rectal distension also generates sensory information that is transmitted via
sacral afferent ﬁbers and ascending afferent pathways in the intermediolateral ﬁmiculus
of the spinal cord (Nathan and Smith, 1953) to cortical and diencephalic centers.
Telencephalic and/or diencephalic centers provide descending input to somatic efferents
to elicit appropriate postural adjustments. If the urge to defecate is suppressed,
descending inputs to motoneurons in the sacral spinal cord mediate voluntary
contractions of the external anal sphincter, puborectalis, and levator ani. Tension in the
rectal wall decreases the stimulus for defecation. In contrast, during defecation, inputs to
somatic efferents effectively increase intrathoracic and abdominal pressure, which assist
evacuation. Increases in intrathoracic and intra-abdominal pressure are mediated by a
descent of the diaphragm and closure of the glottis. The squatting position also aids in
the increase of intra-abdominal pressure.

Descending projections from the brain stem and hypothalamus to preganglionic
neurons of the sacral spinal cord facilitate the principal reﬂexes that underlie defecation.
These include activation of parasympathetic efferent ﬁbers in the distal colon and rectum
and inhibition of the excitatory input to the external anal sphincter and pelvic ﬂoor

muscle. Activation of the sacral parasympathetic outﬂow facilitates excitatory reﬂexes,

 

13
which results in smooth muscle contraction and the propulsion of fecal contents.
Inhibition of motor input to the external anal sphincter and pelvic ﬂoor results in

relaxation of the skeletal muscle.

 

CHAPTER 1
MORPHOLOGY AND ACTIVE AND PASSIVE PROPERTIES OF
NEURONS IN PELVIC PLEXUS GANGLIA.

Introduction

Begining with the pioneering work of Langley and Anderson (1896a, b), it has
been shown that preganglionic ﬁbers in pelvic nerves (arising from the sacral spinal cord)
and hypogastric nerves (arising from the lumbar spinal cord) form synapses with neurons
in pelvic plexus ganglia (Blackman, Crowcroﬁ, Devine, Holman and Yonemura, 1969;
Crowcroﬁ and Szurszewski, 1971; de Groat and Krier, 1978). These neurons are the
origin of post ganglionic parasympathetic ﬁbers that innervate vascular and visceral
smooth muscle of colon, anal canal, urethra, urinary bladder and reproductive organs
(Langley and Anderson, 1896a,b,c). Thus neurons in the pelvic ganglia provide a site for
integrating synaptic inputs arising from preganglionic neurons in the sacral and lumbar
regions of the spinal cord. Their postganglionic ﬁbers innervate effector structures in the
urethra-urinary bladder, colon-rectum and reproductive organs. Parasympathetic nerve
activity from the postganglionic ﬁbers originated from neurons in the pelvic plexus
ganglia is considered to play an important role in the control of smooth muscle tone of
distal colon and urinary bladder.

Anatomy of pelvic plexus ganglia of male guinea-pig. In the male guinea-pig,
the pelvic plexuses are located on both sides of the urinary bladder neck and the distal
colon. The pelvic plexus is composed of a dense network of blood vessels, nerve
branches and a prominent cluster of nerve cells at the junction of the pelvic and
hypogastric nerves (Fig. 2). This nerve cell cluster is named the pelvic plexus ganglia

(Wozniak and Skowronska, 1966; Crowcroft and Szurszewski, 1971). Small isolated
14

 

  
    
   
 

 

at.“ man .a :bnslg 3:11»:ng *-
a“ me with m 1 .. e not: mdmul .. _
lb or. - maniacal :d .weiv
MM" ,

"he .

.zux-rlq unusual

a
, mi 8372‘)“ oi'tisq ?

" la‘l .S awgi"

  

IEBQquﬁ mum

 

  

..(lV 3319mm

.9

 

15

Figure 2. Pelvic-hypogastric plexus. Ganglion cells are scattered throughout the region
where hypogastric and pelvic nerves intersect. HGN, hypogastric nerves; SV, seminal
vesicles; VD., vas deferens; P, prostate gland; C, coagulating gland; B, urinary bladder;
pelv. N., pelvic nerves; L, lumbar root; and S, sacral nerve. (After Szurszewski and King,

1989). a: dorsal ventral view. b: lateral view.

16

..o

 

I

 

17

clusters of cells are also found for a short distance along hypogastric nerves and pelvic
nerves. The former ones now are named hypogastric ganglia while the later ones are
considered to belong to the pelvic plexus ganglia (Crowcroft and Szurszewski, 1971).

Immunohistochemisty of male guinea-pig pelvic plexus ganglia. In male
guinea-pigs, there are three types of ganglion cells, distinguished by their
immunohistochemical properties. The majority of cells (70%), which do not contain
granular vesicles, are considered cholinergic neurons. The other two types containing
either large granular vesicles or granular vesicles of variable size are considered to be
adrenergic or peptide containing neurons. Ninety ﬁve % of the total axon proﬁles within
the ganglia are cholinergic, the remaining 5% are adrenergic. Ninety-nine % of the
synapses are cholinergic whereas 1% are adrenergic (Yokota and Bumstock, 1983).
These studies indicate that the majority of neurons and the majority of axon proﬁles in
pelvic ganglia are cholinergic.

Electron microscopy studies of male guinea-pig pelvic plexus ganglia. When
viewed under the low power electron microscope, each ganglion contains a variable but
small number of ganglion cells. The number was estimated to be about 200 in each
ganglia. Ganglion cell bodies, about 20-30 um in diameter, were tightly packed together
within groups with larger tissue spaces between the groups. Within these larger spaces,
there were ﬁber tracts containing both myelinated and unmyelinated axons (0.5-3.0 pm in
diameter) (Blackman, Crowcroﬁ, Devine, Holman and Yonemura, 1969).

The narrower spaces between individual ganglion cells within a group were
occupied by the capsule cells surrounding each cell body and by small bundles of nerve
ﬁbers ensheathed by Schwann cells. Nerve ﬁbers were classiﬁed as either preganglionic
axons or ganglion cell processes. Clusters of ﬁne ganglion cell processes (0.1-0.3 um)
were also seen within the capsule. At synapses, the pre- and postsynaptic membranes
were thickened and spaced 100-150 A apart for distances up to 5000 A (Blackman,
Crowcroﬁ, Devine, Holman and Yonemura, 1969).

 

18

Electrophysiological and pharmacological studies of pelvic ganglia. Due to the
close location of the pelvic plexus ganglia to the organs, studies of the
electrophysiological properties of neurons in guinea-pig pelvic plexus ganglia have not
been done since the late 19603 (Blackman, Crowcroﬁ, Devine, Holman and Yonemura,
1969) and early 19705 (Crowcroﬁ and Szurszewski, 1971). But even aﬁer these initial
studies, the electrophysiological and pharmacological properties of neurons in pelvic
plexus ganglia remained unclear for another decade. Recently, Krier and colleagues
described the electrophysiological properties of parasympathetic neurons in cat and rabbit
pelvic ganglia and obtained qualitative data regarding voltage-dependent calcium and
potassium currents (Nishimura, Krier and Akasu, 1993). Their studies revealed the
effects of opioid and vasoactive peptides on synaptic transmission and
electrophysiological properties of neurons in pelvic plexus ganglia. Enkephalins were
found to inhibit nicotinic neurotransmission in cat colonic ganglion cells by speciﬁc
interactions with delta-type opioid receptors (Kennedy and Krier, 1987). The vasoactive
peptides, endothelin (ET) and vasoactive intestinal contractor (VIC) modulated the sacral
parasympathetic outﬂow to the colon by affecting neurons in the parasympathetic
ganglia. In summary, ET and VIC inhibited nicotinic synaptic transmission mediated by
nicotinic Ach receptors (Nishimura, Krier and Akasu, 1991), depolarized and then
hyperpolarized the membrane potential associated with a decrease and an increase in
membrane input resistance of neurons (Nishimura, Akasu and Krier, 1993) and reduced
or facilitated voltage-dependent calcium currents (Nishimura, Krier and Akasu, 1991).
ET and VIC also caused an inward followed by an outward receptor-operated current in
parasympathetic neurons. The currents were mimicked by exogenous guanosine 3’, 5’-
cyclic monophosphate (cGMP). Both ET and VIC also caused contractions of the colon
circular muscle (N ishimura, Akasu and Krier, 1992).

The present studies are focused on the role that pelvic plexus ganglion neurons

play in regulation of contractile activity of the colon and are part of a complex study of

 

19

the function of pelvic plexus ganglia. The studies described in the present chapter were
designed to determine the morphology, the active and passive properties and the
components of action potentials of neurons in the male guinea-pig pelvic plexus ganglia.
These data will be compared to those obtained in other autonomic ganglia in different
mammalian species. These studies served basically as a background investigation for the

following studies.

Methods

Experiments were performed on tissues from adult male guinea-pigs (200-500
gm), euthanized by exsanguination following carbon dioxide gas-induced
anesthetization. Pelvic plexus ganglia (PG) were dissected in situ and placed in a single
compartment organ bath. In all preparations, PG were pinned to the ﬂoor of a Sylgard-
lined organ bath. The organ bath was superfused with a modiﬁed Krebs solution (2-3
ml/min) containing (mM): Na“, 137.4; K+, 5.9; Ca2+, 2.5; Mg2+, 1.2; Cl', 134; HCO3',
15.5; H2PO4', 1.2; glucose, 11.5. The solution was preheated to 37-38 °C at the
recording site and equilibrated with 95% 02 - 5% C02 gas mixture. Low Ca2+ high
Mg2+ Krebs solutions contained 0.05 mM Ca2+ and 30 mM Mg”.

Electrophysiological procedures. Intracellular potentials were recorded with
microelectrodes ﬁlled with 3 M-KCl and a tip resistance ranging from 60-110 MD.
Membrane potential was recorded using an ampliﬁer with an active bridge circuit that
allowed current injection into neurons through the recording microelec’aode. Signals
from the microelectrode were displayed on an oscilloscope with digitized memory and
recorded on a pen-writing chart recorder. Data were stored on a videocassette recorder
and the output (individual electronic potentials and action potentials) was plotted with an
X-Y plotter.

Horseradish peroxidase (HRP) injection technique. The horseradish peroxidase
(HRP)-injcction technique was used to study the morphological properties such as soma

 

20

diameters and number of processes of pelvic plexus ganglion neurons. Individual
neurons in pelvic plexus ganglia were ﬁlled by intracellular injection of horseradish
peroxidase through the recording microelectrode (5% HRP (Sigma) in 0.5 M KCl/Iris
buffer solution). Microelectrodes had a tip resistance of 70-140 MQ. Horseradish
peroxidase was injected into neurons by depolarizing currents (1 - 2 nA, 300 ms duration
at 0.5-1 Hz for periods of approximately 15-25 min). Following a 2-hr time period ﬁom
the time of HRP injection into the last cell studied, ganglia were ﬁxed in a 1% (w/v)
glutaraldehyde/ 1% (w/v) paraformaldehyde phosphate-buffered ﬁxative for 45 min and
stored in ice-cold ﬁltered 0.05 M phosphate buffer at pH 7.2-7.4. Fixation and storage
were carried out at 3-5°C.

After ﬁxation, all ganglia to be sectioned was ﬁrst cryoprotected by sequential
equilibration in 10%, 20% and 30% sucrose buffer. Ganglia then were sectioned
horizontally at 40 um on a freezing microtome and mounted on gel coated slides. HRP-
labeled neurons were visualized by incubating the tissue with tetrarnethyl benzidine
dihydrochloride (TMB) for 12 to 18 hr at 4°C. This was followed by incubating the
tissue in 3,3'- diarninobenzidine (DAB) for 5 min at room temperature (22-25°C)
(McRorie, Krier and Adams, 1991). The tissue was rinsed in PBS and dehydrated in
serial dilutions of ethanol followed by xylene. All tissue was then coverslipped in
Permount and allowed to dry for 24 hr. Sections were viewed under light- and dark—ﬁeld
microscopy at magniﬁcations of 100 and 400 X.

Passive properties. The time constant was measured as the time for electrotonic
potential to reach 1-1/e of the maximum. Input resistance was calculated as the ratio of
the amplitude of the steady-state electrotonic potential (mV) and the magnitude of the
hyperpolarizing current pulse (nA) according to Ohm’s law. 5

Active properties. The characteristics of the action potentials, overshoot (mV),
maximum rate of rise (Vmax, V/sec), threshold potential (mV) and action potential

duration at 50% of repolarization (ms), afterspike hyperpolarizing potential duration

 

21

(ASH) at 90% (ms) were calculated by a computer program after analog to digital
conversion. Action potentials were elicited from cells in response to the injection of
depolarizing current pulses across the cell membrane (10 ms pulse duration) (0.1-1.5 nA
current magnitude).

Sodium and calcium componenm of action potentials. Action potentials were
elicited by direct depolarizing current injection (25-30 ms duration) (0.05-1.0 nA).
Tetrodotoxin (TTX, 0.5-1.0 uM), co-conotoxin GVIA (300-700 nM), dihydropyridine
derivative nifedipine hydrochloride (5-15 uM), substitution of normal Krebs solution to
low Ca2+ (0.05 mM) high Mg2+ (30 mM) Krebs solutions and tetraethylammoniurn
bromide (TEA, 30-60 mM) were used. Drugs were dissolved directly into Krebs solution
and were superfused onto the ganglia using a three way stopcock.

Drugs used were horseradish peroxidase (HRP, Sigma) co-conotoxin GVIA
(Peninsula Laboratories), tetrodotoxin (TTX, Sigma), tetraethylammonium bromide
(TEA, Sigma), nifedipine or nicardipine hydrochloride (Sigma).

Data were expressed as mean i SEM. Students t test was used for paired

comparison and p<0.05 was considered as signiﬁcant results.

Results
Morphological properties of neurons in pelvic plexus ganglia.

A total of 60 neurons from nine male guinea-pig pelvic plexus ganglia were ﬁlled
by intracellular injection of HRP that appeared in one or more serial sections without
apparent damage. These neurons were divided into two types according to
morphological criteria. One type had a single long process from the soma and was
classiﬁed as monopolar neurons (33.0%, 20 of 60) (Fig. 3A—F) with a mean soma
diameter of 37.3 :t 0.8 um (ranged from 26.0-53.0 um) and mean total length of the
process of 401.0 :h 61.0 um (mean :1: SEM, ranged from 55.0-990.0 um). The second
type (67.0%, 40 of 60) had multiple short, broad or ﬁne processes which radiated from

 

22

the soma and one long process (Fig. 3G-K). The number of processes in this type of
neuron varied from two to eight. Thirty-ﬁve % (14 of 40) of these neurons had two
processes and were classiﬁed as bipolar neurons. Sixty-ﬁve % (26 of 40) of the neurons
had more than two processes and were classiﬁed as multipolar neurons. These neurons
had soma diameter of 40.2 :1: 1.2 um (mean :t SEM, ranged from 29.0- 58.0 pm). The
mean total length of long process was 373.0 d: 32.5 um (mean i SEM, ranged 48.0-570.0
um) and the mean total length of shorter processes was 171.0 i 48.4 um (mean i SEM,
ranged from 16.0-564.0 pm). These data are summarized in Table 1 and show that the
majority of neurons in pelvic plexus ganglia of male guinea-pigs were multipolar and

bipolar neurons.

Passive and active electrophysiological properties and sodium, calcium components of
action potentials of neurons in guinea-pig pelvic plexus ganglia.

The passive and active electrophysiological properties of neurons in pelvic plexus
ganglia were tested in 89 neurons from isolated pelvic plexus ganglia of 30 guinea-pigs.
These neurons were not spontaneously active and had a resting membrane potential of -
56.3 d: 0.3 mV (-42 - - 82 mV, n=89); membrane input resistance of 51.4 :t 2.5 M!) (21.4
- 121.4 MQ, n=82); membrane time constant of 6.5 i 0.3 ms (3 - 18 ms, n=76) and
amplitude of action potentials of 73.0 i 0.5 mV (54 -112 mV, n=79).

Phasic and tonic properties of pelvic plexus ganglion neurons. Intracellular
injection of suprathreshold depolarizing currents (5 sec., 0.3-0.5 Hz) elicited two types of
responses. In one type of neuron, action potentials occurred for the duration of the
constant depolarizing pulse while the second type ﬁred a single (4 of 38, 11%) or a burst
(34 of 38, 89%) of action potentials at the onset of the depolarizing current. The response
of these neurons was referred to as "tonic discharging" and "phasic discharging",
respectively (Fig. 4). Tonic discharging neurons represented 57% (51 of 89) of neurons

tested, whereas, the remainder represented phasic discharging neurons. The

 

23

Figure 3. Camera lucida drawings of horseradish peroxidase (HRP)-injected neurons in
pelvic plexus ganglia of male guinea-pigs. A-F, monopolar neurons with one long
process. G-K, multipolar neurons with one long process and two or more shorter

processes.

25

.522 no Sum Esqsm ..2.m.m

 

 

 

 

 

 

 

 

 

 

 

amiéusws £3.53
£355 max»: 33mm 5 Eegaezcse awfumei Bummﬁgﬁegxe Egan. .u «55,—.

3:. H 3: 0% H 38 S H «.9. so no as $68 m s a 582::
93 H ogov md H msm So .8 ed $3M ﬂ 38802
.2.m.m H :32 .2.m.m H :82 .2.m.m H :82 8253me mommooocm 8882
o c3852 Mo 89¢.
83085 team @8885 was A; .9255 «Sow
E: K $80ch .0 gang. E

 

26

passive and active electrical properties of phasic and tonic neurons are listed in Table 2.
From HRP-injection experiments, 37% (43 of 116) of neurons injected were phasic
discharging neurons while the majority (63%, 73 of 116) of neurons were tonic ﬁring
neurons. The tonic and phasic neurons showed similar morphology (size of soma, length
and number of processes). The data suggest that the majority of neurons in pelvic plexus
ganglia were tonic neurons. There was no statistically signiﬁcant difference between
these two types of neurons in either passive electrophysiological properties or
morphology. However, the value of after spike hyperpolarization was marginally
signiﬁcantly different between tonic and phasic neurons.

Sodium and calcium components of action potentials. Action potentials that
were evoked by direct depolarizing current injection (15-30 ms duration, 0.05 - 0.6 nA)
were abolished by superfusion of the ganglia with a TTX-containing Krebs solution (0.5-
1.0 uM) (n=6) (Fig. 5). In the presence of TTX, increasing the strength of the
depolarizing current elicited action potentials that had a lower rate of rise and amplitude
than normal action potentials. Superfusion of ganglia with a low Ca2+ (0.05 mM), high-
Mg2+ (30 mM) Krebs solution containing TTX (1 uM) (Fig. 5) or a Krebs solution
containing TTX (1 uM) and m-conotoxin (700 nM) (data not shown) reversibly abolished
both the action potential and the subsequent aﬁerspike hyperpolarization (n=5).

Experiments were conducted to study the calcium component of the action
potentials when potassium currents were blocked by external application of TBA (60
mM) (Fig. 6). Under these conditions, the duration of spike (TEA spike) and aﬁerspike
hyperpolarization were prolonged. The action potential duration at 50% repolarization
before TEA was 2.1 at 0.2 ms and duration at 50% repolarization of TEA spike was 4.9 i
0.3 ms (n=4). Addition of TTX (1 uM) at this time blocked only the initial peak of the
TEA spikes. The plateau of TEA spikes was unchanged. Superfusion of a low Ca2+,
high Mg2+ Krebs solution (n=3) or m-conotoxin (700 nM, n=3) abolished TTX-resistant
TEA spikes (Fig. 6), suggesting that the charge carrier of the TTX-resistant TEA spikes

27

.bo>aoo%o._ canon—ma 23:; can 038 “m 98 < £2wa .3on 320m

298% 5 88:0: 9.3 «o 30:83 .08 m .<= $6-53 .8383 “go wag—09v 88:. 8 mom—8&3— .v £53m

 

 

 

 

 

 

 

 

.8: .8:
$2; 11. ..l .422; l ..I
8- .. I. an- . ..\
E m a. i xii; 4
o

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

28

actuate—gun»: commas?
5?. ”scene? 8250 maﬁa—cave 82% .3 3396 3:98 cocoa .m< ”2mm H 28:. 0.3 82.3

 

 

no a «.9: nd a 32 a5 8:23 $8 mm<
3 H on 3 a 2: 9.5 83%;... :2
as a a: E a «.2 9.5 .2 5 233 9 383a BEE
35 a M: .c :6 a mm... 25 .2 5 233 9 38:0 22.85.
...c a we 3. a 3 as e 2588 25 053502
an H Sn 3 a we. AGE 8.5%. :55 25502
3 a «.8- 2 a 2w. 95 “3:28 2.2205 ”58”.
ﬁns £8 28... ans «=8 28.5
$363.3» .33: kc

gang 33m 9.33 5 .393»: 9.38 35» £535? 8.9.393 BummSemuaﬁehuﬁm .N 035.

29

is calcium. In contrast the dihydropyridine derivative, nifedipine (10 uM) did not alter
the resting membrane potential or the conﬁguration of the action potentials (n=3) (data
not shown).

The data suggest that action potentials in neurons located in pelvic plexus ganglia
contain both voltage dependent sodium and calcium currents. The calcium current was
carried through m-conotoxin—sensitive but not nifedipine-sensitive voltage-gated calcium

channels.

30

Figure 5. Sodium and calcium components of the action potential (column A) and spike
dependence of the afterspike hyperpolarization (column B) in pelvic plexus ganglion
neurons. Aa and Ba, control spike and afterspike hyperpolarization. Application of
tetrodotoxin (TTX) (0.5-1 uM) abolished spike discharge (Ab) and afterspike
hyperpolarization (Bb). In the presence of TTX (0.5-1 uM), by increasing the intensity
of depolarizing current pulse, an action potential with lower amplitude and slower rate of
rise was achieved (Ac) while the afterspike hyperpolarization was not different from the
control (Bc). Superfusion of low Ca2+ (0.05 mM) high Mg2+ (30 mM) Krebs solution
containing TTX (0.5-1uM) abolished both the TTX-resistant spike (Ad) and afterspike

hyperpolarization (Bd). Ae and Be, washed with normal Krebs solution.

 

31

KS Control

 

 

 

 

 

 

Low Ca1r KS/l'I'X
d d
KS Wash
e e

 

. 45,.

10 ms

 

L63

mv.

32

Figure 6. Effects of tetraethylarnmonium (TEA, 60 mM) (a), TEA (60 mM) and
tetrodotoxin (TTX, 0.5-1uM) (b), low Ca2+ (0.05 mM) high Mg2+ (30 mM) containing
TEA (60 mM) and TTX (0.5-luM) (c) and TBA (60 mM), TTX (0.5-1 nM) and a)-
conotoxin GVIA (700 nM) (d) on action potentials recorded from pelvic plexus ganglion
neurons. Action potentials were evoked by brief depolarizing current injection (15-30 ms
duration, 0.2-0.5 nA) through recording electrode. Traces of action potentials before and

during application of test solutions (arrow) were superimposed.

 

33

KS/TEA KS/TEA + TTX
a / i b I

Low Ca2+ KS/TEA KS/l'EA + TTX
+ TTX + m—Conotoxin

I

 

 

 

 

34

Discussion

Intracellular injection and localization of HRP revealed two morphologically
different types of neurons in pelvic plexus ganglia They are monopolar neurons with
only one long process and multipolar neurons with two or more processes. As in other
mammalian parasympathetic ganglia such as guinea-pig bronchial parasympathetic
ganglia and dog intracardiac ganglia (Myers, Undem and Weinreigh, 1990; Xi, Thomas,
Jr., Randall and Wurster, 1991), the majority of neurons in guinea-pig pelvic plexus
ganglia are multipolar neurons. In contrast, multipolar neurons in opossum gallbladder
parasympathetic ganglia (Bauer, Hanani, Muir and Szurszewski, 1991) are less abundant.
The soma diameter resembles the size obtained by Blackman et al. (Blackman,
Crowcroft, Devine, Holman and Yonemura, 1969) using electron microscopy to measure
neurons in guinea-pig pelvic plexus ganglia. The soma diameter and the length of the
process of neurons in pelvic plexus ganglia are similar to these parameters deﬁned for
neurons in cat lumbar paravertebral ganglia, guinea-pig bronchial parasympathetic
ganglia, dog intracardiac ganglia, and opossum gallbladder ganglia (Percy, Walsh and
Krier, 1988; Myers, Undem and Weinreigh, 1990; Xi, Thomas, Jr., Randall and Wurster,
1991; Bauer, Hanani, Muir and Szurszewski, 1991). In the opossum gallbladder ganglia,
only the long process of monopolar neurons extend beyond the ganglion border in a ﬁber
tract while multipolar neurons had their processes remain inside the ganglia. The authors
considered the function of these ganglia to be a simple relay station (Bauer, Hanani, Muir
and Szurszewski, 1991). Unlike the neurons in opossum gallbladder ganglia, in guinea-
pig pelvic plexus ganglia, both monopolar and multipolar neurons in pelvic plexus
ganglia have one or more processes extended beyond the ganglion border and joined. in
the ﬁber tracts. We speculate that both types of neurons in guinea-pig pelvic plexus
ganglia may be involved in more complicated functions rather than acting as a simple

relay station (Bauer, Hanani, Muir and Szurszewski, 1991).

35

Intracellular recording techniques in vitro have previously been used to study the
electrophysiological properties of neurons in guinea-pig pelvic plexus ganglia (Blackman,
Crowcroﬁ, Devine, Holman and Yonemura, 1969; Crowcroﬁ and Szurszeski, 1971), cat
and rabbit pelvic ganglia (Grifﬁth III, Gallagher and Shinnick-Gallagher, 1980;
Nishimura, Akasu and Krier, 1991, 1992, 1993) and cat colonic ganglia (Krier and
Heartrnan, 1984). Electrophysiologically, the values for the resting membrane potential,
input resistance, time constant, threshold to initiate AP and ASH amplitude and duration
resembled those described by Blackman et a1. and Crowcroft and Szurszeski (Blackman,
Crowcroft, Devine, Holman and Yonemura, 1969; Crowcroﬁ and Szurszeski, 1971) for
neurons in guinea-pig pelvic plexus ganglia. The passive and active properties are also
similar to pelvic ganglion neurons in other species and other mammalian parasympathetic
ganglion neurons such as cat pelvic and colonic ganglion neurons (Grifﬁth III, Gallagher
and Shinnick-Gallagher, 1991; Krier and Heartrnan, 1984), guinea-pig bronchial ganglion
neurons (Myers, Undem and Weinreigh, 1990), dog intracardiac ganglion neurons (Xi,
Thomas, Jr., Randall and Wurster, 1991), guinea-pig gall-bladder ganglion neurons
(Mawe, 1990), rat intramural neurons from paratracheal ganglia (Allen and Burnstock,
1990), rat submandibular and cat ciliary ganglia (Skok, 1973). Similar membrane input
resistance is characteristic of the similar size of all these neurons. The resting membrane
potential in mammalian ganglion neurons ranged from -45 to -80 mV. The wide range of
values could be due to differences in the degree of damage to neurons by the
microelectrode. The similarities in resting membrane potential usually reﬂect similar
ionic mechanisms underlying the resting membrane potential. The similarities in time
constant and capacitance suggest that electrical properties and membrane surface area are
similar. The similarities in the passive and active properties of these neurons are not
surprising considering the general agreement that neurophysiological values are similar

for neurons in parasympathetic ganglia distributed throughout the autonomic nervous

36

systems of different mammalian species (Gallagher and Shinnick-Gallagher, 1986; Nishi,
1986).

Neurons in pelvic plexus ganglia were characterized as either phasic or tonic
based on their response to prolonged superthreshold depolarizing current injection. The
present study shows that the majority (60.0%) of neurons in guinea-pig pelvic plexus
ganglia exhibit a tonic ﬁring pattern. Phasic and tonic ﬁring patterns have been described
previously in several ganglia of guinea-pigs and cats. In cat pelvic ganglia, 93% of
neurons were identiﬁed as tonic (Griffith III, Gallagher and Shinnick-Gallagher, 1980).
The proportion of tonic neurons reported in studies of the celiac and superior mesenteric
ganglia varies between about 60% (Kreulen and Szurszewski, 1979) and 35% (Decktor
and Weems, 1983). In studies of inferior mesenteric ganglia, either only tonic ﬁring
patterns (Crowcroft and Szurszeski, 1971) or about 40% tonic, 60% phasic neurons
(Weems and Szurszewski, 1978; Kreulen and Szurszewski, 1979; Julé and Szurszewski,
1983) have been described. Functionally, the prevertebral ganglion neurons have been
described to be mainly involved with motility and secretion in the pelvic organs (Langley
and Anderson, 1896, a, b, c). whether the proportion of tonic or phasic neurons in certain
ganglia will determine the ganglionic function is still unclear. Further studies in inferior
mesenteric ganglia have shown that neurons which project in the hypogastric nerves, 87%
were tonic and about 12% phasic (Cassell, Clark and McLachlan, 1986). In contrast,
72% of neurons in cat renal ganglia discharged phasically (Decktor and Weems, 1981), as
did 96% in the lumbar sympathetic chain (Cassell, Clark and McLachlan, 1986). In the
present study, how many of the neurons with different ﬁring patterns may have been
projecting in the colonic nerves or urethra-urinary bladder nerves was not determined.
However, when the functions served by the components of the outflow from these various
ganglia are considered, it appears that these distinctive discharge patterns may provide a
basis for the broad classiﬁcation of sympathetic and parasympathetic ganglion neurons as

either vasoconstrictor, vasodilator or visceral.

37

Tonic and phasic neurons in sympathetic ganglia (Cassell, Clark and McLachlan,
1986; Dektor and Weems, 1983) and enteric ganglia (Nishi and North, 1973) but not
parasympathetic ganglia (Griffith III, Gallagher and Shinnick-Gallagher, 1980) have been
shown to differ in their active and passive membrane properties. Unlike neurons in
sympathetic ganglia which show differences in passive electrical properties that reﬂect
differences in neuronal geometry as well as differences in the population of membrane
ion channels opened at resting potential (Cassell, Clark and McLachlan, 1986; Dektor and
Weems, 1983), in guinea-pig pelvic plexus ganglia, there is no statistically signiﬁcant
difference between phasic and tonic neurons with respect to passive membrane
properties. The active electrical property study showed that after spike hyperpolarization
in tonic and phasic neurons was marginally different. However, not all the
electrophysiological properties, especially the properties that are considered to be
involved in the mechanism of tonic and phasic ﬁring pattern were studied in detail in the
present study. Based on the study ﬁom inferior mesenteric ganglion neurons (Cassell,
Clark and McLachlan, 1986), the mechanism involved in discharge patterns of the two
populations of neurons resulted from differences in voltage-dependent potassium
channels present in their membranes. The large aﬁerspike hyperpolarization produced by
calcium-activated potassium current might contribute to the slow rates of repetitive ﬁring
in tonic neurons, while the rhythmic ﬁring of tonic neurons is due to the transient A
current (an outward, voltage dependent potassium current). While the high frequency of
the initial discharge in phasic neurons can be explained by the absence of an A current,
the termination of ﬁring seems to be due to M current (outward, voltage-dependent
potassium current which can be blocked by muscarinic receptor agonists). The
magnitude of M-current was signiﬁcantly greater in phasic neurons than in tonic neurons.
In some tonic neurons, M-current could not even be detected. Some phasic neurons
exhibited two ﬁring patterns with either a single action potential or a short burst of action

potentials. These two patterns may resemble the two subclasses of phasic ﬁring neurons

38

found in guinea-pig inferior mesenteric ganglia (King and Szurszewski, 1988) and cat
pelvic ganglia (Grifﬁth III, Gallagher and Shinnick-Gallagher, 1980). Different
magnitudes of the M-current may contribute to these two patterns of ﬁring in phasic
neurons. Tonic and phasic neurons in guinea-pig pelvic plexus ganglia had similar somal
size and dendritic arbors. Monopolar, bipolar and multipolar neurons were all capable of
generating both phasic and tonic discharging patterns. This suggests that the capability of
generating different patterns of action potentials in neurons of pelvic plexus ganglia is not
correlated to the morphological differences.

Action potentials recorded from pelvic plexus ganglion neurons are similar to
other mammalian parasympathetic ganglion neurons (Allen and Burnstock, 1987; Mawe,
1990; Xi, Thomas, Jr., Randall and Wurster, 1991; Nishimura and Krier, 1993). The
depolarizing phase of the action potential is primarily the result of voltage-activated
inward sodium currents, because TTX blocked the action potentials. TTX-resistant
spikes were abolished by substituting normal Krebs solution to a low Ca2+ high Mg2+
Krebs solution or by superfusion with m-conotoxin GVIA. The appearance of a calcium-
dependent spike during superfusion with TTX indicates the presence of a voltage-
dependent calcium current in the rising phase of action potentials as well. A high
concentration of TEA alone broadened action potentials. Under this condition, the
broadened shoulder of the TEA-spike and the TTX-resistant action potential when TTX is
added to TEA-Krebs solution may be due to the blockage of a delayed rectiﬁer which
allows for opening of voltage-sensitive calcium channels. The calcium current was
reversibly blocked by m-conotoxin and was insensitive to nifedipine, suggesting that for
neurons in guinea-pig pelvic plexus ganglia, (o-conotoxin-sensitive N-type but not
nifedipine-sensitive L-type voltage-dependent calcium channels are activated during the
action potential. The N-type calcium channel has been suggested to be the principal
calcium channel subtype through which calcium enters nerve terminals and triggers the

release of many chemical transmitter substances. Action potentials in both tonic and

39

phasic neurons possess a calcium component which is expressed as a prolonged
potassium current. This current has been shown to be larger in tonic neurons than in
phasic neurons of sympathetic ganglia and has been considered as a major contributor of
aﬁerspike hyperpolarization (Cassell, Clark and McLachlan, 1986). However, whether a
larger afterspike hyperpolarization is due to a larger N-type calcium current is not known.
But if this is true, then more calcium will enter tonic neurons than in phasic neurons
through N-type calcium channel. Since calcium entry may also induce long term changes
on neuronal properties by activating the signal transduction pathway, this may also be a
signiﬁcant mechanism for different ﬁring pattern in tonic and phasic neurons. In the
present study, ASH amplitude and duration are greater in tonic neurons than in phasic

neurons, although statistically, these values are marginally signiﬁcant at P = 0.05.

CHAPTER 2
SYNAPTIC TRANSMISSION IN PELVIC PLEXUS GANGLION
NEURONS.

Introduction

Preganglionic ﬁbers in sacral ventral roots and pelvic nerves provide synaptic
inputs consisting of fast excitatory postsynaptic potentials (f-EPSPs) and/or action
potentials, to neurons in cat colonic ganglia (Krier and Hartman, 1984; Nishimura, Krier
and Akasu, 1991) and guinea-pig, cat and rabbit pelvic plexus ganglia (Blackman,
Crowcroft, Devine, Holman and Yonemura, 1969; Crowcroft and Szurszewski, 1971;
McLachlan, 1977; Grifﬁth III, Gallagher and Shinnick-Gallagher, 1980). Pelvic plexus
neurons also receive synaptic inputs from preganglionic ﬁbers in lumbar sympathetic
nerves (hypogastric nerves). Synaptic potentials evoked by electrical stimulation of the
hypogastric nerve and pelvic nerve are blocked by d-tubocurarine, hexamethonium and
dihydro-B-erythroidine (Blackman, Crowcroft, Devine, Holman and Yonemura, 1969;
Crowcroﬁ and Szurszewski, 1971; McLachlan, 1977) suggesting that they are mediated
by nicotinic acetylcholine receptors.

Although central preganglionic sympathetic (hypogastric) and parasympathetic
(pelvic) ﬁbers which provide synaptic inputs to pelvic plexus ganglia have been studied -
(Blackman, Crowcroft, Devine, Holman and Yonemura, 1969; Crowcroﬁ and
Szurszewski, 1971; McLachlan, 1977), it is yet unknown whether neurons in the pelvic
plexus ganglia receive synaptic input(s) from peripheral nerve trunks which connect the
pelvic ganglia with their effector organs, such as the distal colon-rectum and urethra-
urinary bladder. The concept of afferent excitatory synaptic inputs originating from

peripheral organs has previously been shown for neurons in sympathetic prevertebral
40

41

autonomic ganglia (inferior mesenteric, superior mesenteric ganglia and celiac plexus
ganglia) of guinea-pigs (Crowcroft, Holman and Szurszewski, 1971; Weems and
Szurszewski, 1977; Kreulen and Szurszewski, 1979a; Kreulen, Muir and Szurszewski,
1983)

In the present study, we hypothesize that neurons in pelvic plexus ganglia, like
neurons in inferior mesenteric ganglion neurons, receive synaptic inputs originated from
peripheral organs. The aim of the present experiments was to determine whether neurons
in pelvic plexus ganglia receive peripheral synaptic inputs from nerve ﬁbers that connect
distal colon (colonic nerves) and urethra-urinary bladder (urethra-urinary bladder nerves)
with pelvic plexus ganglia. The experiments were performed by using an in vitro
preparation which contained pelvic plexus ganglia and central (pelvic and hypogastric)
and peripheral (colonic and urethra-urinary bladder) nerve trunks. The data suggest that
in addition to the excitatory synaptic inputs received from central preganglionic nerve
ﬁbers, neurons in the pelvic plexus ganglia also receive peripheral synaptic inputs
consisting of f-EPSPs and APs in response to electrical stimulation of peripheral nerve
trunks. Thirty to forty percent of the responses originated ﬁ'om the peripheral organs
while the remainder may be mediated by axon collaterals from efferent and/or afferent

ﬁbers in hypogastric and/or pelvic nerves.

Methods
Experiments were performed on adult male guinea-pigs (200-500 gm), euthanized
by exsanguination following carbon dioxide gas-induced anesthetization. Pelvic plexus
ganglia (PG), together with pelvic nerves (PELN), hypogastric nerve (HGN), urethra-
urinary bladder nerves (UUBN) and colonic-rectal nerves (COL-RECN) were dissected
in situ and placed in a single compartment organ bath. A diagrammatic sketch of the in

vitro preparation is shown in Figure 7. In all preparations, PG were pinned on Sylgard to

42

Figure 7. Diagrammatic sketch of lumbar sympathetic and sacral parasympathetic
pathways to neurons in pelvic plexus ganglia and to distal colon-rectum of male guinea-

pigs. Note: inferior mesenteric ganglia and hypogastric ganglia are not shown.

 

 

.z 50005
22.5-9285

Isl: I . ,, ,, ... , . ‘ . ..... . o=acow
_. . -1. ...:on 0.29.
\ as. . u

Jarv- ..

43

I
’4

/ .2 25.0009»:

 

I

.z _QomzoEQOU a. . .2 28.00

6:00
62-663.

ttttt

44

the ﬂoor of the organ bath. The organ bath was superfused with a modiﬁed Krebs
solution (2-3 mein) containing (mM): Na“, 137.4; K+, 5.9; Co”, 2.5; Mg2+, 1.2; (31',
134; HCO3', 15.5; H2P04', 1.2; glucose, 11.5. The solution was preheated to 37-38 °C at
the recording site and equilibrated with a 95% 02 - 5% C02 gas mixture. Low Ca2+ high
Mg2+ Krebs solutions contained 0.05 mM Ca2+ and 30 mM Mg”.

Electrophysiological procedures. Intracellular potentials were recorded with
microelectrodes ﬁlled with 3 M-KCl and having tip resistances ranging from 60-110 MQ
. Membrane potentials were recorded using an ampliﬁer with an active bridge circuit
allowing current injection into neurons through the recording microelectrode. Signals
from the microelectrode were displayed on an oscilloscope with digitized memory and
recorded on a pen-writing chart recorder. Data were stored on a videocassette data
recorder and the output (individual electrotonic potentials, f-EPSPs, and action
potentials) were plotted (superimposed consecutive records) with an X-Y plotter.

Fast-EPSPs and APs were recorded from neurons in response to electrical
stimulation of preganglionic ﬁbers. Bipolar platinum stimulating electrodes were
positioned on central nerve trunks which connect the pelvic plexus ganglia with the sacral
spinal cord (pelvic nerves) and with inferior mesenteric ganglia (hypogastric nerve) and
peripheral nerve trunks which connect the ganglia with distal colon-rectmn (colonic-
rectal nerves) and with urethra-urinary bladder (urethra-urinary bladder nerves).
Stimulation was produced by rectangular pulses of 0.3-0.5 ms duration at various
frequencies and intensities.

Antidromic potentials were distinguished from synaptic potentials by the
occurrence of all-or-none action potentials without a subthreshold synaptic potential and
by the occurrence of fast rising potentials during high frequency nerve stimulation that
decayed at the rate of the membrane time constant during the injection of hyperpolarizing
current pulses (Hartman and Krier, 1984). Also, antidromic potentials are resistant to

hexamethonium (Kreulen and Szurszewski, 1979b). Drugs were dissolved directly into

45

Krebs solution. Solutions were changed by using a three-way stopcock and superfused
on ganglia.

Chronic surgical procedures. In ﬁve animals, sacral dorsal and ventral roots on
one side (S2 - S4) and hypogastric nerves on both sides (1 -2 cm below the caudal lobes of
the inferior mesenteric ganglia) were transected under ether anesthesia. Following 5 - 27
days of recovery, the animals were reanethestized and surgically prepared for removal of
chronically decentralized pelvic ganglia and the associated nerve trunks for the in vitro
experiments.

Drugs used were hexamethonium bromide (Sigma), dihydro-B-erythroidine
(Sigma), atropine sulfate (Sigma), gamma-aminobutyric acid (GABA, Sigma),

Data were expressed as means d: SEM The signiﬁcance of differences was

assessed by a Chi-Square test for nominal data and P < 0.05 was considered signiﬁcant.

Results

Synaptic potentials and pharmacological studies. Electrical stimulation of
preganglionic ﬁbers in hypogastric nerve and pelvic nerves elicited action potentials
and/or subthreshold f-EPSPs in neurons of pelvic plexus ganglia. Superfusion of the
ganglia with Krebs solution containing hexamethonium (C6) (10 - 100 nM) (n=180),
dihydro-B-erythroidine (10 - 100 pM) (n=40) and gamma-aminobutyric acid (GABA) (10
- 100 pM) (n=5) or by stimulating nerve trunks at frequencies of 5-15 Hz (n=20)
reversibly suppressed orthodromic action potentials and completely abolished fast EPSPs
(Fig. 8). Atropine (0.1 - 1.0 pM) (n=14) did not alter orthodromic action potentials or
fast EPSPs. Similar responses were also obtained by electrical stimulation of peripheral
nerve trunks that connected the pelvic plexus ganglia with distal colon-rectum (colonic-
rectal nerves) and with urethra-urinary bladder (urethra-urinary bladder nerves). The

data suggest that neurons in pelvic plexus ganglia receive synaptic inputs consisting of f-

46

 

 

KS Control KS/c6 10‘5M 10min KS Wash 25min
Urethra-Urinary Bladder N. 0
A B C "IV
Iﬁ -60
Colonic N. 0

 

 

 

 

LC 4 .

l
1

€

 

 

 

 

 

 

mv
G H l
__ l -72
Hypogastric N. 0

J K L "w
__ __ -63

T .__.

20 ms

Figure 8. Synaptic responses of neurons in pelvic plexus ganglia during stimulation of
urethra-urinary bladder nerves (A-C), colonic-rectal nerves (D—F), pelvic nerves (G-1) and
hypogastric nerve (J-L). A, D, G and J, synaptic responses of neurons in Krebs solution
(KS). B, E, H and K, synaptic responses of neurons during superfusion of ganglia with a
Krebs—hexamethonium (C 6) solution (10 nM). C, F, I and L, synaptic responses of
neurons in pelvic plexus ganglia 25 minutes aﬁer superfusion of ganglia with Krebs
solution (KS). Each panel represents superimposed records of two successive responses

to nerve stimulation at maximum intensity.

47

EPSPs and action potentials not only from central nerve trunks (hypogastric nerve and
pelvic nerves) but also ﬁ'om peripheral nerve trunks (colonic-rectal nerves and urethra-
urinary bladder nerves). The synaptic inputs were mediated by acetylcholine nicotinic
receptors.

Figure 9a shows the distribution of synaptic input(s) to neurons in pelvic plexus
ganglia from electrical stimulation of the nerve trunks indicated. Electrical stimulation of
hypogastric, pelvic, colonic-rectal and urethra-urinary bladder nerves, elicited synaptic
responses in 84 % (205 of245), 80 % (102 of 128), 67 % (108 of 161) and 80 % (109 of
135) of neurons tested, respectively.

The data indicate that majority of neurons in pelvic plexus ganglia received
synaptic inputs not only from central nerve trunks (hypogastric nerve and pelvic nerves)
but also from peripheral nerve trimks (colonic-rectal nerves and urethra-urinary bladder
nerves).

Pattern of synaptic potentials. Electrical stimulation of hypogastric nerve, pelvic
nerves, colonic-rectal nerves and urethra-urinary bladder nerves evoked single and
multiple synaptic inputs in pelvic plexus neurons. Single synaptic inputs are deﬁned as
either a single fast excitatory postsynaptic potentials (f-EPSPs) or a single action
potential. Multiple synaptic inputs are deﬁned as either a single action potential and a
single f-EPSP and/or multiple f-EPSPs and/or multiple action potentials.

. Figure 9b shows that > 55% of neurons received multiple synaptic inputs from all
nerve trunks stimulated (75.0%, 153 of 205 from hypogastric nerve; 57.0%, 58 of 102
from pelvic nerves; 55.0%, 59 of 108 from colonic-rectal nerves and 59.0%, 64 of 109
from urethra-urinary bladder nerves). The remainder received single synaptic input
(25.0%, 52 of 205 from hypogastric nerve; 43.0%, 44 of 102 from pelvic nerves; 45.0%,
49 of 108 from colonic-rectal nerves and 41.0%, 45 of 109 from urethra-urinary bladder
nerves). The majority of neurons received multiple synaptic potentials, suggesting

differences in the conduction velocity of the nerve ﬁbers.

48

Figure 9. a: Synaptic inputs to neurons in pelvic plexus ganglia. Height of columns
represents percentage of cells tested which received synaptic input from indicated nerve
trunks. Numbers represent population of neurons tested.

b: Single and multiple synaptic inputs to neurons in pelvic plexus ganglia.
Height of columns represents percentage of cells tested which received either single or
multiple synaptic inputs from indicated nerve trunks. Single synaptic input (shaded
column) consisted of either a single EPSP or a single action potential. Multiple synaptic
inputs (open column) consisted of either a single action potential and a single EPSP and
/or multiple EPSPs and /or multiple action potentials. Numbers represent the population
of neurons which received synaptic response(s). HGN, hypogastric nerves; PELN, pelvic

nerves; COL- RECN, colonic-rectal nerves and UUBN, urethra-urinary bladder nerves.

49

O

 

135

 

 

 

161

 

 

 

128

 

 

 

 

245

 

 

 

100 —

_ V _ _ _ _ _ _ _
O O O O O O O O O
9 8 7 6 5 4 3 2 1

UBmmH 260 do mmoacooumd

O

PELN COL-RECN UUBN

HGN

..D

205

80r

 

9

o _\\\\\\\\\\\\

1

 

SNN\\\\ x

108

 

 

 

 

T\\\\\\\\\\\.

102

 

 

 

 

_\\\\\W \

 

_ _ _ _
O O O O O O
7 6 5 4

ponmob m__oo do oooEooLon.

 

10—
O

N UUBN

’\
\J

PELN COL—RB

HGN

50

Convergent synaptic inputs from central and peripheral nerve trunks to a
single neuron in pelvic plexus ganglia. To study the convergent synaptic inputs from
central and peripheral sources, electrical stimulation was used to activate hypogastric
nerve, pelvic nerves, colonic-rectal nerves and urethra-urinary bladder nerves. The
distribution of neurons that receive synaptic inputs (consisting of f-EPSPs and action
potentials) from one, two, three and four nerve trunks was determined.

Convergent synaptic input was tested in 107 neurons. Figure 10a shows
convergent synaptic input to one neuron in pelvic plexus ganglia during electrical
stimulation of preganglionic ﬁbers in colonic-rectal nerves, urethra-urinary bladder
nerves, pelvic nerves and hypogastric nerve. The distribution of neurons that receive
synaptic inputs (f-EPSPs and action potentials) from one, two, three or four nerve trunks
is shown in Figure 10b. Eighty-two percent of neurons tested (88 of 107) received
convergent synaptic input mediated by acetylcholine nicotinic receptors from ﬁbers
within two, three and four nerve trunks. Fifteen percent of neurons (16 of 107) received
synaptic input from ﬁbers within one nerve trunk. The remainder received no synaptic
input from any of the nerve trunks.

The data indicate that the majority of neurons in pelvic plexus ganglia receive
convergent synaptic inputs from ﬁbers within the hypogastric nerve, pelvic nerves,
colonic-rectal nerves and urethra-urinary bladder nerves.

Origin of synaptic potentials evoked by electrical stimulation of colonic-rectal
nerves and urethra-urinary bladder nerves. To determine whether the synaptic inputs
to neurons in pelvic plexus ganglia elicited by electrical stimulation of colonic-rectal and
urethra-urinary bladder nerves were mediated via collaterals of primary afferent ﬁbers
and preganglionic or postganglionic efferent ﬁbers in hypogastric and pelvic nerves,
synaptic transmission was studied following degeneration of ﬁbers in the hypogastric and

pelvic nerves (decentralization).

51

Figure 10. Convergence of synaptic input(s) to a neuron in pelvic plexus ganglia from
central (hypogastric nerve and pelvic nerve) and peripheral nerve trunks (colonic-rectal
nerves and urethra-urinary bladder nerves).

A: Convergent synaptic input to a single neuron in pelvic plexus ganglia when colonic-
rectal nerve (a), pelvic nerve (b), urethra-urinary bladder (c) and hypogastric nerve (d)
were electrically stimulated. Each panel represents superimposed traces of two
successive responses to nerve stimulation at maximum intensity.

B: Height of columns represents the percentage of cells tested. The numbers (1-4)
represent the number of convergent synaptic inputs to a single neuron. Note: 33% and
38% of neurons tested received convergent synaptic inputs from three or four nerve

trunks, respectively. N=107, total number of neurons tested.

Percentage of Cell: Tested

Ln
0

ﬁ

30

20

10

 

 

 

52

 

 

 

 

 

 

 

P E.

 

 

. (40)
(36) —'
_.

’ (16)

(12)
(3)
[—1
O I 2 3 4

Number of Nerve Trunks

O

3
<

'3

53

 

 

 

 

 

 

 

 

 

   

100 r
’0 ' , 245 128 '35
- 8 so - [_— _ T—

13 161
i ’°' -~
0 °° '
'5 5° _ 35
r ’° *

20 l'

10 l- 35 35

o ____h_———
HGN PELN COL-RECN UUBN

Figure 11. Synaptic input to pelvic plexus ganglion neurons in animals with an intact
nervous system (open column) and aﬁer degeneration of preganglionic ﬁbers in
hypogastric nerves and pelvic nerves (dark column). The height of columns represents
the percentage of cells tested which elicited excitatory synaptic potentials from the
indicated nerve trunks at maximal stimulation intensities. HGN, hypogastric nerve.
PELN, pelvic nerve. COL-RECN, colonic-rectal nerves. UUBN, urethra-urinary bladder

nerves. Numbers on top of the columns represent the population of neurons tested.

54

The experiments were performed in 35 decentralized neurons of pelvic plexus
ganglia from ﬁve guinea-pigs. Figure 11 shows the distribution of f-EPSPs and action
potentials of neurons in animals with an intact nervous system (open column) and after
degeneration of preganglionic ﬁbers in hypogastric nerves and pelvic nerves (dark
column). In animals for which ﬁbers in hypogastric nerves and pelvic nerves had
degenerated, the synaptic potentials elicited by electrical stimulation of hypogastric nerve
and pelvic nerves were nearly abolished (2.8%, 1 of 35) while 40.0% (14 of 35) and
31.0% (11 of 35) of neurons tested received synaptic responses during electrical
stimulation of colonic-rectal and urethra-urinary bladder nerves, respectively. The data
suggest that approximately 30-40 % of synaptic responses originate from peripheral
neurons located either in the myenteric plexus or submucous plexus of the distal colon or
in intramural plexus of urethra-urinary bladder. The data also suggest that axon
collateral ﬁbers from afferent ﬁbers and preganglionic efferent ﬁbers in the hypogastric
nerve and pelvic nerve may also provide synaptic inputs to neurons in pelvic plexus
ganglia.

Table 3 shows the percentage of neurons that received single and multiple
synaptic inputs during electrical stimulation of the respective nerve trunks. Data were
obtained from animals with an intact nervous system and from animals aﬁer chronic
degeneration of the hypogastric and pelvic nerves. Sixty percent of neurons in pelvic
plexus ganglia received multiple synaptic inputs during electrical stimulation of the
peripheral nerve trunks obtained from animals with an intact nervous system, while only
20—30 % of neurons in pelvic plexus ganglia received multiple synaptic inputs during
electrical stimulation of the peripheral nerve trunks obtained from animals after
degeneration of hypogastric and pelvic nerves. The data indicate that synaptic potentials
recorded from neurons after chronic degeneration of hypogastric and pelvic nerves,
consisted primarily of single f-EPSPs or single action potentials. The data in Table 4

indicate that the changes were not due to the alteration of active and passive electrical

55

.320: .832; 35.5-2523 .ZmDD ”mote: 30262260 .ZUm—MJOU

 

Gm be S oeoﬁm Gm ho m3 38.9. as we v8 $3.0m 32 .3 n3 seq—v Zm—DD

 

2m .8 my 95ch Sn mo 08 «$de 82 mo 05 owoﬁm 82 mo 9; $5.9. ZUNMJOU

 

 

 

 

$225 $229. wagon ”3:95 oomﬂnaum
3.35m 032:2 ounmﬁm Ewim ouaaim 032:2 owaaim 2min «JEEP 0202
".9320on 24mm .20: :— EBE Esmxm 20:02 885 :<
2:2 98me ES? «2255‘. 5 32:0 5.3 m «8:2. E 25:52

 

 

 

 
 

dot»: ~33 VS» 8&8an 5 macawsumeemweumﬁmxe coneuueuuou Lake 33 £899. 36E»:
885 § .33 £35.33 5 .263»: 6%.? 3333 £33m» 33:3: 33 235% renaasua .n 035,—.

 

56

.368 £8.83 .«o 598:: .AZV

”mote: 320a .ZAmE mote: 058393 .205 Hosanna—0&3»: exam “can .$m< “3288 cores #2

 

lam—v ﬁu— ﬂ Nd: 85 a.«— an 6.3— SD ﬁe H QoN— A35 62350 $8 Eml<|

 

G: ... a an. 8: a... « v.2 as S « «...: 9.5 82.55 :2
as n. a 3 F9 2 a 3 5.: 3 a 3 9.5 8:28 8% 32.28 =29.
as an a «.2. 99 5 a on. an: 2 a 92. 9.5 «38.8 8:8 .8 82.55
6: 3 a 92 a: 3 a n: as 8 u ..e 9.5 .2 ... 232 a 3:28 2285
G: 8... a a... a: 8.: a n8 :5 86 a and 25 .2 5 3.5.5 2 .355 2285.
G: 3 a a.“ a: 2 a on as 8 a no as .5 .528 25. osseo:
a: ....n a 3n 8: tn « he. as 3 « ...: 5:: 853.8. 59: 8225:
ac S « 2h- 89 ... a v.2- a»: 3 a new. 9.5 season ossaoe mouse
gimme use? zo: .882 8:82.. i
2.5.. + 22. 2V mm a 5o:

     
 

 

 

63.3333» 33: so names» 333 £22m 5 25.52.
3535.83. use 3.83:... 26! .359. s :55»: so 8332; Bﬁmssuagmgusm .v “...—eh.

57

properties of neurons, because there were no signiﬁcant differences between parameters
obtained from animals with an intact nervous system and from animals with degenerated
hypogastric and pelvic nerves on these properties .

Conduction velocities of preganglionic ﬁbers in central nerve trunks and fibers
in peripheral nerve trunks. The calculated conduction velocities (m/s) of preganglionic
ﬁbers in hypogastric nerve, pelvic nerves and nerve ﬁbers in colonic-rectal nerves and
urethra-urinary bladder nerves were determined by measuring the latencies (ms) of
synaptic responses during electrical stimulation of central and peripheral nerve ﬁbers,
subtracting 4.0 ms, a value estimating the ganglionic delay (Hartman and Krier, 1984)
and dividing conduction distance (mm) by this value.

Fig. 12 shows the distribution of conduction velocities from ﬁbers in hypogastric
nerve, pelvic nerves, colonic-rectal nerves and urethra-urinary bladder nerves that provide
synaptic inputs to neurons in pelvic plexus ganglia. Sixty % of ﬁbers from all nerve
trunks had conduction velocities ranging between 2 to 16 m/s. The remainder had the
conduction velocities ranging between 0.1-2.0 m/s. The data indicate that the nerve ﬁbers
in peripheral nerve trunks and preganglionic ﬁbers in central nerve trunks were B and C
type ﬁbers (Skok, 1979).

Antidromic Responses. Electrical stimulation of central (hypogastric and pelvic)
and peripheral (colonic-rectal, urethra-urinary bladder) nerve trunks evoked all or none
antidromic potentials in neurons of pelvic plexus ganglia. The potentials were resistant
to the ganglionic blocking agent, hexamethonium (100 nM) or replaced by all or nothing
fast rising potentials when nerve trunks are stimulated at high frequencies ranging
between 20-50 Hz (Fig. 13). The percentage of neurons tested in pelvic plexus ganglia
that exhibit antidromic potentials to electrical stimulation of central sympathetic
(hypogastric), parasympathetic (pelvic) and peripheral (colonic-rectal, urethra-urinary

bladder) nerve trunks is summarized in Fig. 14a. During stimulation of central nerve

58

- Hypogastric Nerve
[ZZ Pelvic Nerve

 

 

 

 

 

 

 

 

 

 

 

 

 

 

100 l— S Colonic-Rectal Nerve
90 _ Urethra-Urinary Bladder Nerve
80 -
3 7O _
:9 \\
g §
8) SO — x
g 40 - §
X
x /
a: k / X
20 r- / \
s / \
\ s
O, >\ .\
0.1—2.0 m/s 2.0—160 m/S

Conduction Velocity

Figure 12. Calculated conduction velocities of preganglionic ﬁbers which provide

synaptic inputs to neurons in pelvic plexus ganglia. Height of columns represent

percentage of ﬁbers from indicated nerve trunks.

59

Figure 13. Antidromic action potential recorded from one neuron in pelvic plexus
ganglia during electrical stimulation of urethra-urinary bladder nerves at 0.5 Hz (A), 15
Hz (C), 20 Hz (D) and 50 Hz (B) when superfused with a normal Krebs solution and at

0.5 Hz when superfused with a Krebs solution containing hexamethonium (10 nM) (B).

60

ca. .

>8

 

me O—

 

 

 

 

 

 

 

N: can
m
we mm 2: mm
N: odm N: oﬁ
a u
me O— me O—
558 21: coax N: no

61

Figure 14. a: Antidromic response to neurons in pelvic plexus ganglia. Height of
columns represent percentage of cells tested which received antidromic response from
indicated nerve trunks. Numbers represent population of neurons tested. HGN,
hypogastric nerves; PELN, pelvic nerves; COL- RECN, colonic-rectal nerves and UUBN,
urethra-urinary bladder nerves. b: Calculated conduction velocities of nerve ﬁbers which
provide antidromic inputs to neurons in pelvic plexus ganglia. Height of columns

represent percentage of ﬁbers from indicated nerve trunks.

62

Q

135

Percentage of Cells Tested

 

 

HGN PELN COL—RECN UUBN

Nerve Trunks Stimulated

- Hypogastric Nerve

b Pelvic Nerve
100 .. [:l Colonic-Rectal Nerve
,\\\\\V Urethra-Urinary Bladder Nerve

90 —

8 80 r

0)

:9 70 ~

Ll.

~0— 60 *-

O

m 50-

U?

.9 4o -

S

o 30 l-

L

(D

O. 20 ~
10 -

 

 

 

 

 

01-20Im/s ZO—llorn/s
Conduction Velocity

63

trunks, about 8.0 % (19 of 245) of neurons tested responded with antidromic responses to
electrical stimulation of hypogastric nerve, whereas 13.0 % (16 of 128) of neurons tested
responded with antidromic responses to electrical stimulation of pelvic nerves. During
peripheral nerve trunks stimulation, 30.0 % (48 of 161) and 39.0 % (52 of 135) of
neurons tested responded with antidromic responses to electrical stimulation of colonic-
rectal nerves and urethra-urinary bladder nerves respectively.

The calculated conduction velocities (m/s) of postganglionic ﬁbers in colonic—
rectal nerves, urethra-urinary bladder nerves, hypogastric nerve and pelvic nerves were
determined by measuring the latencies (ms) of antidromic potentials during electrical
stimulation of central and peripheral nerve ﬁbers and dividing conduction distance (mm)
by this value.

Fig. 14b shows the distribution of conduction velocities from ﬁbers in hypogastric
nerve, pelvic nerves, colonic-rectal nerves and urethra-urinary bladder nerves that provide
antidromic inputs to neurons in pelvic plexus ganglia. The majority (70—90%) of ﬁbers
from all nerve trunks had conduction velocities ranging between 2 to 13 m/s. The
remainder had conduction velocities ranging between 0.1-2.0 m/s. These data suggest
that the majority of postganglionic ﬁbers in peripheral nerve trunks and central nerve
trunks were B ﬁbers (Skok, 1979).

Discussion
Intracellular recording techniques in vitro have previously been used to study
synaptic transmission ﬁ’om central preganglionic sympathetic (hypogastric) and
parasympathetic (pelvic) nerve ﬁbers to neurons in pelvic plexus ganglia of male guinea-
pigs (Blackman, Crowcroﬁ, Devine, Holman and Yonemura, 1969; Crowcroﬁ and
Szurszeski, 1971; McLachlan, 1977) and cat colonic ganglia (Krier and Hartman, 1984).
The present data are similar to the results obtained in neurons of guinea-pig pelvic plexus

ganglia by other researchers (B lackman, Crowcroft, Devine, Holman and Yonemura,

64

1969; Crowcroft and Szurszeski, 1971; McLachlan, 1977) and the results obtained in cat
colonic ganglia mentioned above (Krier and Hartman, 1984). The synaptic responses
consist of f-EPSPs and orthodromic action potentials and are mediated by cholinergic
nicotinic receptors. However, in addition to synaptic inputs from central nerve trunks, the
present study shows that neurons in pelvic plexus ganglia receive synaptic inputs from
nerve ﬁbers that connect pelvic plexus ganglia with peripheral organs (distal colon and
urethra-urinary bladder). These synaptic responses are also mediated by cholinergic
nicotinic receptors. The present study is the ﬁrst to investigate synaptic transmission
which originates from peripheral organs. In the absence of orthodromic stimulation,
spontaneous f-EPSPs and action potentials are rarely seen. This observation is consistent
with that stated by Blackman et al. (Blackman, Crowcroﬁ, Devine, Holman and
Yonemura, 1969).

There are two patterns of nerve stimulation-evoked synaptic responses from each
nerve ﬁber tract. In the ﬁrst pattern (less than 50%), the superthreshold response of the
neuron is either a single action potential or a single excitatory postsynaptic potential. In
this pattern, the neurons may receive synaptic input from a single nerve ﬁber or several
nerve ﬁbers with similar thresholds and conduction velocities. In the second pattern
(55.0%-75.0%), the superthreshold response of neurons consists of multiple potentials
with either a single action potential and multiple f-EPSPs or a single action potential and
single EPSP or multiple action potentials and multiple f-EPSPs. In this pattern, the
neurons likely receive synaptic inputs from multiple nerve ﬁbers with different
conduction velocities. Both single and multiple patterns of synaptic potentials in guinea-
pig pelvic plexus ganglion neurons are similar to pelvic plexus ganglion neurons in other
species and to other mammalian parasympathetic and sympathetic ganglia. However, the
distribution of these two patterns in different ganglia is different. Generally, the majority
of neurons in parasympathetic ganglia such as adult rat submandibular ganglia (Lichtman,

1977) and cat ciliary ganglia (Martin and Pilar, 1963) were considered to receive single

65

synaptic potentials while majority of neurons in sympathetic ganglia such as guinea pig
and rabbit superior cervical ganglia (Perri, Sacchi and Casella, 1970; Wallis and North,
1978) and guinea-pig inferior mesenteric ganglia (Crowcroﬁ and Szurszewski, 1971)
receive multiple synaptic potentials. Yet, in the cat colonic ganglia (Krier and Hartman,
1984) which is a parasympathetic ganglia, the majority of neurons receive multiple
synaptic potentials. Traditionally, due to their higher percentage of neurons receiving
multiple synaptic input (Crowcroﬁ, Holman and Szurszewski, 1971), sympathetic ganglia
have been considered to be involved in more complicated integration function than
parasympathetic ganglia. However, the present study demonstrates that pelvic plexus
ganglia, a parasympathetic ganglia, had distribution of synaptic input similar to that of
sympathetic ganglia, indicating the role pelvic plexus ganglia plays is more complicated
than a simple relay station like other researchers predicted (Crowcroﬁ, Holman and
Szurszewski, 1971).

The majority of neurons in pelvic plexus ganglia (82.0 %) receive convergent
synaptic inputs from two, three and four nerve ﬁber tracts that involve both central
sympathetic, parasympathetic and peripheral nerves. Fiﬁeen percent of neurons receive
synaptic input from only one nerve trunk. The remainder receive no synaptic inputs. It is
possible that neurons that receive a non-convergent synaptic input may function as simple
relay stations with little capability of modulating synaptic input arising from either the
central nervous system or peripheral organs. In contrast, neurons that receive a
convergent synaptic inputs may be capable of integrating inputs from central sympathetic,
parasympathetic and peripheral nerves. The data also indicate that neurons in pelvic
plexus ganglia integrate inputs from both central and peripheral nerve trunks. Thus it can
be suggested that interactions between the central nervous system and peripheral organs
or interactions between two peripheral organs (colon-rectum and urethra-urinary bladder)

can take place or be regulated at the pelvic plexus ganglia level.

66

Sixty percent of the neurons tested exhibit synaptic potentials during electrical
stimulation of peripheral nerve trunks (colonic-rectal nerves and urethra-urinary bladder
nerves). There are several possible explanations of this ﬁnding. First, axon collateral
ﬁbers in pre-and/or postganglionic axons in hypogastric nerve or pelvic nerve may
provide synaptic innervation to neurons in pelvic plexus ganglia. Second, collaterals of
primary afferent ﬁbers may provide synaptic innervation to neurons in pelvic plexus
ganglia. Third, ﬁbers that provide synaptic inputs to neurons in pelvic plexus ganglia
may originate from peripheral neurons located in enteric ganglia of distal colon and in
intramural ganglia of urinary bladder. There have been a number of previous studies
indicating that neurons in sympathetic prevertebral ganglia (inferior mesenteric and celiac
ganglia) received mechanoreceptive peripheral synaptic input from the colon and small
intestine (Crowcroﬁ, Holman and Szurszewski, 1971; Kreulen and Szurszewski, 1979b).
It is possible, therefore, that neurons in pelvic plexus ganglia receive a similar peripheral
synaptic innervation. In chronic decentralized animals, the synaptic inputs from
peripheral nerve trunks (colonic-rectal and urethra-urinary bladder nerves) are reduced,
but still 30%- 40% of the synaptic responses remain. The results of the present study
indicate that 30%-40% of the synaptic responses from electrical stimulation of peripheral
nerves originate ﬁ'om peripheral organs. Also, we suggest that axon collateral ﬁbers from
afferent and pre/postganglionic ﬁbers in hypogastric and pelvic nerves are involved.
Eighty percent of the remaining responses are single responses with the majority
consisting of single f-EPSPs. The passive and active electrical properties of neurons in
pelvic plexus ganglia are not diﬁ‘erent. I conclude that the pattern of the synaptic
potential change may due to the elimination of axon collaterals from central nerve ﬁbers
and not due to the changes in neuronal electrical properties.

From the conduction velocities, I conclud that the majority of preganglionic ﬁbers
in central nerve trunks are B and C ﬁbers (conduction velocity of B ﬁbers ranging from
2-15 m/sec; conduction velocity of C ﬁbers ranging from 0.1-2.0 m/sec) (Skok, 1973).

67

The present data are similar to the data obtained for preganglionic ﬁbers that synapse in
sympathetic ganglia (inferior mesenteric ganglia, superior cervical ganglia and stellate
ganglia) and parasympathetic ganglia (ciliary ganglia, intracardiac ganglia and
submandibular ganglia) including guinea pig pelvic plexus ganglia (Skok, 1973). Also,
the majority of the nerve ﬁbers in peripheral nerve trunks that synapse with neurons in
pelvic plexus ganglia are B and C ﬁbers. The B and C ﬁbers are generally afferent ﬁbers
which further supports the hypothesis that neurons in pelvic plexus ganglia may receive
synaptic input from afferent ﬁbers that originate from the periphery.

Antidromic potentials were recorded during electrical stimulation of central
sympathetic (hypogastric) and parasympathetic (pelvic) nerves and peripheral nerve
trunks (colonic-rectal and urethra-urinary bladder nerves). The results indicate that
postganglionic neurons located in pelvic plexus ganglia send their axons both back to
central nervous system and to periphery. Antidromic responses recorded from guinea-pig
pelvic plexus ganglion neurons during electrical stimulation of the hypogastric nerve
were also established in other laboratories (Blackman, Crowcroft, Devine, Holman and
Yonemura, 1969). Our data also conﬁrm the anatomy data obtained previously
(McLachlan, 1985), where the application of horseradish peroxidase (HRP) to the distal
end of the transected hypogastric nerve labels some neurons in pelvic plexus ganglia.
Similar anatomical results were also obtained in the rats (N adelhaft and Vera, 1991). It is
not surprising that we found peripheral nerve trunks exhibited antidromic responses thus
indicating that the pelvic plexus ganglion neurons send their postganglionic axons to
distal colon, rectum, urethra-urinary bladder. It has previously been shown that
postganglionic parasympathetic axons may provide phasic and tonic synaptic inputs to
neurons in myenteric plexus or directly innervate colonic smooth muscle, blood vessels
or mucus glands (de Groat and Krier, 1976; Hulter, Jodal and Lundgren, 1969; Wright,

Florey and Jennings, 193 8).

CHAPTER 3
NEURON S IN PELVIC PLEXUS GANGLIA RECEIVE
MECHANOSENSORY INPUT FROM DISTAL COLON.

Introduction
Concept of Neurons in Prevertebral Ganglia Receiving Mechanosensory Input From
Peripheral Organs.

The abdominal ganglia that lie anterior and anteriolateral to the abdominal aorta
are called abdominal prevertebral ganglia. They are divided into four groups: celiac
ganglia, superior mesenteric ganglia, inferior mesenteric ganglia and pelvic -hypogastric
plexus. Neurons in the ﬁrst three groups of prevertebral ganglia are mainly adrenergic
while neurons in pelvic-hypogastric plexus are mainly cholinergic (Szurszewski and
King, 1989)

The concept that neurons in guinea-pig prevertebral ganglia (inferior mesenteric,
superior mesenteric ganglia and celiac plexus ganglia) receive excitatory cholinergic
synaptic inputs consisting of f-EPSPs and action potentials originating from
mechanoreceptors located in the peripheral organs has been demonstrated by using
intracellular recording techniques and in vitro preparations which consist of an isolated
segment of small intestine or distal colon attached to the prevertebral sympathetic ganglia
(Crowcroft, Holman and Szurszewski, 1971; Weems and Szurszewski, 1977; Kreulen
and Szurszewski, 1979a; Kreulen, Muir and Szurszewski, 1983). Furthermore, studies
presented in Chapter 2. have shown that neurons in pelvic plexus ganglia also receive
cholinergic synaptic inputs from ﬁbers within peripheral nerve trunks that connect the

pelvic plexus ganglia with effector organs such as distal colon or urethra-urinary bladder.

68

69

Properties of Visceral Mechanoreceptors in Distal Colon-Rectum

Two types of mechanoreceptors have been described in the gastrointestinal
system (Iggo, 1957). One type, termed tension receptors, is located in serial with smooth
muscle cells and exhibits a low threshold pressure to passive distension. Tension
receptors are spontaneously active and their activities are increased by contractions of
smooth muscle. The second type, termed stretch receptors, is located in parallel with
smooth muscle cells and exhibits a high threshold for activation by passive distension.
These mechanoreceptors are not spontaneously active and can not be activated by active
contractions of the smooth muscle. They are activated by fast rates of passive distension
and exhibit a rapidly adapting discharge of action potentials. The studies from human
rectum showed that the mechanoreceptors located within the smooth muscle layers,
exhibited properties of tension receptors while mechanoreceptors located on mucosa,
exhibited properties of stretch mechanoreceptors (Sun et al., 1990). In guinea-pig distal
colon, the mechanoreceptors that mediate synaptic inputs from colon to neurons in
inferior mesenteric ganglia and from small intestine to neurons in celiac ganglia, inferior
mesenteric ganglia have the properties of tension receptors (Szurszewski and King, 1989)
and not stretch mechanoreceptors, because both distension and contraction of the colon
elicit mechanosensory synaptic input (Kreulen and Peters, 1986). A stretch-activated
mechanoreceptor would not be stimulated by a decrease in intestinal or colonic diameter
during contraction.

Based upon the response of individual primary afferent ﬁbers to increasing levels
of balloon distension, mechanoreceptors in the gastrointestinal tract have been classiﬁed
into three types. They are low threshold mechanoreceptors (LTM), high threshold
mechanoreceptors (HTM) and wide-dynamic range mechanoreceptors (WDM) (Grundy,
1993)

70

Classiﬁcation ofmechanoreceptive primary afferents in cat colon-rectum.

Primary afferent ﬁbers in cat colon-rectum have been classiﬁed on the basis of
their size, conduction velocities and discharge of action potentials. One type are small
diameter myelinated ﬁbers with a median conduction velocity of 8.0 m/s. Their
activation by mechanosensory stimuli causes a phasic discharge of action potentials. The
other type are small diameter non-myelinated ﬁbers with a median conduction velocity of
1.7 m/s. Their activation by mechanosensory stimuli causes a tonic discharge of action
potentials (Janig and Koltzenberg, 1991). Some nonmyelinated afferent ﬁbers (C-ﬁbers)
are activated by mechanosenstive stimuli only after application of bradykinin, hypertonic
KCl or by ischemia (Janig and Koltzenberg, 1991; Haupt,. Jéinig and Kohler, 1983)
suggesting that these are silent C-ﬁbers that are activated by mechanosensory stimuli
during inﬂammation.

From the studies in last chapter, it has been shown that part of the synaptic input
to neurons in pelvic plexus ganglia are purely of peripheral origin. We hypothesize that
the peripheral synaptic inputs are mediated by mechanoreceptors located in the peripheral
organ. The aim of the present study was to determine whether the synaptic inputs
originating from peripheral organs are mediated by mechanoreceptors. The experiments
were performed on a newly designed in vitro pelvic plexus ganglia-colon preparation.
The data show that neurons in pelvic plexus ganglia receive nicotinic cholinergic
synaptic inputs ﬁ'om distal colon. The synaptic inputs are mediated by mechanosensitive
afferent ﬁbers within the colonic-rectal nerves originated ﬁ'om the distal colon. Both

tension and stretch receptors are involved.

Methods
Experiments were performed on adult male guinea-pigs (200-500 gm), euthanized
by exsanguination following carbon dioxide gas-induced anesthetization. Pelvic plexus

ganglia (PG), pelvic nerves, hypogastric nerves, urethra-urinary bladder nerves, colonic-

71

rectal nerves with an attached segment of distal colon-rectum were dissected in situ and
placed in a two compartment organ bath (see Fig. 7). The distal colon (6-10 cm in length,
originating 2 cm orad from the anus) was placed in one compartment of the organ bath
and the PG and attached nerve trunks were placed in the other. The colonic nerves which
connect the colon to PG were draped across a 1 mm thick wall separating the two
compartments and covered by moist strips of tissue paper to prevent dehydration. Both
bath compartments were superfused separately with a modiﬁed Krebs solution containing
(in mM): 117 NaCl, 4.7 KCl, 2.5 CaClz, 1.2 MgClz, 1.2 NaHzPO4, 2.5 NaHCO3 and 11
glucose. Solutions were gassed with 95% 02-5% CO2 and preheated to 37-38°C at the
recording site. In all preparations, pelvic plexus ganglia were pinned onto Sylgard at the
bottom of the chamber.

Measurement of intraluminal colonic pressure. The distal colon-rectum was
cannulated at both ends for distension and measurement of intraluminal colonic/rectal
pressure. The catheter at the orad end of the distal colon-rectum was connected to a
pressure transducer to continuously monitor intraluminal pressure. The catheter at the
aborad end was attached to a Y-valve with one end connected to a calibrated cylindrical
reservoir and the other to a syringe. Basal intraluminal pressures (2-6 cm H20) were set
by the height of Krebs solution in the calibrated reservoir. Colonic distension was
performed by switching the position of the Y-valve from the reservoir to the syringe and
injecting Krebs solution into the colon segment.

Electrophysiological procedures. Intracellular potentials were recorded with
microelectrodes ﬁlled with 3 M- KCl. Tip resistance of the microelectrodes ranged
between 60-110 MQ. Membrane potential was recorded by conventional microelectrode
techniques. An active bridge circuit will allow current injection into neurons through the
recording microelectrodes. Signals from the microelectrode were displayed on an

oscilloscope with digitized memory and recorded on a pen-writing chart recorder.

72

Both mechanical and electrical data (intraluminal pressure, individual electrotonic
potentials, f-EPSPs and action potentials) were stored on a videocassette data recorder
and were plotted on a pen-writing chart recorder.

Drugs were dissolved in Krebs solution and were superﬁised separately in either
the colon-rectal compartment or the ganglia compartment of the organ bath.

Drugs used were hexamethonium bromide (C6, Sigma), dihydro-B-erythroidine
(Sigma), atropine sulfate (Sigma), tetrodotoxin (TTX, Sigma).

Data were expressed as mean i SEM Student’s t test was used for paired

comparisons. p< 0.05 will be considered as signiﬁcant.

Results

Mechanoreceptive synaptic inputs to neurons in pelvic plexus ganglia were
studied in 202 neurons from pelvic plexus ganglia which were attached to a segment of
distal colon-rectum via colonic-rectal nerves. Twenty-seven percent (54 of 202) of
neurons exhibited continuous electrical activities consisting of either f-EPSPs (Fig. 15
panels A and B) or f-EPSPs and action potentials (APs) (Fig. 15, panels C, D, E and F).
In 20% (_11 of 54) of these neurons, the frequency of APs and f-EPSPs and amplitude of
f-EPSPs were increased by distension of the colon segment (Fig. 16) or by applying
punctate mechanoreceptive stimuli (using a glass rod with tip diameter of 2 mm) to the
serosal/longitudinal muscle layer (data not shown). The responses showed adaptation
even when the colon segment was still distended. The majority of neurons (73%, 148 of
202) in pelvic plexus ganglia were electrically silent. In 24% (35 of 148) of these
neurons, passive colonic distension initiated f-EPSPs and APs that rapidly adapted
(compare to the responses in neurons with continuous electrical activities) during the
distending stimulus (Fig. 17). Slow synaptic responses evoked by distal colon distension
were not observed. The properties and distributions of these two types of

mechanosensitive responses are summarized in Table 5.

73

Figure 15. Excitatory synaptic inputs to three neurons (A and B; C and D; E and F) in

pelvic plexus ganglia. Pelvic plexus ganglia were attached to distal colon-rectum via

colonic-rectal nerves.

75

Figure 16. Effect of distension of distal colon on synaptic input to a neuron in pelvic
plexus ganglia. Panel B-E show effect of distension and withdraw of air from distal

colon.

76

.ll BEeﬁBL
D e t. e e E ill

.3 3 - 7 -4- gm

:5 a :28 .6 2m case. a 55:25 .2

 

 

 

 

J

       
     

BEEBL

CE 5
EsceaL Co

lift Ii 'If D ‘FkC—V .D\— II . r» h
1 <

p {I'D Diff In .-
1 jﬁ—‘i ‘ 141 —llll‘4.4 cl 1‘ 411 ill! ‘11 lllul‘l cl:

 

77

nausea deuce 23 “23:38: 3:338: @8365 .«o 3:88 momeoame urge—Sm ”302. 638:2: m

:5: 8e: wovgaxo .U .e:_w§w 3.on 0:32— 5 .850: e 82.: mew—v.83.— HngooebH: .m .8532: 3:28 33:53:: .<

$3.8m 9..qu ":22: a: no.5»: Engage a 8 3209.2 ounam mega: no nomad—3mm. :33 3:: .«o 88mm .2 953...:

E-
Z:

c

o—

o
3T

.00»—
..Illll

7......

.8» ca

Jilin;

 

 

gist}:

 

I

 

 

 

ﬁlzu‘

 

78

Figure 18. Effect of hexamethonium (C 6) (10 uM) on synaptic inputs from distal colon
to a neuron in pelvic plexus ganglia. A, synaptic responses of a neuron in normal Krebs
solution (KS). B, C and D, synaptic inputs to the same neuron S, 10 and 15 minutes aﬁer
superfusion of ganglia with a Krebs-hexamethonium solution (KS/C6, 10 uM). E and F,
recovery of synaptic responses when solution bathing the pelvic plexus ganglia was

returned to normal Krebs solution 10 minutes and 30 minutes, respectively.

79

68¢—

 

:33 mg 5:. cm ....—

 

lbl. .’DD
.5 31' E ’ 'l" VI I DD!» P " LIL! 'PDI’ D’ 'I D brlhlbtht l [g hl l lib! [ll DD

 

iii 11“ A. 41‘ 1411‘
‘ 1‘ 1A. 4“} 1 ‘II‘ : :1 1‘11‘b‘l1 {111‘1111 ‘l‘lJ‘l “ 1 1 1‘ {I ‘i 11‘] 111‘ ‘ 11“.

292 90me SE n- d

$3 ..1..l. ......l......1...ldu.v..l.1...1. —L- ..P H ...L—Ll. F 01.1.3.1... -.. .. 4.0.40.1...» Q79“; 1.- — -—‘l.lh..u1.
:92 00‘8— ...—E A: .U
«I..— lE—PJITPU. ... ... “.11...qu , :L

Enb— oUsmv—la

 

 

 

 

 

 

 

 

 

 

 

.9300 mV— .<

80

The spontaneous electrical activities were reversibly blocked (n=7) by superfusion
of the ganglia compartment of the organ bath with a Krebs solution containing
hexamethonium (C6, 10 - 100 uM) or dihydro-B-erythroidine (10 -50 uM) (Fig. 18). The
data suggest that nicotinic acetylcholine receptors on neurons in pelvic plexus ganglia
were involved in maintaining those spontaneous electrical activities. Also, transaction of
colonic-rectal nerve trunks which connect the colon-rectum with pelvic plexus ganglia
irreversibly abolished spontaneous electrical activities of neurons in pelvic plexus
ganglia, indicating that nerve ﬁbers in colonic-rectal nerves mediated the electrical
activities (n=4) (Fig. 19).

The properties of mechanoreceptive responses evoked in electrically silent
neurons of pelvic plexus ganglia are analde in Figure 20 and 21 and summarized in
Table 5. Figure 20a shows a plot of the relation between instantaneous ﬁring frequency
(1/interval between successive action potentials) and action potential intervals at a
constant distention rate and various constant distension pressures. Figure 20b shows a
plot of the relation between instantaneous ﬁring frequency and action potential intervals
at a constant distension pressure and various constant distension rates. Rate of distension
was calculated as the ratio between the magnitude of initial maximum intraluminal
pressure and the time period needed to reach the maximum intraluminal pressure. The
data show that rapid adaptation appeared within 1/3 of the total response period. Figure
21 shows that neurons exhibited higher maximum instantaneous ﬁring frequencies at
higher intraluminal distending pressure when the distending rate was constant or at faster
rate of distension when distending pressure was constant. For both spontaneously active
and quiescent neurons, the relationship between maximum instantaneous ﬁring frequency
and pressure of distension at a constant distension rate (40 cm HZO/sec) is illustrated in
Figure 22. The data show that compared to quiescent neurons, spontaneously active

neurons were further activated to reach the maximum response at a lower distension

81

pressure. Also, mechanosensitive stimuli evoked higher maximal instantaneous ﬁring

frequency in spontaneously active neurons than in quiescent neurons.

82

Figure 19. Effects of sectioning colonic-rectal nerves on synaptic inputs to a neuron in
pelvic plexus ganglia. Pelvic plexus ganglia were attached to distal colon-rectum via
colonic-rectal nerves. A, B and C, synaptic inputs before sectioning colonic-rectal
nerves. C and D, effect of sectioning some colonic-rectal nerves. E and F, effect of

sectioning additional colonic-rectal nerves.

83

mm-
2:

do... md

E3321;

azuz 38.5225 .2363 .5 J

83oz
_soombesou 3m:

 

(ill/2

”—

84

Figure 20. A: Plot of relation between instantaneous action potential frequency and
action potential interval in four pelvic plexus ganglion neurons at a distension rate of 20
cm HZO/sec (duration, 3 min). Distension pressures: 40 cm H20 (V-V); 20 cm H20 (V-
V); 10 cm H20 (0 - O); 6 cm H20 (0 - O). B: Plot of relation between instantaneous
ﬁring frequency and action potential intervals in ﬁve pelvic plexus ganglion neurons at a
distension pressure of 40 cm H20.(duration, 3 min). Distension rates: 100 cm H20/sec
(O - C); 75 cm HZO/sec (O - O); 60 cm HZO/sec (V - V); 25 cm HZO/sec (V- V); 5 cm
HZO/sec (El - D). Ordinate, instantaneous ﬁring frequency. Abscissa, action potential

interval.

Instantaneous Firing Frequency (Hz)

15

12

 

85

‘QCO
llll
‘QCO
3330
3333

2533-35

Rate-20 en HBO/lea

 

 

 

1516

Action Potential Interval

 

86

Figure 21. Stimulus-response curves for distal colon mechanoreceptors. A: maximum
instantaneous ﬁring frequency plotted against distension rate at four different constant
distension pressures. O - O, 6 cm H20; V- V, 10 cm H20; 0 - C, 15 cm H20; E1 - Cl,
35 cm H20. Ordinate, maximum instantaneous ﬁring frequency. Abscissa, distension
rate. B: maximum instantaneous ﬁring frequency plotted against distension pressure at
three different constant distension rates. V - V, 5 cm HzO/sec; O - O, 40 cm HZO/sec;
O - O, 75 cm HZO/sec. Ordinate, maximum instantaneous ﬁring frequency. Abscissa,

distension pressure.

20

1B

12

20

15

Maximum Instantaneous Firing Frequency (Hz)

10

87

o- o 6cmH20
v- v10cmH20
P C-O15cmH20
a -E.‘55cmH20

. I”///,/”’

 

 

15 30 45 50 75
Distension Rate (cm H30/sec)

o - o 75 cm H20/Sec
O - 0 40 cm Hzo/sec
v - '05 cm HzO/sec

é/i/

/§/§”l’§
9

 

. 1/4 1 1 r

O 10 20 30 4O

 

Pressure (cm H30)

88

Distension Rate=40 cm Hzo/sec

30l-

/-i

251-
20-
15.

10 - {/}
//
1

51- 9/
0 V4 1 I I

O 1 O 20 50 40

Pressure (cm H30)

Maximum Instantaneous Firing Frequency (Hz)

 

Figure 22. Stimulus-response curves for distal colon mechanoreceptors. Maximum

instantaneous ﬁring frequency is plotted against distension pressure for two types of

mechanoreceptors mediated responses to spontaneous active (0-0) and quiescent (V-V)
neurons in pelvic plexus ganglia. Ordinate, maximum instantaneous ﬁring frequency.

Abscissa, distension pressure.

89

.582 3 sum e323 ..2.m.m

 

 

 

 

 

 

 

 

 

 

q: 23. ago 2.3...” 83% a: on ages 88% 8: SR 3282
I835
2: son :23 3. as. see 238 $2.. 5 3.8 as... E is 3832 3.3
coma—oh «3883mm
8": as 3%.... 8.5. .93. Sadness 3:8 .835»: Begin. 82.32
89.3. .9 Seize. 8:83. 93:8.— 38 8.: 8393. s .283. e s .25

i

 

 

4:35am .2383 Maﬁa 5 .383»: o. «€39... oaumgoaoeeﬁus \o @3553. .m 035—.

90

Discussion

The present study shows that fast cholinergic nicotinic synaptic responses can be
further activated or initiated by passive distal colon distension in a portion of neurons
located in the pelvic plexus ganglia. The data demonstrate directly the existence of the
afferent limb of pelvic plexus-pelvic organ reﬂexes involving neurons in pelvic plexus
ganglia and mediated by mechanoreceptors located on the wall of the distal colon.
Nearly 30% of neurons in the pelvic plexus ganglia exhibit continuous spontaneous
electrical activity when a segment of distal colon is attached to the ganglia. In contrast,
as shown in Chapters 1 and 2, spontaneous electrical activity was rarely seen in neurons
of an isolated pelvic plexus ganglia. This can be considered as indirect evidence that
continuous spontaneous electrical activity, recorded from neurons in pelvic plexus
ganglia, is associated with a sensory input from the distal colon. Transecting the colonic-
rectal nerves, which connect the distal colon-rectum with the pelvic plexus ganglia,
abolishes the spontaneous electrical activity in neurons of pelvic plexus ganglia
indicating directly that this activity originates from the distal colon-rectum. This concept
is also supported by the fact that spontaneous activity of neurons in the pelvic plexus
ganglia is reversibly blocked by nicotinic receptor antagonists, applied to the ganglia,
suggesting that the electrical activities are due to synaptic input mediated by cholinergic
nicotinic receptors.

Comparing the response of neurons in the pelvic plexus ganglia to the response of
neurons in prevertebral sympathetic ganglia, several similarities and differences can be
established. Similar to the spontaneously active neurons in the pelvic plexus ganglia,
spontaneously active neurons in the prevertebral sympathetic ganglia receive
mechanosensitive inputs mediated by tension mechanoreceptors. The tension
mechanoreceptors providing synaptic input to neurons in the sympathetic ganglia can be

activated by spontaneous contractions of the smooth muscle layers and exhibit low

91

thresholds to passive distension and no adaptation (Szurszewski and Weems, 1976;
Kreulen and Szurszewski, 1979; Szurszewski and King, 1989). In contrast, the
spontaneously active neurons in pelvic plexus ganglia, respond to passive distension of
the distal colon with tonic discharge of action potentials and exhibit rapid adaptation. It
is possible that continuous spontaneous electrical activity recorded from neurons in pelvic
plexus ganglia is due to the activation of tension mechanoreceptors, while the increase in
electrical activity, evoked by passive distension, represents the summarized response to
activation of both tension and stretch mechanoreceptors. Stretch receptors are
characterized as not spontaneously active, not activated by active contractions of the
smooth muscle and exhibit relatively higher threshold, rapidly adapting discharge of
action potentials (Iggo, 1957). This may explain why spontaneously active neurons show
tonic discharge and adaptation to passive colon distension. The involvement of mixed
mechanoreceptors, both tension as well as stretch receptors can be suggested. Both
tension and stretch receptors have been shown to exist in the gastrointestinal system
(Iggo, 1957; Sengupta and Gebhart, 1994).

The majority of neurons in pelvic plexus ganglia are quiescent and a portion of
them are activated by rapid colon distension. This type of synaptic response rapidly
adapts and correlates positively with both the rate of distension and pressure of
distension. Stretch receptors but not tension receptors are considered to be involved.
Based upon the response of individual primary afferent ﬁbers to increasing levels of
balloon distension, mechanoreceptors in the gastrointestinal tract have been classiﬁed
into three types: low threshold mechanoreceptors (LTM), high threshold
mechanoreceptors (HTM) and wide-dynamic range mechanoreceptors (WDM) (Grundy,
1993). Comparing the stimulus-response curves for esophageal mechanoreceptors
(discharge frequency vs. distension pressure) (Grundy, 1993) to the data for the distal
colon mechanoreceptors shown in the present study (Fig. 22), it can be suggested that

both LTM and WDM are involved in the mechanosensory afferent inputs to

92

spontaneously active neurons and quiescent neurons in pelvic plexus ganglia,
respectively.

Type B mechanosensitive sacral afferents (8 mm/sec) exhibit phasic responses to
distension and type C (1.7 mm/sec) mechanosensitive sacral afferents exhibit a tonic
response to distension. From the data presented in Chapter 2, nerve ﬁbers that mediate
synaptic inputs to neurons in pelvic plexus ganglia are type B and C nerve ﬁbers. This is
consistent with the present study where both tonic and phasic responses to distal colon
distension were observed. The data suggest that mechanosensitive synaptic inputs to
neurons in pelvic plexus ganglia are mediated by both type B and type C afferent ﬁbers.
Whether the type B and type C nerve ﬁbers in peripheral nerve trunks are the same type B
and type C ﬁbers in the sacral afferents needs to be further elucidated. However, studies
in myenteric neurons (Wood, 1989) have shown that two types of mechanosensitive units
behave like visceral mechanoreceptors that discharge in primary afferents. One has the
properties of fast-adapting (phasic) mechanoreceptors and the other behaves like slowly
adapting (tonic) receptors

Unlike inferior mesenteric ganglia, where the majority (70%) of neurons receive
mechanoreceptor mediated synaptic inputs from distal colon (Szurszewski and Weems,
1976), in pelvic plexus ganglia only a small percentage (25%) of neurons receive
mechanosensitive synaptic inputs from distal colon. This may be due to either of the
following possibilities: 1) neurons in pelvic plexus ganglia innervate multiple peripheral
organs with only a small percentage of neurons innervating the distal colon; or 2). the
majority of neurons in the pelvic plexus ganglia are efferent neurons. Neurons in inferior
mesenteric ganglia receive both fast and slow synaptic potentials during colon distension.
The fast synaptic potentials are mediated by nicotinic acetylcholine receptors, while the
slow synaptic responses are noncholinergic. The distension-evoked noncholinergic slow
synaptic responses are partially mediated by substance P (Kreulen and Peters, 1986;

Peters and Kreulen, 1986). Substance P may be released from the axon collaterals of the

93

primary afferent ﬁbers with cell bodies located in dorsal root ganglia or from substance P
containing neurons located in myenteric plexus. In contrast, neurons in pelvic plexus
ganglia respond to colon distension only with fast synaptic potentials mediated by

nicotinic acetylcholine receptors indicating that only cholinergic afferents are involved.

CHAPTER 4
NEURON S IN PELVIC PLEXUS GANGLIA MEDIATE REFLEX
CONTRACTIONS BETWEEN SEGMENTS OF COLON AND
BETWEEN COLON AND URINARY BLADDER.

Introduction
Sacral Parasympathetic Reﬂex Path ways

Continence and defecation are the major functions of the colon and rectum under
normal physiological conditions. These functions are regulated by extrinsic spinal
visceral nerves (lumbar colonic and pelvic nerves) and local intrinsic nerve plexuses
(myenteric and submucous plexus). In humans and experimental animals, the integrity of
the parasympathetic pathways is essential for normal performance of defecation and
continence. The importance of the sacral parasympathetic nerves has been shown in
studies where pelvic nerves or sacral ventral roots were sectioned and the experimental
animals were allowed to recover. The absence of this pathway, following chronic
bilateral section of the pelvic nerves in cats, resulted in dilation of the colon, resembling
megacolon and slowing of gastrointestinal transit (Adamson and Aird, 1932). In humans,
destruction of the sacral ventral roots and pelvic nerves leads to fecal incontinence and to
disturbances in defecation (Marzuk, 1985).

Data from physiological studies show that sacral parasympathetic ﬁbers play an
important role in colonic motility and defecation. Electrical stimulation of sacral ventral
roots causes contractions of both longitudinal and circular muscle layers of the colon and
rectum which often result in defecation. The studies performed in the cat by de Groat
and Krier during the 19703 (de Groat and Krier, 1975, 1976 and 1978) brought 3

Signiﬁcant contribution to understanding the physiology of the large intestine. It was
94

95

suggested that reﬂex activation of the sacral parasympathetic efferent pathway by
distension, electrical stimulation or ﬁictional stimulation leads to activation of primary
afferent ﬁbers and reﬂex ﬁring in sacral parasympathetic efferent ﬁbers. These reﬂexes
enhance colonic motility leading to evacuation of colonic contents. The studies also
demonstrated that the sacral parasympathetic reﬂexes are organized in the sacral spinal
cord.

Based on the available experimental data, the organization of two possible sacral
parasympathetic reﬂex pathways which can modulate colonic motility is illustrated in
Figure 23. Figure 23a shows afferent ﬁbers in colon circular (CM) and longitudinal
muscle (LM) whose cell bodies are located in sacral dorsal root ganglia (DRG). They
synapse with preganglionic efferent neurons in the sacral spinal cord. Preganglionic
cholinergic neurons (CN) in the sacral spinal cord synapse with cholinergic neurons in
the pelvic plexus ganglia. Neurons in the pelvic plexus synapse with the cholinergic
(CN) and nonadrenergic, noncholinergic neurons (N AN C) located in the myenteric
plexus. Fibers from CN and NANC neurons in the myenteric plexus innervate CM and
LM.

Sacral parasympathetic reﬂexes may also be organized in the pelvic plexus
ganglia. Figure 23b shows afferent neurons (darkened circles) and their ﬁbers in colon
circular (CM) and longitudinal muscle (LM) that directly synapse with cholinergic
neurons (CN) in pelvic plexus ganglia. The CN neurons synapse on smooth muscle
and/or with cholinergic neurons located in the intramural plexus of the urinary bladder
and the myenteric plexus of the colon. Fast excitatory cholinergic mechanosensory
afferent inputs originating from distal colon have been demonstrated in neurons in pelvic

plexus ganglia (see Chapter 3).

96

Figure 23. A: Diagrammatic sketch of sacral parasympathetic central reﬂex pathway to
colon. The afferent ﬁbers have their cell bodies located in sacral dorsal root ganglia, and
terminals located in colon-rectum. The preganglionic ﬁbers from the sacral spinal cord
synapse either directly on colon smooth muscle and/or indirectly via myenteric plexus
cholinergic (CN) or nonadrenergic, noncholinergic (NANC) neurons. CM, circular
muscle; LM, longitudinal muscle; MP, myenteric plexus; DRG, dorsal root ganglia.

B: Diagrammatic sketch of sacral parasympathetic reﬂex pathway to colon and
urinary bladder via pelvic plexus ganglia. The afferent neurons located in myenteric
plexus and directly synapse with cholinergic neurons (CN) in pelvic plexus ganglia.
These cholinergic neurons synapse either directly on smooth muscle and/or indirectly
mediated by intrinsic plexus neurons of the urinary bladder or myenteric plexus neurons

of colon. CM, circular muscle; LM, longitudinal muscle; MP, myenteric plexus;

97

    

Pelvic plexus ganglia

 

 

Sacral spinal cord

Pelvic plexus ganglia

 

 

 

Distal colon Urethra-urinary bladder

98

Peripheral Reﬂexes Mediated by Prevertebral Ganglia.

Before the late 19605, prevertebral ganglia were viewed primarily as stations
relaying and distributing incoming central instructions to peripheral destinations. We
know now that this viewpoint is incomplete. In mammals, it has been shown that
autonomic ganglia participate in peripheral reﬂexes, integrate synaptic messages arriving
from various sources and utilize several neurotransmitters (Szurszewski and King, 1989).

The major impetus for renewed interest in the concept of peripheral reﬂex activity
through prevertebral ganglia came from Garry (Garry, 1933, 1934), who showed that
spinal ablation above the level of LI and below L 5, followed by dorsal root rhizotomy of
all the lumbar segments failed to alter colonic movements in the cat to any signiﬁcant
degree. These ﬁndings cast doubt on the involvement of the spinal cord as the principal
site of communication for afferent traffic. It is important to point out that Garry did not
open the abdominal cavity and that his results reﬂected the activity of a normal, dynamic
gut. In these experiments, Garry also showed that removing the inferior mesenteric
ganglia considerably increased colonic motility. Subsequently, it was shown that
extirpation of the inferior mesenteric ganglia in dogs also increased colonic motility
(Lawson, 1934; Lawson and Holt, 1937). Thus the possible involvement of afferent
ﬁbers in peripheral reﬂexes synapsing on ganglion cells in prevertebral ganglia remained
alive.

The experiments by Kuntz and Saccomanna (Kuntz and Saccomanna, 1944)
provide the ﬁrst concrete evidence for peripheral reﬂexes through a prevertebral ganglia.
They found that distension of the distal half of a transacted colon caused inhibition of the
proximal half of the colon in cats, in which the spinal cord from the last thoracic segment
onward was removed, the lumbar sympathetic chain was removed from both sides, the
intermesenteric and hypogastric nerves were cut, and the colon was transected at its

midregion. Semba (Semba, 1954) also showed that an inhibitory intesto-intestinal reﬂex

99

did not depend on the continuity of intrinsic nerve plexuses in the gut nor on the spinal
cord but depended on the integrity of the celiac plexus.

Inhibitory colon-colonic and intesto-intestinal reﬂexes, which are mediated by
mechanoreceptors located in myenteric plexus of the distal colon and the small intestine
and by neurons located in prevertebral sympathetic ganglia (inferior mesenteric ganglia
and celiac ganglia, respectively) have also been directly demonstrated in in vitro colon-
inferior mesenteric ganglia or small intestine-celiac ganglia preparations (Crowcroft,

Holman and Szurszewski, 1971; Kreulen and Szurszewski, 1979b).

Eﬂecm of capsaicin on primary aﬂ'erent ﬁbers.

Capsaicin is the pungent ingredient in red peppers of the genus Capsicum.
Chemically, it is a derivative of vanillyl amide, 8-methyl-N-vanillyl-6-nonenamide and
has a molecular weight of 305.42. Hogyes (1878) was the ﬁrst to state that the pungent
and irritant action of capsicol, an extract of capsaicin, is mainly mediated by sensory
nerves. In 19608, a Hungarian investigator, N. Jancso, discovered that capsaicin, the pure
substance, also exerts a long-term sensory receptor blocking action which can be used as
an important probe to investigate the function of afferent neurons (Jancso, 1960, 1968).
The function of primary afferent neurons is to receive and transmit information from the
internal and external environment and thereby contribute to the organism’s ability to
maintain homeostasis. The cell bodies of primary afferent neurons are located in the
spinal cord (dorsal root) or cranial sensory ganglia and send ﬁbers in both central and
peripheral directions. They contain tachykinins (including substance P) and calcitonin
gene related peptide (CGRP) (Holzer, 1988). The ending of the peripheral ﬁbers may be
either the receptors themselves or connected to special structures, providing the sensory
information. Most of the current knowledge about primary afferent neurons is based on
the application of new experimental techniques and capsaicin, which has proven to be an

important pharmacological tool in sensory neuroscience.

100

Primary afferent ﬁbers have been divided into three groups: (a) thick myelinated
(Aa, B)-ﬁbers have highest conduction velocities and carry non—nociceptive mechanical
information from skin and muscle. (b) thin myelinated (A8)-ﬁbers have intermediate
velocities and carry both nociceptive (mechanonociceptors and polymodal nociceptors)
and nonnociceptive (mechanoreceptors, cold receptors) and (c) thin myelinated (C)-
ﬁbers, have the slowest conduction velocities and are primarily nociceptors (polymodal
nociceptors, chemonociceptors) which respond to noxious mechanical, thermal and/or
chemical stimuli.

Acute capsaicin application depolarizes nonmyelinated (C)-ﬁbers and small
diameter myelinated sensory terminals (A8)-ﬁbers (Heyman and Ran, 1985; Marsh et al.,
1987; Bevan and Forbers, 1988). After the acute application of capsaicin to axons of
sensory neurons, nerve conduction through the treated segment is blocked (Petsche et al.,
1983; Handwerker et al., 1984; Brugger et al., 1990). Chronic administration of
capsaicin to neonatal animals depletes tachykinins and CGRP from sensory ﬁbers whose
cell bodies are located in dorsal root ganglia and further destroys them (Buck and Burks,
1986). Also, systemic administration of capsaicin to adult guinea-pigs produces
extensive axon terminal degeneration in the dorsal spinal cord (J ancso, et al., 1987).
Depletion of tachykinins and CGRP and ultrastructural signs of axon degeneration are
appreciable for at least 1 year (Papka et al., 1984). In contrast, capsaicin does not deplete
substance P and CGRP from enteric neurons in adult animals (I-ones and Barber, 1981).

The mechanism for the action of capsaicin on primary afferent neurons and their
axons has been established. Capsaicin-evoked depolarization involves an increase in
cation conductance (N a+ and Ca2+) though a receptor-mediated mechanism (Marsh et al.,
1988; Docherty, et al., 1991). Membrane depolarization causes further Na+ and Ca2+
inﬂux by activating voltage-dependent sodium and calcium channels (Wood et al., 1988).
The increase in intracellular calcium due to calcium entry, secondarily causes a

prolonged inhibition of voltage-activated calcium currents (Docherty et. al., 1991) and

101

initiates a neurotoxic process (Marsh et al., 1988; Jancso et. al., 1984). Long lasting
inhibition of voltage-activated calcium channels may contribute to the mechanism of
capsaicin inhibition of neurotransmitter release from terminals of primary afferent
neurons and neuronal toxicity may contribute to the degeneration of primary afferent
neurons.

Previous studies have shown that acute capsaicin application in vitro can
selectively “desensitize” primary sensory C ﬁbers (Jancsé, 197 7) and has been shown to
abolish the primary afferent ﬁber mediated noncholinergic (second-phase) smooth
muscle contraction in response to vagus nerve stimulation or electrical ﬁeld stimulation
(BF S) (Undem et al., 1990).

To test the involvement of axon collaterals ﬁ'om primary afferent ﬁbers in reﬂex
contractions between segments of colon and between colon and urinary bladder, acute
capsaicin treatment at the site of pelvic plexus ganglia will be used. Capsaicin will
deplete tachykinins and CGRP from the primary afferent ﬁbers and eliminate the possible
involvement of axon collaterals ﬁom primary afferent in the reﬂex contractions.

Neuropeptides in primary sensory neurons. It has been known that CGRP and
takykinins are putative transmitters of sensory neurons. CGRP is a 37 amino acid
peptide encoded in the calcitonin gene. Its expression is dependent on tissue-speciﬁc
alternative RNA processing (Amara, Jonas, Rosenfeld, Ong and Evans, 1982; Rosenfeld,
Mermod, Amara, Swanson, Sawchenko, Rivier, Vale and Evans, 1983).
Immunohistochemical studies have revealed that CGRP is present throughout the central
and peripheral nervous systems. The action of CGRP depends on the existence of
speciﬁc CGRP receptors in target tissues. Using double-labeling techniques it was
shown that CGRP and substance P coexist in primary afferent neurons. All substance P
containing primary afferent neurons also contain CGRP, but there is also a population of
nerve cells that contains only CGRP but not substance P (Lee, Takarni, Kawai, Girgis,

Hillyard, Macintype, Emson and Tohyarna, 1985; Skoﬁtsch and Jacobowitz, 1986;

102

Gibbins, F umess, Costa, MacIntyre and Girgis, 1985; Gibbins, Fumess and Costa, 1987).
The distribution of CGRP in the central and peripheral nervous system and its co-
localization in some neurons with substance P or acetylcholine suggests several possible
roles in autonomic sensory and motor functions as well as in inﬂammation (Goodman
and Iversen, 1986; Sharkey, 1992). CGRP also has effects in the gastrointestinal system.
In the guinea-pig, CGRP induced contraction of ileal and colonic longitudinal smooth
muscle in a dose—dependent (10-200 nM) manner (Goodman and Iversen, 1986).
Chemically, the tachykinins (TKs) are a family of peptides that share the common
C-terminal sequence, Phe-X-Gly-Leu-Met-NHZ.(Ersparner, 1981). In mammals, at least
three different TKs namely substance P (SP), neurokinin A(NKA) and neurokinin B
(NKB) are likely neurotransmitters both in central and peripheral nervous system (Maggi,
1990). Three neurokinin receptor subtypes (N K-l, NK-2 and NK-3) have been cloned,
sequenced and pharmacologically characterized. Although each of the endogenous
tachykinins can stimulate all of the receptors, there is a selectivity of SP, NKA and NKB
for NK-l, NK-2 and NK-3 receptors, respectively (Maggi, 1990). NK—1 receptors are
present on neurons, smooth muscle cells and neuroendocrine cells. NK—2 receptors occur
primarily on smooth muscle cells while NK-3 receptors are located primarily on neurons
(Buck and Butcher, 1986; Buck, Pruss, Krstenansky, Robinson and Stauderrnan, 1988;
Yau and Youther, 1982). Tachykinin-induced contractile responses through the
activation of tachykinin receptors have been shown in dog and guinea-pig ileum (Chahl,
1982; Shuttleworth, Sanders and Keef, 1993; Daniel, Parish, Watson, Fox-Threlkld,
Regoli and Rainsford, 1995). SP- and NKB-induced release of acetylcholine has also
been shown in guinea-pig small intestine (Yau and Youther, 1982; Yau, Mandel, Dorsett
and Youther, 1992). In autonomic ganglia, SP and NKB induce membrane
depolarization, as it was shown in bronchial parasympathetic ganglia (Myers and Undem,

1993). SP also induces membrane depolarization, associated with an increase in input

103
resistance in guinea-pig inferior mesenteric ganglia (Dun and Minota, 1981; Krier and
Szurszewski, 1982).

Since neurons in pelvic plexus ganglia receive mechanoreceptor mediated
synaptic inputs from peripheral organs as shown in Chapter 3, we hypothesize that
neurons in pelvic plexus ganglia mediate visceral organ reﬂexes. The aim of the present
section of the study is to determine whether pelvic plexus ganglia and nerve ﬁbers which
connect the pelvic plexus ganglia with peripheral organs (distal colon and urethra-urinary
bladder) can mediate reﬂexes between colon segments and between colon and urethra-
urinary bladder. We also aim to determine, if capsaicin-sensitive primary afferent axon
collateral ﬁbers are involved in these peripheral reﬂexes. The data suggest that neurons
in pelvic plexus ganglia are able to mediate excitatory reﬂex responses between two
isolated segment of distal colon and between distal colon and urinary bladder. These
peripheral reﬂexes do not involve capsaicin sensitive axon collaterals from primary

afferent ﬁbers.

Methods

Experiments were performed on adult male guinea-pigs (200-500 gm), euthanized
by exsanguination following carbon dioxide gas-induced anesthesia. Pelvic plexus
ganglia (PG), pelvic nerves, hypogastric nerves, urethra-urinary bladder nerves, colonic-
rectal nerves with an attached segment of distal colon-rectum and/or urethra-urinary
bladder were dissected in situ and placed in a two or three compartment organ bath.

Measurement of intraluminal colonic pressure and urinary bladder pressure.
The distal colon-rectum was canulated at both ends for distension and measurement of
intraluminal colonic/rectal pressures. The catheter at the orad end of the distal colon-
rectum was connected to a pressure transducer to continuously monitor intraluminal
pressures. The catheter at the aborad end was attached to a Y-valve, with one end

connected to a calibrated cylindrical reservoir and the other to a syringe. Basal

104

intraluminal pressures (2-6 cm H20) were set by the height of Krebs solution (KS) in the
calibrated reservoir. Colonic distension was accomplished by switching the position of
the Y-valve from the reservoir to the syringe and injecting Krebs solution into the colon
segment. The urinary bladder was cannulated through the urethra for either distension or
measurement of intraluminal pressure. The pelvic plexus ganglia were pinned to the
ﬂoor of the organ bath. All bath compartments were perfused separately with a modiﬁed
Krebs solution at 37 °C, bubbled with 95% 02,, 5% C02, of the following composition
(mM): 117 NaCl, 4.7 KCl, 2.5 CaClZ, 1.2 MgClz, 1.2 NaHZPO4, 2.5 NaHCO3 and 11
glucose.

Experiments to study reﬂexes between segments of distal colon and between
colon and urinary bladder. Intraluminal pressure recording techniques were used to
study in vitro the contractile responses of the colon and urinary bladder. In these
experiments, two isolated segments of the colon or one segment of the colon and urinary
bladder were attached by peripheral nerve trunks (colonic nerves, urethra/urinary bladder
nerves; see Fig. 24 a, b) to the pelvic plexus ganglia. In the experiments where colon-
colonic reﬂexes were studied, the in vitro preparation was placed in a three compartment
organ bath. One colon segment was placed in one compartment, the other segment and
pelvic plexus ganglia were placed separately in the other two compartments. In the
experiments where colon-urinary bladder reﬂexes were studied, the colon segment and
pelvic plexus ganglia were placed in one compartment of the organ bath while the
urethra-urinary bladder was placed in the other. The nerve ﬁbers which connect the
peripheral organs and pelvic plexus ganglia were draped across a 1 mm thick wall
separating the compartments and covered by moist strips of tissue paper to prevent
dehydration. In experiments where the reﬂexes were studied by direct activation of the
peripheral nerve ﬁbers, stimulating electrodes were positioned on colonic nerves that
connected one segment of the isolated distal colon with pelvic plexus ganglia while

intraluminal colonic pressure was recorded from another distal colon segment to study

105

Figure 24. Diagrammatic sketch of in vitro preparations. A, Experiment to study
reﬂexes between two segments of colon and B, reﬂexes between the colon and urinary
bladder. (A), Two isolated segments of colon are connected to pelvic plexus ganglia by
colonic-rectal nerves. (B), Colon and urinary bladder are connected to pelvic plexus

ganglia via colonic-rectal nerves and urethra-urinary bladder nerves respectively.

106

 

107

the colon-colonic reﬂexes. To study colon-minary bladder reﬂexes, the stimulating
electrodes were positioned on colonic-rectal nerves and intraltuninal pressure in the
urinary bladder was recorded, or stimulating electrodes were positioned on urethra-
urinary bladder nerves while the intraluminal colonic pressure was measured. Electrical
stimulation was applied at frequencies ranging between 0.5 to 20 Hz (duration 0.5 ms at
supramaximal intensities). Intraluminal pressure was recorded from the other colon
segment or urinary bladder. Passive distension of the colon occurred in increments of 1
ml to maximum volume of 6 ml. Frequency dependent contractile response curves were
constructed by expressing the contractile responses at different frequencies as a
percentage of the maximal contractile responses to pelvic nerve stimulation.

Drugs were administered by separate superfusion to either the colon
compartment, urethra-urinary bladder compartment or the ganglia compartment of the in
vitro organ bath.

Data for mechanical activity was stored on a videocassette data recorder and the
output (intraluminal pressure) was plotted on a pen-writing chart recorder.

Data were expressed as mean i SEM. Student’s t test was used for paired

comparison. p<0.05 was considered as signiﬁcant.

Results
Passive distension of one colonic segment or electrical stimulation of its colonic
nerves (3-20 Hz at supramaximal voltage) evoked contractile responses of the other
colonic segment (Figure 25A, B) and the urinary bladder (Figure 26). Electrical
stimulation of urethra-urinary bladder nerves (5- 20 Hz) also evoked colon contractile
responses (Figure 27). The contractile responses were frequency dependent (Fig. 28, 29,
30) and were abolished by: 1) sectioning the colonic nerves that connect the pelvic plexus

ganglia with colon segment which had its intraluminal pressure recorded or the urethra-

108

Figure 25. Contractile responses of distal colon segment during passive distension of the
proximal colon segment by injecting 3.5 ml of Krebs solution into the lumen (A) and
electrical stimulation of colonic nerve ﬁbers attached to proximal colon segment (B) at 10
Hz (a), 15 Hz (b) and 20 Hz. (C): contractile responses of distal colon segment evoked
by electrical stimulation (20 Hz, 0.5 ms pulse duration, supramaximal voltage for 45 sec)
of colonic nerves attached to proximal colon segment in the absence (left column) and
presence (right column) of tetrodotoxin (TTX) (a), hexamethonium (C 6) (b), atropine (c)

and after sectioning the colonic nerves ((1).

109

3.5 ml

”met.

10 Hz

c KS Control

20 Hz

b t
W” "Tm

20 Hz

15 Hz

10
C
ch H20
0

"r——t

20 Hz 40 sec

TTXluM

e e
20 Hz

C5 100 HM
W

O O
20 Hz

Atropine 1 uM
W
0 o
20 Hz

Colonic Nerve
Transectron 5

110

Figure 26. Contractile responses of urinary bladder during passive distension of colon-
rectum by injecting of a 5 ml Krebs solution into the lumen (a) and electrical stimulation
of colonic-rectal nerves at frequency of 10 Hz (b, c, d). Left panel represents the urinary
bladder contractile responses in a normal Krebs solution. Right panel represents the
urinary bladder contractile responses during superfusion of the urinary bladder with a
Krebs solution containing tetrodotoxin (TTX, 1 uM) (a, c) and atropine (1 uM) (b) and
during superfusion of pelvic plexus ganglia with a Krebs solution containing
hexamethonium (C 6, 100 nM) (d).

 

a Control

W

t
b
O
10 Hz
c
o
10 Hz
d
O
10 Hz

Colonic/Rectal N.

TTX 1 [AM 15 min

W

l Sml i

.—-I
10 sec

Atropine 1 uM

WW

0
10 Hz

TTXIuM

O
10 Hz

C6 100 NM

Icm H20
0

10 Hz

50 Sec.

112

Figure 27. A: Contractile responses of colon-rectum in normal Krebs solution evoked by
electrical stimulation of urethra-urinary bladder nerves at frequency of 20 Hz.

B: Contractile responses of colon-rectum when electrical stimulation of
urethra-urinary bladder nerves during superfusion of the colon-rectum with a Krebs
solution containing tetrodotoxin (TTX, 1 uM) (top trace), atropine (1 uM) (bottom trace)
and during superfusion of the pelvic plexus ganglia with a Krebs solution containing

hexamethonium (C 6, 100 nM) (middle trace).

113

KS Control B TTX 1 11M

W W

20 Hz 20 Hz

. C6 100 uM.
O O o O
20 Hz 20 Hz

Atropine 1 uM

LMA'MVKNJL. ml 520 ‘

o o o
20 Hz 20 Hz

 

114

Figure 28. Effects of electrical stimulation of colonic nerve ﬁbers originating from
proximal colon segment on contractile responses recorded from distal colon segment (0.5
ms pulse duration, supramaximal voltage for 45 sec). A and B, frequency dependent
contractile responses of distal colon segment during electrical stimulation of pelvic
nerves and nerve ﬁbers originating from proximal colon segment at indicated stimulating
frequencies, respectively. Time period between two ﬁlled circles shows time of electrical
stimulation. C: Frequency response curves of distal colon segment contractile responses.
I, represent pelvic nerve stimulation. Cl, represent the urethra-urinary bladder nerves
stimulation. Ordinate, contractile responses expressed as a percentage of maximal pelvic
nerve-evoked contractions. Abscissa, stimulation frequency. Values are mean i- S. E.

Mean, n=4.

115

Anzv Eco—.02... 5.3.55.3
cu m p o— n o

T q d 4 O

 

 

O

nuodeoa umurjxnn OMON awed jo ebngueaud

 

 

 

a: 8 n: n. N= 2 a: m E n Nz m E . 5. W...
O o 0 o 0
w W 73> at 5272» .3.4.7)</
2 z 2%.. m
n0: ONO N: n- ”I O- ON: n. ON: m0

 

116

Figure 29. Effects of electrical stimulation of urethra-urinary bladder nerves on
contractile responses recorded from distal colon segment (0.5 ms pulse duration,
supramaximal voltage for 30 sec). Time period between two ﬁlled circles shows time of
electrical stimulation. A and B, frequency dependent contractile responses of colon-
rectum during electrical stimulation of urethra-urinary bladder nerves and pelvic nerves
at indicated stimulating frequencies, respectively. (B). C: Frequency response curves of
distal colon-rectum contractile responses. 0, represents pelvic nerve stimulation. 0,
represents urethra-urinary bladder nerve stimulation. Ordinate, contractile responses
expressed as a percentage of maximal pelvic nerve-evoked contractions. Abscissa,

stimulation frequency. Values are mean i S. E. Mean, n=5.

117

3: Flat... Saleem

 

 

 

nqm o.~ m..— o._ 07 oo .6
w
.8 »
ion ”
m
..on W
100
\ LON m
x. m
8
.\*\‘\‘ 10¢ m
Pco. u
'DIBOIO
:01. U
.00me
1
":8 n: n— 3.0— O— NI” RIM NH—m N=- Nzo n6
O 00 00 00
m_.. J.
O m
:—
.Zu.>_o.— my
N02“— N02:—

 

.z .82... b2...2§.§: <

118

Figure 30. Effects of electrical stimulation of colonic-rectal nerves on contractile
responses recorded from urinary bladder (0.5 ms pulse duration, supramaximal voltage
for 10 see). A and B, frequency dependent contractile responses of urinary bladder
during electrical stimulation of colonic-rectal nerves and pelvic nerves at indicated
stimulating frequencies, respectively. C: Frequency response curve of urinary bladder
contractile responses. Y, represents pelvic nerve stimulation. V, represents colonic-rectal
nerve stimulation. Ordinate, contractile responses expressed as a percentage of maximal
pelvic nerve-evoked contractions. Abscissa, stimulation frequency. Values are mean i

S. E. Mean, n=5.

9.
:1:
o

o
0
IS Hz

w .l.
i i

O
8Hz

ill.

0
3Hz

~49

A ColonidReclal N.

NLAM
5 Hz
B Pelvic N

all—r
0.5 Hz

25
5080c.

119

DD
ll
DP

 

5...

gssassesae

L

15

25

1
20

1
5

 

O
O

eeuodug wnwpum we" awed go ebozueaed

Stimulation Frequency (Hz)

120

urinary bladder nerves that connected the pelvic plexus ganglia with urethra-urinary
bladder which had its intraluminal pressure recorded, 2) superﬁrsion of pelvic plexus
ganglia with Krebs solution containing hexamethonium (10-100 uM), 3) superfusion of
colon or urinary bladder with Krebs solution containing tetrodotoxin (1 nM) and 4)
atropine (1 uM) (Fig. 25C, 26, 27) These data indicated that the contractions were
neurogenic, mediated by ﬁbers in CN and UUBN and involved nicotinic acetylcholine
receptors and muscarinic receptors.

Effects of Acute Capsaicin Treatment. Effects of capsaicin on colon contraction
evoked by electrical stimulation of urethra-urinary bladder nerve orcolonic nerves which
attached to one of the two separated segments of the distal colon were studied. Capsaicin
was used to determine if axon collaterals from primary afferents in pelvic and hypogastric
nerve, which are components of the colonic and urethra-urinary bladder nerves, were
involved in peripheral nerve stimulation-evoked colon contraction. The protocol used for
acute "desensitization" of primary afferent ﬁbers with cell bodies in the dorsal root
ganglia involved treatment with 10 uM capsaicin for 45 min, followed by a 60 min
washout period. Acute capsaicin "desensitization" was used to abolish the responses
mediated by primary afferents but not by SP and CGRP-containing intramural neurons
(Myers and Undem, 1991). Superfusion of pelvic plexus ganglia with a Krebs solution
containing capsaicin (10 nM, 45-65 min, followed by 60-120 min wash) had no effect on
electrical stimulation (8-15 Hz, 0.5 ms duration, supramaximal voltage) of distal colonic
nerve-evoked contraction of the proximal segment of distal colon (n=2) and on colon
contraction evoked by electrical stimulation of urethra-urinary bladder nerves (n=3) (Fig.
31). The data suggest that capsaicin-sensitive primary afferent ﬁbers with cell bodies
located in dorsal root ganglia, were not involved in these electrically-induced

contractions.

121

 

 

120.-
0
g 1001- ,§
0
a.
03
o?
.. 80-
O
b
C
8
‘_ 50r-
O
...:
C
0
g 40%
O
a.
20; . i . . J . I... .
Control 5 20 35 so 65 (min) 2hr

wash

Figure 31. Effect of capsaicin (10 nM) on urethra-urinary bladder nerves (15 Hz, 35 sec.)
evoked contractile responses of guinea-pi g distal colon segment when superfused on
pelvic plexus ganglia only, in vitro. Ordinate, percent of control contractile responses of

distal colon. Abscissa, time period of capsaicin application.

122

Discussion

The present study shows that in preparations where peripheral organs were
connected to each other via only the pelvic plexus ganglia and peripheral nerve trunks,
electrical stimulation of nerve ﬁbers connecting one segment of distal colon with pelvic
plexus ganglia or distension of that colon segment evoked contractile responses of the
other colon segment. Also, electrical stimulation of colonic nerves or distention of a
colonic segment evokes contractile responses of urinary bladder. Furthermore, electrical
stimulation of urethra-urinary bladder nerve evokes distal colon contraction. The reﬂex
responses are blocked by peripheral nerve trunk transection, superfusion of
hexamethonium on pelvic plexus ganglia and tetrodotoxin and atropine on the peripheral
organ. This indicates that the excitatory reﬂexes between colon segments and between
colon and urethra-urinary bladder are neurogenic, mediated through peripheral nerve
ﬁbers and pelvic plexus ganglia by cholinergic nicotinic and muscarinic receptors. An
inhibitory colon-colonic reﬂex mediated by neurons in inferior mesenteric ganglia has
previously been shown (Kreulen and Szurszewski, 1979b). It is known that the majority
of neurons in pelvic plexus ganglia are cholinergic (Yokota and Burnstock, 1983) while
the majority of neurons in inferior mesenteric ganglia are adrenergic (Szurszewski and
Weems, 1976). Generally, afferent input to neurons in inferior mesenteric ganglia will
increase sympathetic discharge to the colon which has been shown to decrease colonic
contraction while afferent input to neurons in pelvic plexus ganglia will increase
cholinergic discharge to the colon which has been shown to increase colonic contraction.
This provides an explanation of the fact that neurons in pelvic plexus ganglia mediate
excitatory reﬂexes and neurons in inferior mesenteric ganglia mediate inhibitory reﬂexes.
The data also demonstrate that peripheral reﬂexes can be realized not only through
prevertebral sympathetic ganglia but also through parasympathetic ganglia. In normal

defecation and continence process or in patients with spinal cord transection, the

123

existence of this reﬂex pathway may be physiologically or pathologically signiﬁcant.
Under physiological condition, two general reﬂex pathways have been described. One of
the reﬂex pathways is mediated through central nervous system at spinal cord level. The
efferent limb of this reﬂex pathway involves both sympathetic, parasympathetic and
somatic nervous system. This reﬂex pathway involves sending information to the brain.
Another reﬂex pathway is locally deﬁned. The reﬂex center is located on the wall of the
intestine (mucosal and submucosal ganglia). This reﬂex is also called peristaltic reﬂex
(Costa and Furness, 1976). The autonomic ganglia mediated reﬂex may be the third
reﬂex pathway that mediates normal defecation and continence. Excitatory reﬂexes
mediated by pelvic plexus ganglia may contribute to defecation while the inhibitory
reﬂex mediated by inferior mesenteric ganglia may contribute to continence. Under
pathological conditions, e. g. in spinal cord transacted patients, the reﬂex through central
nervous system is abolished, defecation and continence are mainly dependent upon local
peristaltic reﬂex and autonomic ganglia mediated reﬂex. Pelvic plexus ganglia mediated
excitatory reﬂex may play a role in helping the defecation in these patients. Also from
the inter organ reﬂex experiments, it has been shown that pelvic plexus ganglia mediate
excitatory reﬂex between distal colon and urinary bladder. Under physiological
conditions, the defecation reﬂex will initiate the micturition reﬂex and micturition reﬂex
will also initiate the defecation reﬂex. The level of this type of reﬂex has been shown to
be at the spinal cord (de Groat, Nadelhaft, Milne, Booth, Morgan and Thor, 1981). Based
on the results of the present studies, we speculate that pelvic plexus ganglia may also
involve in simultaneous micturition and defecation reﬂexes.

Although the general function of inferior mesenteric ganglia and pelvic plexus
ganglia is to mediate inhibitory and excitatory reﬂexes, respectively, it is possible that
opposite function of these ganglia also be obtained under certain conditions. It has been
shown that when the colon smooth muscle tone is low, stimulation of postganglionic

sympathetic ﬁbers induces contraction of the circular layer of colon smooth muscle. This

124

response is mediated by co-release of noradrenaline and ATP or a related purine
nucleotide ﬁ'om post ganglionic sympathetic nerve ﬁbers part of which originate from
inferior mesenteric ganglia, acting on postjunctional al-adrenoreceptors and P2X-
purinoceptors (V enkova and Krier, 1993, 1994). When colon smooth muscle tone is
high, stimulation of postganglionic parasympathetic nerve ﬁbers induced relaxation of
colon smooth muscle. This relaxation is mediated by nonadrenergic, noncholinergic
(NANC) enteric neurons (Zeitlin, Johnson, Fasth, Hulten and Eaglesom, 1978).
Cholinergic postganglionic parasympathetic nerve ﬁbers originated from pelvic plexus
ganglia is one source of activation of the NAN C inhibitory neurons. Although inferior
mesenteric ganglia and pelvic plexus ganglia have been demonstrated to mediate
excitatory and inhibitory reﬂexes, respectively when colon or intestine are under normal
muscle tone , I speculate that these reﬂexes can also be altered when the smooth muscle
condition is different.

Acute capsaicin treatment depletes tachykinins (including substance P) and
calcitonin gene-related peptides from primary afferent terminals (Myers and Undem,
1991) in the hypogastric and pelvic nerves which may give axon collaterals to pelvic
plexus ganglia. The data show that acute capsaicin treatment of pelvic plexus ganglia has
no effect on nerve stimulation-evoked excitatory reﬂex responses, suggesting that axon
collaterals from primary afferent ﬁbers are not involved in pelvic plexus ganglia mediated
peripheral reﬂexes. The present data also in accordance with data presented in Chapter 3,
which show no slow synaptic potentials in pelvic plexus ganglion neurons during passive
distension of the distal colon segment. If capsaicin-sensitive primary afferent ﬁbers were
involved in the afferent arc of reﬂexes, tachykinins released from the capsaicin sensitive
primary afferent terminals during distension should have caused slow membrane
depolarization, as shown in neurons of inferior mesenteric ganglia (Kreulen and Peters,

1986). Application of tachykinins including NKB and substance P have also been shown

125

to depolarize inferior mesenteric ganglion neurons and bronchial parasympathetic
ganglion neurons (Dun and Minota, 1981; Myers and Undem, 1993).

Since the central nervous system is also a source of excitatory synaptic input to
pelvic plexus ganglion neurons through the pelvic and hypogastric nerves, it is possible
that pelvic plexus ganglia is capable of modifying the level of mechanoreceptor input it
receives ﬁom the distal colon by integrating the peripheral input with input from the
central nervous system.

In summary, the present studies show that excitatory reﬂexes between colon
segments and between the colon and the urinary bladder can be mediated by neurons in
pelvic plexus ganglia. The peripheral reﬂexes are mediated by nicotinic and muscarinic
cholinergic neurotransmission. Axon collaterals from capsaicin-sensitive primary

afferent ﬁbers are not involved in these pathways.

CHAPTER 5
THE INTERACTION OF INFLAMIVIATORY MEDIATORS WITH
PELVIC PARASYMPATHETIC SYSTEM.

Introduction
Inﬂammation of the gastrointestinal system.

Tissue injury or infection is accompanied by a complex series of homeostatic
reactions, involving the immune, circulatory and nervous system, that is termed
inﬂammation. In cutaneous tissues, the processes of inﬂammation lead to four well
pronounced clinical signs: heat, redness, edema and pain (Gauldie, 1991). These
responses are caused by vasodilation, plasma extravasation and sensitization and
activation of nociceptive C-ﬁber nerve endings (Sharkey, 1992). It is also well known
that visceral inﬂammation is associated with persistent pain and dramatic gastrointestinal
motility changes as a consequence of an exaggerated reﬂex activity. Enhanced
sensitivity of the viscera in inﬂammation has been well documented. Mechanical stimuli
such as light pressure or distension do not elicit any sensation or reﬂex when applied to
the healthy mucosa of gastrointestinal or urinary tract. However, when these tissues
become chronically inﬂamed, their sensitivity markedly increases and formerly
unnoticeable stimuli can elicit considerable pain, discomfort and reﬂex motility changes .
(Janig and Koltzenburg, 1990).

The development of prolonged hyperalgesia and hyperreﬂexia are prominent
ﬁndings of the inﬂammatory process in visceral organs resulting in the presence of
discomfort, exaggerated perception of pain, abdominal cramps, dysuria, urgency,
increased frequency of micturition and frequent bouts of severe diarrhea. Thus, visceral

tissues can become the source of pain and on-going tenderness during inﬂammation,
126

127

although they are largely insensitive to a wide variety of stimuli under normal non-
inﬂamed conditions (Mclellan and Goodell, 1943; Wolf, 1965).

Spinal visceral afferents supplying the colon and the urinary bladder are involved
in neural reﬂex regulation of continence and evacuation and in perception of different
sensations occurring under physiological and pathological conditions (de Groat,
Nadelhaft, Milne, Booth, Morgan and Thor, 1981; Haupt, Jéinig and Kohler, 1983;
Habler, Janig and Koltzenburg, 1988). How visceral primary afferents encode peripheral
events and how central neurons decode these messages leading to appropriate reﬂex
patterns and sensations have been studied both under physiological and pathological
conditions. Distension and contraction of colon and urinary bladder elicit autonomic
reﬂexes and sensations in humans and animals. The sensations are generally diffuse and
consist of feelings of fullness and urgency to void at low intraluminal pressures and are
gradually supplanted by discomfort and pain when pressures exceed physiological values.
Pathological changes during inﬂammation, may elicit spontaneous pain that is different
in character from visceral pain elicited under physiological conditions (Janig,
Koltzenburg, 1989; Jinig and Morrison, 1986).

For neurophysiological investigations, people have studied the visceral sensory
mechanism under pathological conditions. To stimulate an acute inﬂammatory process,
algesic chemicals, such as bradykinin or capsaicin have been applied topically onto
afferent endings or have been injected intra-arterially into the vascular supply of the
organ (Bahns, Ernsberger, Janig and Nelke, 1986; Haupt, Janig and Kohler, 1983). These
substances activate the majority of mechanosensitive afferents that also respond to
distension of the organ. To investigate the neural substrates of visceral inﬂammatory
hyperalgesia and hyperreﬂexia, intraluminal injection of chemical irritants such as
mustard or turpentine oil was performed (Habler, Janig and Koltzenburg, 1988, 1989).
Mechanosensitive vesical afferents are vigorously excited by either irritant. Both

inﬂammatory algesic chemicals and inﬂammatory chemical irritants produced

128

hyperalgesia and hyperreﬂexia that resembled the inﬂammatory disease of visceral
organs.

Inﬂammatory bowel diseases (IBD) (Crohn's disease and ulcerative colitis) are a
group of chronic illnesses characterized by periods of active disease followed by
remission and relapse with symptoms of bouts of diarrhea, cramping of the abdomen and
fever. The IBD responses include inﬁltration of plasma cells, mast cells, macrophages
and lymphocytes into the mucosa and, although it is a chronic disease, a prominent acute
phase involving local granulocyte immigration (Kirsner and Shorter, 1988) also occurs.
The original observation that ganglion cells were increased in the enteric nervous system
in IBD, established a neural involvement in the diseases (Sharkey, 1992). In addition to
the neural hyperplasia found in some tissues, necrosis, degeneration of ganglion cells and
axons have also been observed (Koch, Sonnenberg and Carney, 1991). These
pathological abnormalities appear to be dependent on the disease (Crohn's vs. ulcerative
colitis), the region of the intestinal wall (mucosa vs. muscle) and whether or not the tissue
was from a site of active disease (Koch, Sonnenberg and Carney, 1991). Intensive
studies involving measurements of neurotransmitter content in the gut wall of IBD
patients have prompted a systemic study of the impact of inﬂammation on
neurotransmitter content and release in the myenteric plexus in animal models. The
studies have shown that inﬂammation decreases acetylcholine release from myenteric
terminals in the inﬂammatory intestine of rats (Collins, Hurst, Main, Stanley, Khan,
Blennerhassett and Swain, 1992). Furthermore, the inhibition of release of acetylcholine
is mediated by the inﬂammatory mediator, interleukin-10. Increased production of
interleukin-1 occurs in patients with active IBD (Ligumsky, Simon, Karrneli and
Rachemilewitz, 1990) and interleukin-1 is expressed early during the course of
experimental colitis in rabbits (Cominelli, Nast, Clark, Schindler, Lierena, Eysselein,

Thompson and Dinarello, 1990) where it may play a protective role (Cominelli, Nast,

129

Lierena, Dinarello and Zipser, 1990). These ﬁndings prompted an evaluation of the
ability of IL-1 [3 to inﬂuence neurotransmitter content and release in rat myenteric plexus.

The studies in the previous chapters have shown that neurons in pelvic plexus
ganglia are able to mediate peripheral visceral organ reﬂexes. It is important to study this
reﬂex not only under physiological conditions but also under pathological conditions
such as inflammation. From the broad spectrum of inﬂammatory mediators, bradykinin
and interleukin-113 were selected in the present study because of: 1) their prominent
effects as inﬂammatory mediators; 2) the speciﬁc effects of bradykinin on
mechanosensitive receptors located in the visceral organs; 3) the important role that
interleukin-113 plays as a communicator between immune system and nervous system

during inﬂammation.

Actions of bradykinin on visceral afferent ﬁbers.

Bradykinin (BK) is a nonapeptide which is released from ischemic and inﬂamed
tissues (Keele and Armstrong, 1964). Normally, blood and tissue levels of BK are very
low. Upon injury, a cascade of biochemical reactions is initiated, which results in the
proteolytic generation of BK and a further kinin (kallidin; lys-BK) from high molecular
weight precursors, the kininogens, found in blood and tissue (Miller, 1987). The actions
of BK are mediated through at least two different types of receptor designated as BKl
and BK2 (Regoli, Rhaleb, Dion and Drapeau, 1990). Most biological actions of BK seem
to be mediated by BKZ-receptors. BK can elicit either contraction or relaxation of
vascular smooth muscle depending on the animal species and site of action. BK also
produces contraction of smooth muscles of the respiratory and gastrointestinal tract and
the uterus. In several species exogenously applied kinins have been shown to preduce
the cardinal symptoms of inﬂammation (Wihelrn, 1973). BK is also called a
vasoneuroactive agent because it causes vasodilation and increases capillary permeability

leading to edema (Guzman et a1, 1962; Handwerker, 1980; Lim et a1, 1962). BK was

130

found to release prostaglandins and leukotrienes through activation of phospholipase A2
(Griesbacher and Lembeck, 1987). Exciting nociceptors and stimulating nociceptive
afferent nerves, BK mediates pain and hyperalgesia (Juan and Lembeck, 1974; Steranka,
Manning, de Haas, F erkany, Borosky, Connor, Vavrek, Stewart, and Snyder, 1988).

Bradykinin is known to be one of the most potent algesic substances formed in
the body and is a potent inﬂammatory mediator in inducing acute inﬂammation. The
hyperalgesic and pro-inﬂammatory activity of BK is either due to a direct action of BK
on a subpopulation of nociceptive primary afferent nerve terminals through the activation
of a receptor-coupled enzyme phospholipase C (Haupt et al., 1983; Miller, 1987;
Mizumura et al., 1990) or indirectly through BK-induced release of other inﬂammatory
mediators including cytokines (F erreira, Lorenzetti, Cunha and Poole, 1993). In turn,
cytokines further induce the release of hyperalgesic mediators including BK. Activation
of phospholipase C results in the production of two intracellular messengers, inositol
1,4,5-triphosphate (1P3) which stimulates intracellular calcium release and diacylglycerol
(DAG) which activates protein kinase C. Protein kinase C phosphorylates membrane ion
channel proteins to induce changes in ion permeability (Kaczrnarek, 1987). The action of
BK on visceral afferent ﬁbers may also be due to the release of prostaglandins
(Kumazawa and Mizumura, 1980). For example, prostaglandin E2 sensitizes primary
afferent terminals to BK (Mense, 1977; Handwerker, 1976; Chahl and Iggo, 1977; Ohno
et al., 1984; Mizumura et a1, 1987). The pain and inﬂammation induced by bradykinin
are blocked by aspirin or other acidic non-steroidal anti-inﬂammatory drugs, which are
known to inhibit cyclooxygenase and to suppress the formation of prostaglandins and
thromboxanes. These data indicate that bradykinin produces pain and inﬂammation in
part through the formation of prostaglandins (Katori, Hori, Uchida, Tanaka and Harada,
1986)

Although BK is used as a nociceptive stimulus (Guilbaud et al., 1976), it does not

stimulate only nociceptors. BK also excites afferents that most likely are not involved in

131

nociception. For example, BK stimulates pulmonary and cardiac vagal C-ﬁbers
(Kaufman et al., 1980 a, b), vagal afferents of the duodenum (Cottrell and Iggo, 1984),
and cutaneous low-threshold mechanoreceptors (Beck and Handwerker, 1974).
Regarding the BK action on sensory afferents, there are two possibilities: 1) it can act
directly on chemosensitive nerve endings (Baker et al., 1980; Cervero and Sharkey, 1988;
Kauﬁnan et al., 1980b; Longhurst et al., 1984) or 2) it can stimulate pure
mechanoreceptors secondary to muscle contraction (Floyd et al., 1977; Kaufman et al.,
1980a). However, it has been shown that all colonic distension-sensitive
mechanoreceptors can be stimulated by a direct action of BK on nerve endings (Haupt et
al., 1983). BK-induced depolarization of primary afferent nerve terminals has been
shown in neonatal rats. The depolarization is due to an increase in sodium permeability
of the membrane, which is mediated via the activation of PKC and DAG (Dunn and
Rang, 1990).

Application of bradykinin to rat dorsal root ganglion neurons (Burgess, Mullaney,
McNeil], Dunn and Rang, 1989), guinea-pig nodose ganglia (Undem and Weinreich,
1993) and hamster submandibular ganglion neurons (Suzuki, 1992) caused a
depolarization associated with an inward current and an increase in membrane
conductance that was probably due to the opening of the sodium channels. Like the
mechanism involved in depolarization of the primary afferent terminals, opening of the
sodium channels is mediated by PKC and DAG (Burgess, Mullaney, McNeill, Dunn and
Rang, 1989). No hyperpolarization or outward current was detected. In dorsal root
ganglion neurons, bradykinin also increases calcium inﬂux and intracellular calcium
concentration (Burgess, Mullaney, McNeill, Dunn and Rang, 1989) and cyclic-GMP
concentration (McGehee, Goy and Oxford, 1992). The activation of guanylate cyclase by
bradykinin is mediated by the inﬂux of calcium (Burgess, Mullaney, McNeill, Coote,
Minhas and Wood, 1989). The inﬂux of calcium is nifedipine sensitive indicating that L-

type voltage activated calcium channels are involved. Furthermore, bradykinin

132

stimulates phosphoinositide hydrolysis and mobilization of arachidonic acid in dorsal
root ganglion neurons (Gammon, Allen and Morell, 1989). Elevated cyclic GMP
inhibited bradykinin-induced formation of inositol phosphates, pointing a possible role of
cyclic GMP in the regulation of polyphosphoinositide turnover in dorsal root ganglion
neurons (Burgess, Mullaney, McNeil], Coote, Minhas and Wood, 1989).

The action of bradykinin on smooth muscle was characterized in guinea-pig
taenia ceci. The bradykinin response was characterized by an initial hyperpolarization
and suppression of spike activity followed by a sustained depolarization and an increased
spike activity. Changes in membrane activity were accompanied by inhibition of phasic
contractions followed by an increase in muscle tone and the development of phasic
contractions. Membrane conductance was decreased during the sustained depolarization.
The multiple actions of bradykinin are mediated via BK2 receptors and involve: 1)
calcium mobilization associated with activation of potassium channels; 2) calcium
release ﬁ'om intracellular stores and 3) receptor-activated sodium channels (Hertog,
Nelemans and den Akker, 1988).

The development of tachyphylaxis or desensitization to BK is also a well-
recognized phenomenon. The pain evoked by application of BK on blister base skin
diminishes in intensity with repeated applications (Elliott et al., 1960). Similar
tachyphylaxis has also been observed in cardiac sympathetic afferents of the cat (Baker,
Coleridge, Coleridge and Nerdrumet, 1980). It was suggested (McGehee, Goy and
Oxford, 1992) that a nitric oxide-cyclic GMP pathway is involved in the mechanism of
desensitization to BK at the receptor or G-protein level because: 1) desensitization to BK
was enhanced by nitroprusside (NP) and reduced by nitric oxide synthase (NOS)
inhibitors; 2) the effects of NOS inhibitors were overcome by 8 Br-cyclic GMP or L-
arginine; and 3) 8 Br-cyclic GMP-evoked modiﬁcation of desensitization to BK required

receptor occupancy.

133

Bradykinin will be used in the experiments involved in this section of the
dissertation, since it is known to activate mechanosensitive primary afferent ﬁbers and to
exert pro-inﬂammatory effects in the gastrointestinal system. The effects of BK on
mechanoreceptor mediated responses in the in vitro distal colon-pelvic plexus ganglia
preparation will be studied to elucidate if BK affects the afferent arc of the peripheral
reﬂexes, which are mediated by neurons in pelvic plexus ganglia. The data show that
application of bradykinin to the distal colon initiates and/or potentiates the
mechanosensitive synaptic inputs to neurons in the pelvic plexus ganglia. The action of
BK is due to either a direct effect on afferent terminals or to secondary activation of

mechanoreceptors located in the distal colon by bradykinin-induced colon contractions .

Inﬂammatory mediator interleukin-1 ,8

At a conference held in the summer of 1984 in London, Guido Majno pointed out
some 100 molecules, including about 50 lymphokines, which could claim to mediate
some aspect of the inﬂammatory process. By the year 2000, this number would be
considerably greater if the current rate of discovery were maintained (Majno, 1985). Ovef
the last few years, evidence for an ordered control of the inﬂammatory process has begun
to emerge in the sense that inﬂammation is one of the integral parts of the body defense
mechanisms wherein the "50" or so lymphokines exercise considerable inﬂuence. Majno
used the term lymphokine in the way others use cytokine, to deﬁne polypeptide
mediators, released ﬁ'om a variety of cell types (not just lymphocytes), which activate,
modulate and control various aspects of body defense and repair. Biologists had long
suspected the existence of an overall control process for activating and co-ordinating the
body defense which now appears to involve the synthesis and release .of a number of
cytokines, particularly interleukin 1 (IL-1) (Billingham, 1987).

Historically, IL-1 was “born” in 1972 during the course of in vitro studies by

Gery and Waksman (Gery and Waksman, 1972). It was described as a necessary factor

134

for thymocyte and lymphocyte proliferation, and was given the title of “lymphocyte
activation factor”. Around the same time, other research groups were working on factors
which activated other aspects of the inﬂammatory and defense mechanisms. Wood and
coworkers were working on a factor necessary for B cell proliferation and antibody
production, which they termed B cell activation factor (Wood and Gaul, 1974). Atkins,
among others, was attempting to purify endogenous pyrogen (Bemheirn, Block and
Atkins, 1979), and Karnpschmidt was working with leucocyte endogenous mediator
(LEM) which activates the acute phase reaction, particularly the increased hepatic
synthesis of acute phase plasma proteins (Kampschmidt, 1981). Other factors discovered
in the mid to late-19708 like mononuclear cell factor (MCF) and catabolin (Dayer,
Robinson and Krane, 1977; Dingle, Saklatvala, Hembry, Tyler, Fell and Jubb, 1979),
possessed the ability to stimulate proteolytic enzymes and prostaglandins. The
difficulties surrounding the attempts to physically separate endogenous pyrogen from
LEM, lead to the suggestion that they might be the same molecule. Following the
sharing of puriﬁed products between the various research groups, it rapidly became
apparent that they were all studying essentially the same molecule. The various
synonyms became included in the general title interleukin 1 in a specially convened
workshop in 1979 (Aarden, Brunner and Cerottini, 1979), differentiating these properties
from interleukin 2, which is a T cell product and is a growth factor for other
lymphocytes.

It is now known that IL-1 is a 17.5 killodalton polypeptide synthesized and
released by blood activated macrophages, endothelial cells, astrocytes and vascular
smooth muscle cells during infection, injury, stress or antigenic challenge (Dinarello,
1985; 1988). It is a prominent member of a group of polypeptide mediators called
cytokines. IL-1 , initially synthesized as a 31 KD precursor that is subsequently processed
to a 17.5 KD molecule, exists in two forms, interleukin-la and interleukin-113. There are

26% amino acid homology between the two forms which are encoded by separate genes

135

on chromosome 2. Interleukin-1a remains cell associated while interleukin-113 is
secreted by blood monocytes/macrophages, T-lymphocytes, ﬁbroblasts, endothelial cells,
epithelial cells, adrenal chromafﬁn cells, astrocytes, skin keratinocytes and smooth
muscle cells. Synthesis of interleukin-113 mRNA is predominant over interleukin-1a in
rodent and human cells (Dinarello, 1988). Two subtypes of interleukin-1 receptor have
been cloned and are called type I and type II (Sims, March, Cosman, Widmer,
MacDonald, McMahan, Grubin, Wignall, Jackson and Call, 1988). The type I receptor is
a 80 KD glycoprotein and the type II receptor is a 68 KD glycoprotein. Both types have
two B-pleated sheets that are connected by disulﬁde bonds, which places them into the
irnmunoglobulin superfamily. There is 28% homology in extracellular domains between
the two receptor types. The type I receptor has 213 amino acid residues in its
cytoplasmic domain while the type II receptor has only 29 amino acid residues in its
cytoplasmic domain. These data strongly indicate that only the type I receptor is capable
of transducing a signal and it can produce all of the biological effects attributed to IL-1
(Stylianou, O’Neill, Rawlinson, Edbrooke, Woo and Saklatvala, 1992; Sims, Gayle,
Slack, Alderson, Bird, Giri, Colotta, Re, Mantovani and Shanebeck, 1993).

Following the binding of IL-1 to its receptor, two pathways for signal
transduction have been proposed, (i): cyclic-AMP activation of protein kinase A; and (ii):
phosphatidylcholine hydrolysis (Mizel, 1989). The inﬂammatory properties of IL-1 have
been shown to induce the release of interleukin-1 itself, other cytokines, prostaglandin
E2, collagenase and phospholipase A2 from immune cells, endothelial cells and smooth
muscle cells and induce immune cell proliferation (Mizel, 1989). It also acts as a potent
chemoattractant for leukocytes into the sites of active inﬂammation and involves tissue
repair by increasing ﬁbroblast proliferation and collagen synthesis (Eastgate, Symons and
Duff, 1988).

A naturally occurring inhibitor of the biological activities of IL-1 has been
puriﬁed and prepared by Synergen Inc. It is a 25 KD protein with 26% homology to IL-

136

115 and 19% homology to IL-la. It does not interact with IL-loc or IL-18, but binds to
type I IL-1 receptor with an afﬁnity approximately equal to IL-lot or IL-18. This protein
is called IL-1 receptor antagonist (IL-Ira) (Hannum, Wilcox, Arend, Joslin and Dripps,
1990; Eisenberg, Evan, Arend, Verderber, Brewer, Hannum and Thompson, 1990). The
current evidence indicates that the inhibitory action of IL-lra results from competition
with IL—la or IL-18 to the type I receptor and that the binding of IL-lra to this receptor
does not result in signal transduction (Dripps, Verderber, Ng, Thompson and Eisenberg,
1991; Granowitz, Clark, Mancilla and Dinarello, 1991). It has been shown that IL-lra is
released in vivo during inﬂammation and during the natural course of many diseases
(Dinarello and Thompson, 1991). Administered experimentally, IL-lra has been
demonstrated to block IL-1 activity both in vitro and in vivo. In rabbits, pretreatment
with IL-lra has been shown to prevent death resulting from septic shock (Ohlsson, Bj ork,
Bergenfeldt, Hageman and Thompson, 1990), to prevent the development of immune
complex-induced colitis (Cominell, Nast, Clark, Schindler, Lierena, Eysselein,
Thompson and Dinarello, 1990) and to block cerebrospinal ﬂuid inﬂammation induced
by intracerebroventricular administration of IL-1 (Rarnilo, Mustafa, Porter, Saez-Llorens,
Mertsola, Olsen, Luby, Beutler and McCracken, 1990). Currently, pre-clinical and
clinical studies are being carried out on the possible therapeutic uses of IL-lra in the
treatments of sepsis, cachexia, rheumatoid arthritis and chronic myelogenous leukemia.
The nervous and immune systems consist of complex networks of cells that
monitor speciﬁc signals and respond in a speciﬁc manner to these signals. The nervous
system, through the production of neuroregulators (neurotransmitters, neuromodulators
and neuropeptides), can regulate speciﬁc immune system functions. The immune system
produces a variety of chemical factors that exert speciﬁc irnmunoregulatory actions.
These immunoregulators (irnmunomodulators and irnmunopeptides, which are substances

with either stimulating or inhibiting activities on certain functions of the immune system)

137

also possess speciﬁc neuromodulatory activities. This establishes a bi-directional
communication between the nervous and immune systems.

In the central nervous system, brain intrinsic and blood-derived macrophages
(Jordan and Thomas, 1988), cerebrovascular endothelial cells (Billingham, 1987),
microglia (Giulian, Allen, Baker and Tomozawa, 1986), astrocytes millingham, 1987)
and even neurons (Breder, Dinarello and Saper, 1989) have the ability to synthesize and
release IL-lB in response to appropriate stimuli. IL-1 [3 causes fever production
(Nakashima, Hori, Mori, Kuriyama and Kizuno, 1989), suppresses food intake
(McCarthy, Kluger and Vander, 1986; Mrosovsky, Molony, Conn and Kluger, 1989;
Plata-Salaman, Oomura and Kai, 1988) and alters slow wave sleep (Dinarello, 1985). In
hypothalamic brain slices, IL-lB increases or decreases neuron ﬁring of glucosensitive
and thermosensitive neurons and depolarizes or hyperpolarizes supraoptic neurons
(Nakashima, Hori, Mori, Kuriyama and Mizuno, 1989; Kuriyama, .Hori, Mori and
Nakashima, 1990; Li, Inenage, Kawano, Kannan and Yamashita, 1992). The effects of
IL-18 on thermosensitive neurons were blocked by sodium salicylate, an aspirin like drug
and a cyclooxygenase inhibitor which is known to block prostaglandin E synthesis from
arachidonic acid but were not blocked by naloxone (Hori et al., 1988). In rat
hippocampal neurons, IL-lB inhibits long term potentiation (Bellinger, Madamba and
Siggins, 1993) and depresses a voltage dependent calcium current through protein kinase
C signal transduction pathway (Palat-Salamén and Ffrench-Mullen, 1992; 1994). The
action of IL-1 B on primary afferents has been known to generally and critically involve in
hyperalgesia induced by inﬂammatory processes (F erreira, Lorenzetti, Cunha and Poole,
1993; Watkins, Wiertelak, Goehler, Smith, Martin and Maier, 1994).

In the gastrointestinal tract, IL-lB inhibits basal and evoked release of
norepinephrine and acetylcholine ﬁom myenteric nerve terminals (Collins, Hurst, Main,

Stanley, Khan, Blennerhassett and Swain, 1992; Hurst and Collins, 1993; Main,

138

Blennerhassett and Collins, 1993). Its actions on the extrinsic autonomic neurons that
innervate the gastrointestinal tract are yet not known.

The aim of the present study was to determine the effects of interleukin-113 on the
active and passive membrane properties and synaptic transmission of neurons in guinea-
pig pelvic plexus ganglia. These neurons are the origin of postganglionic ﬁbers that
innervate vascular and visceral smooth muscle of colon, anal canal, urethra, urinary
bladder and reproductive organs. They receive central synaptic inputs from ﬁbers in
pelvic and hypogastric nerves and peripheral synaptic inputs from ﬁbers in urethra-
urinary bladder and colonic-rectal nerves (Lin and Krier, 1993). Utilizing in vitro
intracellular microelectrode techniques, we show that IL-10 causes membrane
depolarization, membrane hyperpolarization, blockade of orthodromic action potentials
and depression of fast excitatory postsynaptic potentials (f-EPSPs) in a portion of the
neurons tested in the pelvic plexus ganglia. The blockade of orthodromic action
potentials and the depression of f-EPSPs by IL-lB may be due to inhibition of

acetylcholine release from presynaptic terminals.

Methods
Experiments designed to study the effects of bradykinin on mechanosensitive input
from distal colon to neurons in pelvic plexus ganglia.

Experiments were performed on adult male guinea-pigs (200-5 00 gm), euthanized
by exsanguination following carbon dioxide gas-induced anesthesia. Pelvic plexus
ganglia (PG), pelvic nerves, hypogastric nerves, urethra-urinary bladder nerves, colonic-
rectal nerves with an attached segment of distal colon-rectum were dissected in situ and
placed in a two compartment organ bath (see Fig. 7). The distal colon (6-10 cm in length,
originating 2 cm orad from the anus) was placed in one compartment of the organ bath
and the PG and attached nerve trunks were placed in the other. The colonic nerves which

connect the colon to PG were draped across a 1 mm thick wall separating the two

139

compartments and covered by moist strips of tissue paper to prevent dehydration. Both
bath compartments were superﬁrsed separately with a modiﬁed Krebs solution containing
(in mM): 117 NaCl, 4.7 KCl, 2.5 CaClz, 1.2 MgClz, 1.2 NaHva 2.5 NaHCO3 and 11
glucose. Solutions were gassed with 95% 02-5% CO2 and preheated to 37-38°C at the
recording site. In all preparations, pelvic plexus ganglia were pinned onto Sylgard at the
bottom of the chamber.

Measurement of intraluminal colonic pressure. The distal colon-rectum was
canulated at both ends for distension and measurement of intralmninal colonic/rectal
pressure. The catheter at the orad end of the distal colon-rectum was connected to a
pressure transducer to continuously monitor intraluminal pressure. The catheter at the
aborad end was attached to a Y-valve with one end connected to a calibrated cylindrical
reservoir and the other to a syringe. Basal intraluminal pressures (2—6 cm H20) were set
by the height of Krebs solution (KS) in the calibrated reservoir. Colonic distension was
made by switching the position of Y-valve from the reservoir to the syringe and injecting
KS into the colon segment.

Drugs was administered by superfusion separately to either the colon-rectal
compartment or to the ganglia compartment of the organ bath.

Electrophysiological procedures. Transmembrane potentials were recorded with
microelectrode ﬁlled with 3 M- KCl. Tip resistance of microelectrodes ranged between
60-110 M9. Membrane potential was recorded by conventional microelectrode
techniques. An active bridge circuit will allow current injection into neurons through the
recording microelectrodes. Signals from the microelectrode was displayed on an
oscilloscope with digitized memory and recorded on a pen-writing chart recorder.

Both mechanical and electrical data were stored on a videocassette data recorder
and the output (intraluminal pressure, individual electrotonic potentials, f-EPSPs and

action potentials) was plotted on a pen-writing chart recorder.

140

Bradykinin (1-10 nM) was superfused over the colon. Colonic-rectal pressure
was monitored continuously during the period of intracellular recording from neurons in
pelvic plexus ganglia. The effect of active colon-rectum contraction and passive
distension of distal colon-rectum on the electrical activities of PG neurons before and

during superfusion of bradykinin were studied.

Experiments designed to test the effects of human recombinant interleukin-M on
neurons in pelvic plexus ganglia.

Experiments were performed on male guinea-pigs weighing 200-500 gm.
Animals were euthanized by exsanguination following a sharp blow to the head. Pelvic
plexus ganglia with attached central (pelvic and hypogastric) and peripheral (urethra-
urinary bladder) nerves were dissected in situ and placed in an organ bath. Pelvic plexus
ganglia were pinned onto Sylgard at the bottom of a chamber (0.5 ml) and continuously
superﬁrsed with a Krebs solution (2-3 ml/min) containing (in mM): 117 NaCl, 4.7 KCl,
2.5 CaClz, 1.2 MgClz, 1.2 NaHzPO4, 25 NaHCO3 and 11 glucose. Solutions were gassed
with 95% 02-5% C02 and preheated to 36-37 °C at the recording site.

Microelectrodes were ﬁlled with 3 M KCl. The tip resistances of microelectrodes
ranged from 25 to 60 MO. Membrane potentials were recorded by conventional
microelectrode techniques. An active bridge circuit allowed current injection into
neurons through a recording microelectrode. Membrane current was recorded by a
discontinuous single-electrode voltage clamp method with an Axoclarnp 2A (Axon
Instruments) ampliﬁer. Sampling frequency ranged between 3 and 5 kHz with a 70-30
duty cycle. The reversal potentials were estimated from IN plots using a least squares
linear regression analysis. Signals ﬁom the microelectrode were displayed on an
oscilloscope, recorded on a pen-writing chart recorder and stored on a videocassette data

recorder.

141

Fast excitatory postsynaptic potentials (f-EPSPs) and associated action potentials
were evoked by electrical stimulation of ﬁbers in pelvic, hypogastric or urethra-urinary
bladder nerves (0.3 - 0.5 ms duration, frequency 0.3 - 0.5 Hz). Acetylcholine (100 - 200
[AM in Krebs solution) was pressure ejected (20 psi for 10 - 250 ms) onto the surface of
neurons from a micropipette (tip diameter ~ 5 - 10 um). Pressure application of Ach
caused a fast membrane depolarization (fast Ach potential) and slow membrane
depolarization (slow Ach potential). Fast Ach potentials were blocked by
hexamethonium (10 - 100 nM), indicating that they are mediated by nicotinic receptors.
Slow Ach potentials were abolished by atropine (300 nM), indicating that they are
mediated by muscarinic receptors. Membrane input resistance was determined by
measuring the steady state voltage evoked by injection of hyperpolarizing current across
the cell membrane (30 - 200 ms duration) and calculated as the ratio between the
magnitude of steady state voltage and hyperpolarizing current. Data are expressed as
mean :t SE.

Human recombinant interleukin-113 (hrIL-lB, 1 ug/ml in 0.1% of bovine serum
albumin) was dissolved in Dulbecco's phosphate buffer sterile solution. It was applied by
either superfusing the ganglia with a Krebs solution containing hrIL-lB (20 - 80 ng/ml) or
by pressure ejection (l ug/ml, 20 psi for 200 - 1200 ms) onto the surface of neurons.
Other drugs were dissolved directly into the Krebs solution and applied by superfusion.
A speciﬁc IL-1 receptor antagonist was used to test the speciﬁcity of hrIL-lB evoked
responses. Bovine serum albumin (BSA, 0.1%) had no effect on active and passive
membrane properties of neurons or on synaptic transmission.

Drugs used were human recombinant interleukin-113 (hrIL-lB) (R & D system,
Inc.), human recombinant interleukin-1 receptor antagonist (hrIL-lra) (Synergen 'Inc.),
bovine serum albumin (B SA), tetraethylammonium chloride (TEA), tetrodotoxin (TTX),
co-conotoxin GVIA, yohimbine hydrochloride, hexamethonium bromide, atropine sulfate,

acetylcholine chloride and naloxone hydrochloride (Sigma).

142

Data were expressed as mean i SEM.

Results
Eﬂ‘ects of bradykinin on colon contraction and mechanoreceptor mediated synaptic
input to neurons in pelvic plexus ganglia.

Superfusion of the distal colon segment with Krebs solution containing
bradykinin (1-10 nM, 20 min) induced a tonic contraction. Associated with the colon
contraction, f-EPSPs and action potentials were initiated or the frequency of spontaneous
f-EPSPs and action potentials and the amplitude of f-EPSPs were increased in pelvic
plexus ganglion neurons in electrically quiescent (2 of 12 preparations) or with minimum
spontaneous electrical activities (2 of 12 preparations) (Fig. 32). The remainder neurons
had no response to BK application. In the four neurons that responded to colon
application of BK, passive distension of the colon-rectum initiated or increased electrical
activities (Fig. 33) while before the application of bradykinin, passive colonic distension
did not increase electrical activities in these neurons. These data suggest that bradykinin
sensitized the mechanosensory input from distal colon to neurons in pelvic plexus
ganglia. This effect may have been due to a direct action of bradykinin on
mechanosensory terminals of peripheral afferent ﬁbers as well as an indirect activation of
mechanoreceptors by stimulation of smooth muscle contraction. Also, the existence of
polymodal receptors which are sensitive both to mechanical stimuli and chemostimuli,
can not be excluded. When bradykinin (10 nM) was superﬁlsed on pelvic plexus ganglia
only (n=3), it had no effect on membrane potential and membrane input resistance of the
neurons, indicating that increased electrical activities of pelvic plexus ganglion neurons
was not due to a direct effect of bradykinin on these neurons. The lack of response of
pelvic plexus ganglion neurons to BK may be due to the lack of BK receptors on these

neurons.

143

Figure 32. Effects of bradykinin (10 nM) on spontaneous colon contractions and
electrical activities of one neuron in pelvic plexus ganglia. In each panel, top traces
represent intracolonic pressures and bottom traces represent electrical activities recorded
from one neuron in pelvic plexus ganglia. A, Intracolonic pressure and electrical
activities of a neuron when the colon was superfused with a normal Krebs Solution (KS).
B, C, Intracolonic pressure and electrical activities of the same neuron during superfusion

of the colon with a Krebs solution containing bradykinin (10 nM).

 

144

 

 

 

 

 

 

 

 

 

 

 

   

2.- TI. - . l 4... r l i
:33... :- - z _ _ _. _
>5
0 IIIIll-III
2o 2 5518.63.55 n u
- _ . _ . - i l
o llllll ll
oazHEo
n

2: 2 5553.5. .5... m m

.8390 m! <

145

Figure 33. Effect of colon-rectum distension on synaptic inputs (fast excitatory
postsynaptic potentials) of a neuron in pelvic plexus ganglia before (A), during (B, C) and
after (D) bradykinin (10 nM) was superfused on colon-rectum. A, effect of colon-rectum
distension on synaptic potentials when colon-rectum and pelvic plexus ganglia were
superfused with a normal Krebs solution. B and C, effect of colon-rectum distension on
, synaptic potentials of the same neuron when colon-rectum was superfused with a Krebs
solution containing bradykinin (10 nM) for 6 min and 8 min, respectively. D, effect of
colon-rectum distension on synaptic potentials in the same neuron when colon-rectum

was superfused with a normal Krebs solution for 30 min.

146

A KS Control

 

atoll-1,0
0

 

Lg. ‘ A— A“ _A

B 6 min. KS/Bradykinin 10 nM

 

 

 

1 Lmllt 1W
C 8 min. KS/Bradykinin 10 nM

 

 

mm! m MWLIJJ m I u u

D 30 min KS Wash

 

 

 

147

Effects of human recombinant interleukin-M on membrane potential and membrane
input resistance

Pressure or bath application of human recombinant interleukin-113 (hrIL-IB)
caused membrane depolarization (54%; 36 of 66) and membrane hyperpolarization (30%;
20 of 66) of neurons tested in guinea-pig pelvic plexus ganglia. HrIL-IB had no effect on
the remainder of cells tested.

Depolarization In 28 of the neurons responding with membrane depolarization,
hrIL-IB (1 ug/ml) was applied by pressure injection (200-1200 ms, 20 psi). In these
neurons, the membrane depolarization was associated with a decrease in membrane input
resistance (46.0 d: 4.5%) (Fig. 34A, B). HrIL-lB-evoked depolarizations (1 ug/ml, 20 psi,
900 ms) had a mean time to peak of 1.2 :1: 0.2 min, a mean peak amplitude of 11.8 :t 1.9
mV and a mean duration of 4.6 :t 0.9 min. The effect of pressure-injected IL-lB was
dose-dependent (Fig. 35A). Pressure application of hrIL-IB (1 ug/ml, 600 ms duration,
20 psi, at holding potentials of -56 to -69 mV) also caused inward currents with a mean
time to peak, peak amplitude and duration of 0.9 d: 0.2 min, 440 i 72 pA and 4.1 i 1.5
min, respectively (Figure 34C). In 8 of the neurons responding with membrane
depolarization, hrIL-lB was applied by superfusion (20 - 80 ng/ml; 5 min.). At a
concentration of 40 ng/ml, membrane depolarizations had a mean time to peak of 2.7 :1:
0.7 min, a mean peak amplitude of 14.2 :1: 2.9 mV and a mean duration of 9.6 :t 2.8 min.

Hyperpolarization Pressure application of hrIL-lB (1 ug/ml, 600 ms duration, 20
psi) evoked membrane hyperpolarization with a mean time to peak, peak amplitude and
duration of 22.1 :t 8.8 sec, 8.2 d: 0.9 mV and 1.2 :1: 0.3 min, respectively. HrIL-lB-evoked
membrane hyperpolarization was associated with an increase in membrane input
resistance (31.1 i: 1.2%) (Fig. 36A, B). The hyperpolarization of membrane potential by
hrIL-lB was dose dependent (Fig. 35B). Pressure application of hrIL-lB (1 ug/ml, 500
ms duration, 20 psi, at holding potentials of -48 to -72 mV) also caused an outward

membrane current with a mean time to peak, peak amplitude and duration of 18.3 :h 3.3

148

Figure 34. Effects of human recombinant interleukin-113 (hrIL-lB) on resting membrane
potential, membrane input resistance and membrane current of two neurons in pelvic
plexus ganglia. A: continuous recording of membrane potential from one neuron. Top
and bottom traces, membrane potential and injected current through a microelectrode,
respectively. Resting membrane potential was -60 mV. HrIL-lB was pressure ejected (1
pig/ml, 300 ms duration, 20 psi.). Membrane input resistance, calculated as the ratio
between the amplitude of steady state voltage of hyperpolarizing pulse and current
injected into the cell decreased (b, c, d) during the hrIL-lB-evoked membrane
depolarization. B: expanded records of hyperpolarizing electronic potentials. Records a-
e were obtained at the time marked by respective letters in A. C: continuous recording of
membrane current from another neuron. Inward membrane current was evoked by
pressure ejection of hrIL-lB (1 ug/ml, 600 ms duration, 20 psi.). Holding potential was -

58 mV (resting membrane potential).

 

149

 

 

 

 

 

 

I 252

<Q8mHll 11: ll lllllllllllllH lllllll will!” 1‘ l l |l\..

>5 mm:

as OH U

>5 S H |> l: |\l|/. >

o e o n a
m
.on cm
I
o
E éééééggggéég
N O nah—mu..—

4

8- ..

.. .... .... _______________ ______=______:___ .2 _s . _ : =_:______........ .
o e e
>5 0

150

Figure 35. Effects of human recombinant interleukin-18 (hrIL-IB) on resting membrane
potential. Ordinate, change in membrane potential amplitude. Abscissa, duration of
pressure ejection of hrIL-IB (l ug/ml, 20 psi.), expressed in ms. A: depolarization effect
of hrIL-IB. B: hyperpolarization effect of hrIL-lB. Vertical bars and number above

symbols are standard error of mean and number of cells tested, respectively.

Change In Membrane Potential Amplitude

14

12

10

 

...
N

i
d

151

Depolarization

 

 

 

(5)
(8)
(6)
(6)
(3)
Hyperpolarization
(a) (3) (5)
(5) ___..
/i’§ é
(2)/é
l
(4)
(5)
(6)
Q
200 :00 GOLD 830 10100 72.00

Duration (ms)

152

Figure 36. Effects of human recombinant interleukin-113 (hrIL-IB) on resting membrane
potential, membrane input resistance and membrane current of two neurons in pelvic
plexus ganglia. A: continuous recording of membrane potential ﬁ'om one neuron. Top
and bottom traces, membrane potential and injected current through a microelectrode,
respectively. Resting membrane potential was -66 mV. I-IrIL-IB was pressure ejected (1
jig/ml, 600 ms duration, 20 psi.). Membrane input resistance was calculated as the ratio
between the amplitude of steady state voltage of hyperpolarizing pulse and current
injected into the cell. Membrane input resistance increased (b) during the hrIL-lB-evoked
hyperpolarization. B: expanded records of hyperpolarizing electronic potentials. Records
a—c were obtained at the time marked by respective letters in A. C: continuous recording
of membrane current from another neuron. Outward membrane current was evoked by
pressure ejection of hrIL-IB (1 ug/ml, 600 ms duration, 20 psi.). Holding potential was -

64 mV (resting membrane potential).

153

.oom o. _ 4
I

.8 8o H l I Illlnlhﬂiﬂr .12... e-

 

 

mEO—

 

 

 

 

s. .3 7 rl H rl l H _lu

Eag> \z/ll l\\\\i\(l/\ln

 

m

.03. ca

Tl.

S. no H 5:::::::_:_::__::::__::___:_:::=:_::_::_z:_::_:___:_:::__==_=::_=::::::::E
. 95:.

 

>EO_H

154

sec, 0.7 :t 0.2 nA and 1.0 :t 0.5 min respectively (Fig. 36 C). The voltage sensitivity of
hrIL-l B-evoked outward currents was tested by clamping the neuron to various potentials
ranging from -10 to -100 mV and measuring the amplitude of the current. HrIL-IB-
induced outward current was decreased by depolarization and increased by
hyperpolarization of the membrane potential (Fig. 37). The extrapolated reversal
potential was -5 i- 4.5 mV (n=3).

Membrane potential, depolarization, hyperpolarization and associated changes in
membrane input resistance evoked by hrIL-lB were not altered by hexamethonium (100
nM), atropine (500 nM), yohimbine (300 nM), naloxone (1 nM), (n-conotoxin (GVIA,
300 - 700 nM), tetrodotoxin (TTX, l - 5 nM) or tetraethylammonium (TBA, 30 - 80 mM)
(n=3 for each agent) (Fig. 38).

Tachyphylaxis to the hrIL-lB-evoked membrane depolarization occurred after the
ﬁrst bath (20 - 80 ng/ml, 3 - 6 min) or pressure application (1 pg/ml, 200 - 1200 ms, 20
psi). Subsequent membrane responses to hr-IL-IB were either smaller in magnitude or
there were no effects (n=28). Tachyphylaxis was more pronounced for the hr-IL-lB-

evoked membrane depolarization than membrane hyperpolarization.

Eﬂ‘ect of hrIL-Iﬂ on Synaptic Transmission

Orthodromic action potentials and fast excitatory postsynaptic potentials (f-
EPSPs). Stimulation of ﬁbers in pelvic, hypogastric and urethra-urinary bladder nerves
at supramaximal intensities evoked f-EPSPs and orthodromic action potentials that were
reversibly blocked by superfusing the. ganglia with a low calcium (0.05 mM), high
magnesium (30 mM) Krebs solution (n=3) or a Krebs solution containing
hexamethonium (10 - 100 nM, n=3) or dihydro-B-erythroidine (5 - 100 nM, ni3). In
contrast, atropine (300 - 500 nM) did not alter orthodromic action potentials and f-EPSPs.

This indicates that the responses were mediated via nicotinic acetylcholine receptors.

155

.>E m T 33 33:83 3338 cam—aubxo 2C. .3: can .me cNV 3.39 2:22.” 3388 a
3 380.5 mm? GER: 5 32:5 .mzmqaw mas—n 320m B :83: a 88m 3232 R: 35.56 805 .3 33:65.0. ow§o>

Ba 3 ”3:23 wags. ﬁesta a 355° 3&8 canes 33:5 £93055 “assess 85: .2 05mm

.03 2 a. 8a 3.52
I 4

_ozcuuom ocuanoz <9 com ...—... llllllll >E a.
93 a can 3... 8| 8.. 8... .
. /. . ql . . id ﬁe

 

 

 

 

 

 

..oou M
w
m
.2. m llnlﬂlﬂwlzz-
o
nu
108 w
w 42%.
.53 w
m.
.82 m l llnﬁHNllSeﬁ-
O
gOON—
Ill»? hilllulmllnl' E
m

 

156

Figure 38. A: Effects of tetrodotoxin (TTX) and m-conotoxin on action potentials evoked
by direct depolarizing current injection (20 ms, 0.2-0.6 nA). a, control action potential.
b, application of TTX (5 nM) abolished action potential discharge. c, in the presence of
TTX (5 uM), an action potential with lower amplitude and slower rate of rise was
achieved by increasing the intensity of depolarizing pulse. d, superfusion of m-conotoxin
(700 nM) abolished TTX resistant action potential. B: Effects of human recombinant
interleukin-113 (1 ug/ml, 300 ms) on membrane potential and input resistance of the same
neuron in the presence of both TTX (5 nM) and m-conotoxin (700 nM).

157

.uom om

2: 8m .8? ﬂ SEE.
<

>5 3 H

 

.4?de IIIIIIEIJ i—IIIIIFIILII IIIJIIILI.

E- l/\ku . .

>E

 

C if :ESoccOé. + g vas—km! E 3.5.50 mV—

158

 

 

#3 4o - (a)

:3

:t.’

'5.

5 30 (4)

CL ..

U)

0.

LL!

‘27:

o 20 -

“- (2)

a...

O

8

3.3 10 -

:9

‘E

"' (2)

‘E

m 0 "

o

L.

(it, 1 1 l I J
200 400 600 800

Duration (ms)

Figure 39. Effect of human recombinant interleukin-113 (hrIL-IB) on amplitude of fast
excitatory postsynaptic potentials (f-EPSPs). Ordinate, percent inhibition of f-EPSP
amplitude. Abscissa, duration of pressure ejection of hrIL-lB (l ug/ml, 20 psi.),

expressed in ms. Vertical bars and number above symbols are standard error of mean and

number of cells tested, respectively.

159

Pressure application of hrIL-lB (1 ug/ml, 400 - 1200 ms. 20 psi) caused a
depression of f-EPSPs and blockage of orthodromic action potentials in 44 % (7 of 16) of
neurons tested. When nerve ﬁbers were stimulated at intensities that initiated only f-
EPSPs, the maximum depression of f-EPSPs amplitude by hrIL-lB (1 pg/ml, 800 ms, 20
psi) was 30.3 2!: 7.5 %. The action of hrIL-IB on the amplitude of f-EPSPs occurred in a
dose-dependent manner (Fig. 39).

Bath application of hrIL-lB (40 - 80 ng/ml) also caused a depression of f-EPSPs
amplitude in 42% (3 of 7) of neurons tested (Fig. 40). The maximum depression of f-
EPSPs amplitude caused by hrIL-lB (80 ng/ml) was 88.5 i 9.7%. The depression of f-
EPSPs by hrIL-IB also occurred when there was no change in transmembrane potential
(n=2) or when hrIL-lB-evoked depolarizations (n=4) or hyperpolarizations (n=3) were
nulliﬁed by injecting hyperpolarizing or depolarizing current through the microelectrode,
respectively (Fig. 40).

Effect of hrIL-Iﬂ on acetylcholine potentials mediated by nicotinic receptors.
HrIL-IB may depress the amplitude of f-EPSPs by altering the sensitivity of the
postsynaptic cholinoceptors to endogenously released Ach. To test this possibility, we
examined the effects of hrIL-lB on membrane potential changes caused by exogenous
Ach. The effect of hrIL-lB (4O - 80 ng/ml) was tested on nicotinic fast Ach potentials (n
=3). The Krebs solution contained tetrodotoxin (300 nM) to block action potentials that
are associated with fast Ach potentials and atropine (300 nM) to block slow Ach
potentials mediated by muscarinic receptors. Figure 41 shows that HrIL-lB (80 ng/ml)
caused membrane depolarization associated with a decrease in the amplitude of fast Ach
potentials recorded from one neuron. Aﬁer the hrIL-lB-evoked membrane depolarization
was nulliﬁed by direct injection of hyperpolarizing current, the amplitude of fast Ach

potentials was not changed.

160

Specificity of hrIL-Iﬂ effects

IL-1 receptor antagonist (IL-Ira) binds to IL-1 receptors and inhibits the binding
of IL-lB (Dinarello and Thompson, 1991). Superfusion of IL-lra (IO-4O ug/ml, 5 ~10
min) alone did not alter the passive and active membrane properties and synaptic
transmission of neurons in pelvic plexus ganglia (n=7). However, it blocked the effects
of hrIL-lB on membrane depolarization (n=3), hyperpolarization (n=3) (Fig. 42) and
synaptic transmission of neurons in pelvic plexus ganglia (n=2) (Fig. 43). IL-lra (20
ug/ml and 40 ug/ml) caused a concentration-dependent suppression of the hrIL-lB-
evoked membrane hyperpolarization by 65.2 d: 3.2% and 96.0 i 4.6%, respectively. The
data indicate that the effects of hrIL-lB on resting membrane potential and on synaptic

transmission are mediated by speciﬁc interaction with IL-1 receptors.

Discussion
The eﬂ'ect of bradykinin on mechanosensory input(s) to neurons in pelvic plexus
ganglia.

The data show that application of bradykinin to the distal colon initiated or
increased the mechanosensory inputs to a portion of neurons in the pelvic plexus ganglia.
The effects was associated with bradykinin-evoked contractions of colon. Bradykinin
had no effects on electrical activities and membrane properties of neurons in pelvic
plexus ganglia, when applied only to the ganglia site. The data indicate that bradykinin
sensitizes the mechnosensory input to neurons in pelvic plexus ganglia from peripheral
organs. The actions of bradykinin may be due to direct depolarization of afferent
terminals by activating the silent mechanoreceptors that only can be activated during
inﬂammation. Another possibility is that bradykinin interacts with a certain type of
polymodal receptors, which can be activated by both mechanical and chemical stimuli
(Haupt, J anig and Kohler, 1983; Janig and Koltzenberg, 1991). The actions of bradykinin

161

Figure 40. Effects of human recombinant interleukin-18 (hrIL-IB) on fast excitatory
postsynaptic potentials (f-EPSPs) of one neuron in pelvic plexus ganglia. Horizontal
lines, time period for superfusion of hrIL-IB (80 ng/ml). Resting membrane potential
was -56 mV. A: continuous recordings of membrane potential and f-EPSPs (upward
deﬂections) evoked by pelvic nerve stimulation. HrIL-lB reduced the amplitude of f-
EPSPs during depolarization and when membrane potential was at the resting level (b).
During time period between ﬁlled circles, hrIL-lB—evoked depolarization was nulliﬁed by
injection of anodal direct current. B: trace obtained 5 min. after removal of hrIL-lB. C:
continuous recording of B. Second application of hrIL-IB did not evoke membrane
depolarization but reduced the amplitude of f-EPSPs. lD: expanded records of f-EPSPs at
the time marked by respective letters in traces A, B and C. In each record, 5 consecutive

responses are superimposed.

162

 

 

 

 

 

163

Figure 41. Effect of human recombinant interleukin-113 (hrIL-IB) on fast acetylcholine
(Ach) potentials of one neuron in pelvic plexus ganglia. A: fast Ach potentials were
recorded in Krebs solution containing tetrodotoxin (300 nM) and atropine (300 nM).
Horizontal line, time period of application of hrIL-lB (80 ng/ml). During time period
between ﬁlled circles, hrIL-lB-evoked depolarization was nulliﬁed by injection of anodal
direct current. Resting membrane potential was -57 mV. Traces a and b, continuous
recording of fast Ach potentials. Trace c, trace obtained 5 min. after b. B: amplitude of
successive fast Ach potentials for neuron in A before (0), during (0) and after (0)
application of hrIL-lB. Note, amplitude of Ach potential was similar to the control when

hrIL-lB-induced depolarization was nulliﬁed.

164

AsEv 2:;
m

em ~N m~ vu @ m...

.00» 9-
TI.

2: c. H

1

(Am) cpnmdwv

 

 

165

Figure 42. Effects of IL-1 receptor antagonist (IL-Ira) on pressure application of human
interleukin-113 (hrIL-lB) induced membrane hyperpolarization in one neuron. Top traces:
hrIL-IB (1 ug/ml, 20 psi, 900 ms) caused membrane hyperpolarization when pelvic plexus
ganglia were superfused with a normal Krebs solution. Middle traces: hrIL-lB (lug/ml,
20 psi, 900 ms) caused membrane hyperpolarization 6 min. after pelvic plexus ganglia
were superﬁlsed with a Krebs solution containing IL-lra 20 ug/ml (a) and IL-lra 40
ug/ml (b). Bottom traces: hrIL-lB (lug/ml, 20 psi, 900 ms) caused membrane
hyperpolarization 10 min. after pelvic plexus ganglia were superfused with a normal

Krebs solution again. Membrane potential was -58 mV.

166

b
A A
‘ A
‘ A A
“”1"” tum-m [1ng

20 sec.

167

Figure 43. Effects of human interleukin-113 (hrIL-lB) on synaptic potentials of one
neuron in pelvic plexus ganglia in the absence and presence of IL-1 receptor antagonist
(IL-Ira). A and B: continuous recordings of membrane potential and synaptic potentials
evoked by electrical stimulation of urethra-urinary bladder nerves (large upward
deﬂections) before and after pressure application of hrIL-lB (1 ug/ml, 20 psi, 300 ms X 4,
30 sec. interval) in the absence (A) and presence (B) of IL-lra 40 ug/ml. Horizontal line,
time period for superfusion of IL-lra. Membrane potential was -64 mV. C: expanded
records of synaptic potentials at the time marked by respective letters in traces A and B.

In each record, 5 consecutive responses were superimposed.

168

m5 cm

.23

 

 

 

169

may also be due to activation of mechanoreceptors by the smooth muscle contraction
caused by bradykinin.

Bradykinin is released during inﬂammation and plays an important role in
mediating hyperalgesia and hyperreﬂexia through the central nervous system by
activating on visceral primary afferents (Keele and Armstrong, 1964, Haupt, J anig and
Kohler, 1983; Janig and Koltzenberg, 1991). The present study demonstrates that
bradykinin is also capable of mediating hyperreﬂexia through the prevertebral
parasympathetic ganglia by activating peripheral afferent ﬁbers which originate from
myenteric plexus ganglia. Axon collaterals from primary afferent ﬁbers are not involved,
based upon the results obtained with capsaicin in Chapter 4. However, the involvement
of axon collaterals from primary afferent ﬁbers to neurons in the myenteric plexus
ganglia can not be ruled out. It is possible that bradykinin acts on the same type of
primary afferent terminals to induce or facilitate both peripheral and central reﬂexes. The
axon collaterals from the primary afferents synapse with cholinergic neurons in myenteric
plexus and those cholinergic neurons further synapse with neurons in the pelvic plexus
ganglia.

The present study suggests that hyperreﬂexia in the visceral organs, induced by
inﬂammation, is mediated through both central nervous system and pelvic plexus ganglia.
It also implicates that neurons in pelvic plexus ganglia may play an important role under

both physiological and pathological conditions.

The effects of human recombinant interleukin-1,6 on synaptic transmission and on the
active and passive properties of neurons in pelvic plexus ganglia.

The data show that hrIL-lB causes membrane depolarization corresponding to an
inward membrane current and hyperpolarization corresponding to an outward membrane
current in neurons located in the guinea-pig pelvic plexus ganglia. Membrane

depolarization and membrane hyperpolarization are associated with a decrease and an

170

increase in membrane input resistance, respectively. These effects are mediated by IL-10
receptors, since the hrIL-lB-evoked responses are blocked by, the IL-lB antagonist. In
contrast, the actions of hrIL-lB are not altered by blocking nicotinic, muscarinic, a2-
adrenergic or opioid receptors. Also, blockade of voltage dependent sodium and
potassium channels or co-conotoxin sensitive voltage-dependent calcium channels does
not affect the hrIL-lB-evoked responses.

The hrIL- 1 B induced outward membrane current has a reversal potential of -5 i
4.5 mV. The ionic mechanism of the hrIL-lB-evoked outward current was not further
studied in the present dissertation work. The possible mechanism for the hrIL-lB-evoked
outward current could be: 1) an increase in K+ or Cl' conductance; 2) activation of Na+-
K+ pump; 3) decrease in resting Na+ conductance and 4) decrease in Ca2+ conductance.
Based upon the reversal potential which was obtained near -5 mV and the input resistance
change associated with the hyperpolarization (increase input resistance and decrease
membrane conductance), the possible involvement of potassium channels can be ruled
out as the reversal potential for potassium current is near -85 mV. Furthermore, the hrIL-
IB-evoked membrane hyperpolarization or outward membrane current are associated with
an increase in membrane input resistance (decrease membrane conductance), which
excludes the possibility of increase in K+ conductance. The equilibrium potential for C1'
is -15 mV in myenteric and sympathetic ganglion neurons when 2 M KCl ﬁlled
electrodes are used. At resting membrane potential (-50 - -7 5 mV), there will be an
inward current mediated by C1' efflux. This resting Cl' current may be signiﬁcant in
contributing to the resting membrane potential in neurons of pelvic plexus ganglia.
Inhibition of this resting Cl' current will cause membrane hyperpolarization associated
with a decrease in input resistance. Thus based upon the reversal potential and the resting
Cl' current mechanism, hrIL- 1 B-evoked membrane hyperpolarization or outward
membrane current can be due to blocking the resting Cl' current. However, based upon

the reversal potential and membrane input resistance change, it is also reasonable to

171

suggest that a decrease in a cation conductance may be involved in the hrIL-lB-evoked
membrane hyperpolarization or outward membrane current. To further determine the
ionic mechanism of hrIL-lB-induced hyperpolarization, ionic substitution experiments
should be performed. In Aplysia neurons, hrIL-lB causes membrane hyperpolarization
and an outward membrane current with the reversal potential near +15 mV. Ion
substitution and pharmacology studies show that low [Ca2+]o, Ca2+ channel blocking
cation, increasing [K+]O, TEA, 4-aminopyridine (4-AP), low [Cl']o, ouabain and TTX
have no effects on hrIL- 113-evoked outward membrane current. Thus it is concluded that
the hrIL- l B-evoked outward membrane current is independent of a decrease in Caz+
conductance, an increase in K+ or Cl' conductance, an increase in Ca2+ activated K+
conductance and an activation of Na“-K+ pump. In contrast, a decrease in [Na+]0 reduced
hrIL-lB-evoked outward membrane current, indicating that in Aplysia neurons, the effect
of hrIL-IB is exclusively due to a decrease in resting Na+ conductance. (Sawada, Hara
and Maeno, 1991). A similar decrease in resting Na+ conductance and decrease in Ca2+
conductance and increase in C1' conductance with net membrane input resistance increase
may also be involved in hrIL-llfevoked outward membrane current or membrane
hyperpolarization in neurons of pelvic plexus ganglia.

The ionic mechanism for hrIL-lB-evoked inward current and membrane
depolarization was also not further studied in the present study. However, the following
mechanisms may be involved. (1) an increase Na+ and/or Ca2+ conductance; 2) a
decrease in K" conductance; 3) an increase in resting Cl' conductance and 4) an inhibition
of Na+-K+ pump. The hrIL- l B—evoked membrane depolarization or intward membrane
current are associated with a decrease in membrane input resistance (increase membrane
conductance), therefore it is unlikely decrease in K” conductance is involved. However,
it is possible that increase in resting Cl' conductance and increase in cation conductance
are involved in the mechanism for hrIL-lB-induced depolarization associated with a

decrease in membrane input resistance.

172

The actions of hrIL-lB on the membrane potential and membrane input resistance
of parasympathetic neurons in pelvic plexus ganglia are similar to the actions of hrIL-IB
found in other neurons. In rat hypothalamic supraoptic neurons, IL- 1 B causes membrane
depolarization and hyperpolarization and increases neuronal excitability (Li, Inenage,
Kawano, Kannan and Yamashita, 1992). In hippocampal pyramidal neurons, hrIL-lB
reduces voltage dependent-calcium currents (Plata-Salaman and Ffrench-Mullen, 1992).

The effects of hrIL-lB are similar to the actions of other inﬂammatory mediators.
For example, platelet-activating factor causes membrane depolarization in cultured rat
myenteric neurons (Willard, 1992). Recombinant human tumor necrosis factor (rhTNF)
causes a slow inward current associated with a decrease in membrane conductance in
Aplysia abdominal ganglion neurons (Swada, Hara and Maeno, 1990). Human
recombinant interleukin-2 depressed the voltage-dependent inward calcium current in
guinea-pig hippocampal CA1 neurons (Plata-Salaman and Ffrench-Mullen, 1993). In
guinea-pigs actively antigen sensitized airway tissue, antigen challenge-caused membrane
depolarization and change in membrane excitability in bronchial parasympathetic
ganglion neurons caused by the release of inﬂammatory mediator(s) including IL-IB
(Myers, Undem and Weinreich, 1991).

The present results also indicate that hrIL-IB depresses the f-EPSPs and blocks
orthodromic action potentials evoked in a portion of neurons in pelvic plexus ganglia by
electrical stimulation of pelvic, hypogastric and urethra-urinary bladder nerves. This
suggests that there are IL-lB receptors located presynaptically, that act to depress
electrically evoked release of acetylcholine from preganglionic ﬁbers in pelvic,
hypogastric and urethra-urinary bladder nerves. The depression of the f-EPSPs occurred
when the hrIL-lB-evoked membrane depolarizations or hyperpolarizations were nulliﬁed
by direct hypepolarizing or depolarizing current injection, respectively. The effects are
not due to alteration of the sensitivity of the nicotinic postsynaptic receptors, since hrIL-

IB has virtually no effect on nicotinic depolarizations evoked by acetylcholine. The data

173

show that hrIL- l B modulates not only the synaptic inputs from the central nervous
system, but also the synaptic inputs from nerve trunks that connect peripheral organs to
pelvic plexus ganglia. A presynaptic action of hrIL-lB has been suggested in rat
myenteric nerves, where hrIL-lB suppresses the evoked release of acetylcholine or
norepinephrine (Collins, Hurst, Main, Stainley, Khan, Blennerhassett and Swain, 1992;
Hurst and Collins, 1993; Main, Blennerhassett and Collins, 1993). In the rat
hippocampus neurons, hrIL-lB inhibits long-term potentiation (Bellinger, 1993) and
decreases the release of acetylcholine (Rada, Mark, Vitek, Mangano, Blume, Beer and
Hoebel, 1991) .

In summary, the results of the present study show that IL-lB affects the electrical
properties and synaptic transmission of neurons in pelvic plexus ganglia. It is possible
that during inﬂammation, IL-lB and other cytokines are released from blood activated
macrophages (Dinarello, 1985, 1988), macrophage-like cells (F aussone-Pellegrini,
Pantalone and Cortesini, 1990; Jessen and Thuneberg, 1991), vascular smooth muscles
(Warner, Auger and Libby, 1987) or neurons in pelvic plexus ganglia (Freidin, Bennett
and Kessler) to directly affect sacral parasympathetic efferent pathways. Furthermore, I
speculate that alteration in sacral parasympathetic reﬂexes during inﬂammation involves
the action(s) of inﬂammatory mediator(s) on both sacral visceral afferent (Habler, J anig
and Koltzenburg, 1990) and sacral parasympathetic efferent pathways.

Overall, the studies in this chapter show that inﬂammatory mediators such as
bradykinin and interleukin-18 are capable of modulating pelvic plexus ganglia-mediated
peripheral reﬂexes at both afferent limb (bradykinin) and center station (pelvic plexus
ganglia) of the reﬂexes (interleukin-113). It has been shown that both bradykinin and
interleukin-113 are released during inﬂammation and the release of interleukin-18 can
further induce the release of bradykinin and other hyperalgesic mediators (F erreira,
Lorenzetti, Cunha and Poole, 1993). Furthermore, both bradykinin and interleukin-113 are

involved in inducting hyperalgesia by directly acting on afferent terminals (Juan and

174

Lembeck, 1974; Steranka, Manning, de Haas, Ferkany, Borosky, Connor, Vavrek,
Stewart, and Snyder, 1988; F erreira, Lorenzetti, Cunha and Poole, 1993; Watkins,
Wiertelak, Goehler, Smith, Martin and Maier, 1994). The studies suggest that bradykinin
and interleukin-18 are involved in the mechanisms of hyperalgesia during inﬂammation.
It is possible that bradykinin and/or interleukin-113 are involved in producing discomfort
and pain in patients with inﬂammatory bowel disease (IBD).

Besides its action on afferent terminals, interleukin-10 has been shown to inhibit
cholinergic synaptic transmission in neurons of pelvic plexus ganglia from both central
and peripheral nerves. Studies in the myenteric plexus have also shown that interleukin-
1B suppressed the release of acetylcholine from myenteric terminals (Collins, Hurst,
Main, Stanley, Khan, Blennerhassett and Swain, 1992). These suggest that interleukin-113
may be also involved in inhibition of inﬂammation-induced hyperreﬂexia. The
protective role of interleukin-113 in IBD has been suggested (Cominelli, Nast, Lierena,
Dinarello and Zipser, 1990). Hyperalgesia and hyperreﬂexia are both present in patients
with IBD. Bradykinin has been known to be involved in both hyperalgesia and
hyperreﬂexia. My dissertation work has demonstrated that hyperreﬂexia can be mediated
not only at the level of the central nervous system but also at the level of autonomic
ganglia. The involvement of the pelvic plexus ganglia in mediating excitatory reﬂexes
and the effects of bradykinin on the afferent limb of this reﬂex may contribute in part of
the hyperreﬂexia in patients with IBD. Others have demonstrated hyperalgisia caused by
interleukin-113 (Ferreira, Lorenzetti, Cunha and Poole, 1993; Watkins, Wiertelak,
Goehler, Smith, Martin and Maier, 1994). However, the functional role of interleukin-113
in regulation of central and peripheral reﬂexes at physiological condition and in patients

with IBD is uncertain and remains to be further studied.

SUMMARY

1. The passive and active electrophysiological properties of neurons in pelvic
plexus ganglia, isolated from male guinea-pigs, were investigated. The neurons were not
spontaneously active and had a resting membrane potential of -56.3 i 0.3 mV (n=89);
membrane input resistance of 51.4 i 2.5 MG (n=82); membrane time constant of 6.5 :1:
0.3 ms (n=76); and amplitude of action potentials of 73.0 i 0.5 (n=79). The membrane
properties of neurons in the pelvic plexus ganglia are similar to the properties obtained in
neurons located in other parasympathetic ganglia.

2. Intracellular recordings were obtained in vitro from neurons in pelvic plexus
ganglia of male guinea-pigs with attached central (hypogastric and pelvic) and peripheral
(colonic-rectal and urethra-urinary bladder) nerves and with (n=202) or without (n=245)
an attached segment of distal colon via colonic-rectal nerves.

3. When a segment of distal colon was not attached to the pelvic plexus ganglia,
neurons in pelvic plexus ganglia received synaptic input(s) consisting of fast excitatory
postsynaptic potentials (f-EPSPs) and/or action potentials during electrical stimulation of
hypogastric nerve (84%, 205 of 245), pelvic nerves (80%, 102 of 128), colonic-rectal
nerves (67%, 108 of 161) and urethra-urinary bladder (80%, 109 of 135) nerves. The
synaptic responses were reversibly blocked by hexamethonium (C6, 10 - 100 nM) and
dihydro-B-erythroidine (10 - 50 nM) but not by atropine (0.1 - 1.0 nM) indicating that
nicotinic but not muscarinic acetylcholine receptors are involved.

4. When pelvic plexus ganglia were attached to a segment of distal colon via
colonic-rectal nerves, 27% (54 of 202) of the neurons tested exhibited continuous

electrical activities consisting of f-EPSPs and action potentials which were reversibly

175

176

blocked by C6 (10 - 100 nM) applied to the ganglia compartment of the organ bath and
irreversibly abolished by sectioning the colonic-rectal nerves.

5. Distension of distal colon segment increased the amplitude and frequency of f-
EPSPs or the frequency of action potentials in neurons which exhibited continuous
electrical activities. In quiescent neurons, distension of the distal colon initiated f-EPSPs
and/or action potentials. The results suggest that neurons in pelvic plexus ganglia receive
sensory inputs from mechanoreceptors located in the distal colon wall. These sensory
inputs involve nicotinic acetylcholine receptors.

6. Superfusion of the distal colon segment with a Krebs solution containing
bradykinin (1 nM) initiated f-EPSPs and action potentials in quiescent neurons of pelvic
plexus ganglia. The ﬁequency and the amplitude of f-EPSPs was further increased by
colon distension. The data suggest that bradykinin sensitizes the mechanosensory input
from distal colon-rectum to neurons in pelvic plexus ganglia.

7. Reﬂexes to colon and urinary bladder in male guinea-pigs were studied in vitro
using preparations with pelvic plexus ganglia attached to two separated segments of distal
colon via colonic nerves or to distal colon and urethra-urinary bladder via colonic nerves
and urethra-urinary bladder nerves. Passive distension of one colon segment or electrical
stimulation of colonic nerves (3-20 Hz) evoked contractile responses of the other colonic
segment and urinary bladder. Electrical stimulation of urethra-urinary bladder nerves (3-
20 Hz) also evoked colon contractile responses. The data suggest that neurons in pelvic
plexus ganglia mediate excitatory reﬂexes to colon and urinary bladder.

8. The contractile response to passive distension of one of the colon segments or
to electrical stimulation of colonic or urethra-urinary bladder nerves were blocked by
tetrodotoxin (TTX, 1 uM), hexarnetonium (C6,10 nM) and atropine (1 nM). Also, these
responses were abolished by sectioning the colonic or urethra-urinary bladder nerves.

The data indicate that reﬂex contractions involve cholinergic nerve ﬁbers which go with

177

the colonic and urethra-urinary bladder nerves and activate nicotinic and muscarinic
acetylcholine receptors.

9. The action of human recombinant interleukin-113 (hrIL-lB) was tested on
neurons in isolated guinea-pig pelvic plexus ganglia, utilizing in vitro intracellular
microelectrode recording techniques. Pressure or bath application of hrIL-lB caused
membrane depolarization (3-32 mV) and an inward current (0.5-1.0 nA) in 54% of
neurons tested. The depolarization was associated with a decrease in membrane input
resistance. HrIL-lB also caused membrane hyperpolarization (3-14 mV) and an outward
current (0.3-0.8 nA) in 30% of neurons tested. The hyperpolarization was associated with
an increase in membrane input resistance.

10. Membrane depolarization, hyperpolarization and associated changes in
membrane input resistance evoked by hrIL-IB were resistant to hexamethonium (100
nM), atropine (500 nM), yohimbine (300 nM) or naloxone (1 nM) as well as to
tetrodotoxin (5 uM), to-conotoxin (700 nM) and tetraethylammonium (30 mM). The
effects of hrIL-lB on membrane potential and membrane input resistance were blocked
by the IL-lB receptor antagonist (10-40 rig/ml). The data indicate that the effects of hrIL-
IB on the membrane potential and input resistance of neurons in pelvic plexus ganglia are
mediated by speciﬁc IL-l receptors and do not involve activation of nicotinic and
muscarinic acetylcholine receptors, az-adrenergic and opioid receptors as well as changes
in the transmembrane currents carried through voltage-dependent Na+, Ca2+ or K+
channels.

11. Electrical stimulation of pelvic, hypogastric or urethra-urinary bladder nerves
evoked f-EPSPs and action potentials, which were reversibly blocked by hexamethonium
(10-100 nM) or dihydro-B-erythroidine (5-100 nM), indicating that the responses are due
to activation of cholinergic nicotinic receptors. In 44% of neurons tested (n=7), hrIL-IB
inhibited the orthodromic action potentials and f-EPSPs in a dose-dependent fashion.

The effects of hrIL-IB on synaptic transmission were blocked by the IL-lB receptor

178

antagonist (40 jig/ml). HrIL-IB had no effects on the depolarizations due to activation of
nicotinic receptors by pressure application of acetylcholine. The data suggest that hrIL-IB
does not alter the sensitivity postsynaptic nicotinic acetylcholine receptors. Thus, the
inhibition of cholinergic synaptic transmission in pelvic plexus ganglia by hrIL-lB is
most probably due to activation of IL-lB receptors on presynaptic nerve terminals, which
inhibit the release of acetylcholine.

In summary, the data obtained in the present studies show that neurons in pelvic
plexus ganglia receive mechanoreceptor-mediated synaptic inputs from distal colon.
These neurons are capable of mediating reﬂexes between colon segments and between
the colon and the urinary bladder. The data also suggest that inﬂammatory mediators,
such as bradykinin and interleukin-10, may regulate these peripheral reﬂexes at both, the
afferent site and at the level of the pelvic plexus ganglia. The pelvic plexus ganglia-
mediated visceral organ reﬂexes may play a signiﬁcant role in modulating the defecation

and micturition reﬂexes under both physiological and pathological conditions.

SIGNIFICANCE OF THE STUDY

Classically, afferent information from distal colon, the rectum and anal canal is
received in the sacral and lumbar regions of the spinal cord via the pelvic and pudendal
nerve and via the hypogastric and inferior lumbar splanchnic nerves. These afferent
ﬁbers are involved in the initiation of spinal autonomic and somatic reﬂex mechanisms
and the transmission of visceral and somatic sensation to higher centers in the brain.
Afferent information from distal colon-rectum is also received by the intramural plexus
ganglion neurons and involved in local gastrointestinal reﬂexes. Both types of reﬂexes
participate in defecation. The present study established a novel reﬂex pathway which
may also be involved in autonomic and defecation reﬂexes. The afferents of this pathway
send information to neurons in pelvic plexus ganglia, a parasympathetic ganglia. Most
pelvic plexus ganglia efferents innervate visceral organs of the pelvic region, and mediate
colon-colonic reﬂex and colon-urinary bladder reﬂex. These pathways may participate in
the defecation reﬂex, micturition reﬂex and regulate distal colon and urinary bladder
motilities under either physiological (normal defecation) and pathological conditions (eg.
defecation under spinal cord injury condition). Since the present studies also
demonstrated that inﬂammatory mediators are able to modulate the afferent information
to neurons in pelvic plexus ganglia and alter pelvic plexus ganglia-mediated reﬂexes, this
reﬂex pathway may play a signiﬁcant role in patients with inﬂammatory visceral organ
diseases.

The present dissertation work have been published in abstract form (Lin and
Krier, 1993a,b, 1994, 1995) and the work in Chapter 1, 2, 3, and 4 are in preparation for
publication as a full manuscript. The dissertation work in Chapter 5 has been accepted as

a full manuscript and will be published in American Journal of Physiology, 1995.
179

BIBLIOGRAPHY

BIBLIOGRAPHY

Aarden, L. A., T. K. Brunner, J. C. Cerottini (1979) Revised nomenclature for antigen-
non-speciﬁc T-cell proliferation and helper factors. J. Immunol. 123: 2928-2929
(letter).

Adamson, W. A. D. and I. Aird (1932) Megacolon: evidence in favor of a neurogenic
origin. Br. J. Surg. 20: 220-233.

Amara, S. G., V. Jonas, M. G. Rosenfeld, E. S. Ong and R. M. Evans (1982) Alterative
RNA processing in calcitonin gene expression generates mRN A encoding
different polypeptide products. Nature. 298:240-244.

Baker, D. G., H. M. Coleridge, J. C. Coleridge and T. Nerdrum (1980) Search for cardiac
nociceptor: stimulation by bradykinin of sympathetic afferent nerve endings in the
heart of the cat. J. Physio]. Lond. 306: 519-536.

Bahns, E., U. Emsberger, W. Janig and A. Nelke (1986) Functional characteristics of
lumbar visceral afferent ﬁbers from the urinary bladder and urethra in the cat.
Pﬂu'g. Arch. 407: 510-518.

Bauer, A. J ., M. Hanani, T. C. Muir and J. H. Szurszewski (1991) Intracellular recordings
from gallbladder ganglia of opossums. Am. J. Physiol. 260:G299-G306.

Bellinger, F. P., S. Madamba and G. R. Siggins (1993) Interleukin 18 inhibits synaptic
strength and long-terrn potentiation in the rat CA1 hippocampus. Brain Res. 628:

227-234.

Billingharn, M. E. J. (1987) cytokine as inﬂammatory mediators. British Med. Bull.
43(2): 350-370.

Bemheirn, H. A., L. H. Block and E. Atkins (1979) Fever: pathogenesis,
pathophysiology, and purpose. Ann. Intern. Med. 91: 261-270.

Bevan, S. J. and C. A. Forbes (1988) Membrane effects of capsaicin on rat dorsal root
ganglion neurons in cell culture. J. Physiol. (Land). 398: 28P.

Blackman, J. G., J. P. Crowcroft, C. E. Devine, M. E. Holman and K. Yonemura (1969)
Transmission from preganglionic ﬁberes in the hypogastric nerve to peripheral
ganglia of male guinea- pigs. J. Physio]. 201:723-743.

180 '

181

Breder, C. D., C. A. Dinarello and C. B. Saper (1988) Interleukin-1 immunoreactivation
of the human hypothalamus. Science. 240(4850): 321-324.

Brugger, F. , R. H. Evans and N. S. Hawkins (1990) Effects of N-methyl-Daspartate
antagonists and spantide on spinal reﬂexes and responses to substance P and

capsaicin in isolated spinal cord preparations from mouse and rat. Neuroscience
36:611-622.

Buck, S. H. and E. Burcher ( 1986) The tachykinins: a family of peptides with a brood of
receptors. Trends Pharmacol. Sci. 7: 65-68.

Buck, S. H. and T. F. Burks (1986) The neuropharmacology of capsaicin: a review of
some recent observations. Pharmacol. Rev. 38:179-226.

Buck, S. H., R. M. Pruss, J. L. Krstenansky, P. J. Robinson and K. A. Stauderman (1988)
A tachykinin peptide receptor joins an elite club. Trends Pharmacol. Sci. 9: 3-5.

Burgess, G. M., I. Mullaney, M. McNeill, P. R. Coote, A. Minhas and J. N. Wood (1989)
Activation of guanylate cyclase by bradykinin in rat sensory neurons is mediated
by calcium inﬂux: possible role of the increase in cyclic GMP. J. Neurochem.
53(4): 1212-1218.

Burgess, G. M., I. Mullaney, M. McNeill, P. M. Dunn and H. P. Rang (1989) Second
messengers involved in the mecdhanism of action of bradykinin in sensory
neurons in culture. J. Neurosci. 9(9): 3314-3325.

Caranikas, S., J. Mizrahi, P. D'Orléans-Juste and D. Regoli (1982) Antagonist of
substance P. Eur. J. Pharmacol. 77: 205-206.

Cevero, F. and K. A. Sharkey (1988) An electrophysiological and anatomical study of
intestinal afferent ﬁbers in the rat. J. Physio]. Land. 401 :381-397.

Chahl, L. A. (1982) Evidence that the contractile response of guinea-pig ileum to
capsaicin is due to substance P release. Naunyn-Schmiedeberg's Arch.
Pharmacol. 319: 212-215.

Chahl, L. A. and A. Iggo (1977) The effects of bradykinin and PGE2 on rat cutaneous
afferent nerve activity. Br. J. Pharmacol. 59:343-347.

Collins, S. M., S. M. Hurst, C. Main, E. Stanley, 1. Khan, P. Blennerhassett and M. Swain
(1992) Effect of inﬂammation on enteric nerves: cytokine-induced changes in
neurotransmitter content and release. Ann. N. Y. Acad. Sci. 664: 415-424.

Cominelli, F ., C. C. Nast, B. D. Clark, R. Schindler, R. Lierena, V. E. Eysselein, R. C.
Thompson and C. A. Dinarello (1990) Interleukin 1 (IL-1) gene expression,

182

synthesis, and effect of speciﬁc IL-1 receptor blockade in Rabbit immune
complex colitis. J. Clin. Invest. 86:972-980.

Cominelli, F ., C. C. Nast, R. Lierena, C. A. Dinarello and R. D. Zipser (1990) Interleukin
1 suppresses inﬂammation in rabit colitis mediation by endogenous
prostaglandins. J. Clin. Invest. 85:582-586.

Cooke, A. R. (1975) Control of gastric emptying and motility. Gastraenterolagy
68:804-816.

Costa, M. and J. B. Furness (1976) The peristaltic reﬂex: an analysis of nerve pathways
and their pharmacology. Naunyn-Schmiedeberg's Arch. Pharmacol. 294:47-60.

Cottrell, D. F. and A. Iggo (1984) The responses of duodenal tension receptors in sheep
to pentagastrin, cholecystokinin and other drugs. J. Physiol. Land. 354: 477-495.

Crowcroft, P. J., M. E. Holman and J. H. Szurszewski (1971) Exitatory input from the
distal colon to the inferior mesenteric ganglion in the guinea-pig. .1. Physio].
219:443-461.

Crowcroft, P. J. and J. H. Szurszewski (1971) A study of the inferior mesenteric and
pelvic ganglia of guinea-pigs with intracellular electrodes. J. Physiol. 219: 421-
441.

Dalsgaard, C.-J., M. Vincent, T. Schultzberg, T. Hokfelt, L.-G. Elfvin, L. Terenius and G.
J. Dockray (1983) Capsaicin-induced depletion of substance P-like
immunoreactivity in guinea pig sympathetic ganglia. J. Autonomic Nervous
System. 9: 595-606.

Daniel, E. E., M. B. Parrish, E. G. Watson, J. E. T. Fox-Threlkeld, I). Regoll and K. D.
Rainsford (1995) The tachykinin receptors including contractile responses of
canine ileum circular muscle. Am. J. Physio]. 268 (Gastrointest. Liver Physiol.
31): Gl61-G170.

Davey, N. J. and P. H. Ellaway (1989) Facilitation of individual y-motoneurones by the
discharge of single slowly adapting type 1 mechanoreceptors in cats. J. Physio].
411:97-114.

Davies, P. J ., H. A. Coleman and H. C. Parkington (1993) Phasic and tonic ﬁringneurons
in the pelvic plexus of the female guinea-pig. X2001 Congress of the International
Union of Physiological sciences 96.1/P, 84.

Dayer, J. M., D. R. Robinson and S. M. Krane (1977) Prostaglandin production by
rheumatoid synovial cells: stimulation by a factor from human mononuclear cells.
J. Exp. Med. 145: 1399-1404.

183

Decktor, D. L. and W. A. Weems (1981) A study of renal-efferent neurons and their
neural connexions within cat renal ganglia using intracellular electrodes. J.
Physio]. 321 :61 1-626.

Decktor, D. L. and W. A. Weems (1983) An intracellular characterization of neurons and
neural connexions within the left coeliac ganglion of cats. J. Physio]. 341 :197-
211.

de Groat, W. C. and J. Krier (1975) Preganglionic C-ﬁbers: A major component of the
sacral autonomic outﬂow to the colon of the cat. Pﬂz‘igers Arch. 359:171-176.

de Groat, W. C. and J. Krier (1976) An electrophysiological study of the sacral
parasympathetic pathway to the colon of the cat. J. Physiol. Land 260: 425-445.

de Groat, W. C. and J. Krier (1978) The sacral parasympathetic reﬂex pathway regulating
colonic motility and defecation in cat. J. Physiol. Land. 276: 481-500.

de Groat, W. C., I.Nade1haft, R. J. Milne, A. M. Booth, C. Morgan and C. K. Thor
(1981) Organization of the sacral parasympathetic reﬂex pathways to the urinary
bladder and large intestine. J. Autonomic Nervous System. 3: 135-160.

Dinarello, C. A. (1985) An update on human interleukin-1: from molecular biology to
clinical relevance. J. Clin Immunol. 5: 287-297.

Dinarello, C. A (1988) Biology of interleukin 1. FASEB 2(2): 108-115.

Dinarello, C. A and R. C. Thompson (1991) Blocking IL-l: interleukin 1 receptor
antagonist in vivo and in vitro. Immuna]. Today. 12(11): 404-410.

Dingle J. T., J. Saklatvala, R. Hembry, J. Tyler, H. V. Fell and R. A. Jubb (1979) A
cartilage catabolic factor from synovium. Biochem. J. 184: 177-180.

Docherty, R. J ., B. Robertson ans S. Bevan (1991) Capsaicin causes prolonged inhibition
of voltage-activated calcium currents in adult rat dorsal root ganglion neurons in
culture. Neuroscience. 40 (2): 513-521.

Dripps, D. J ., E. Verderber, R. K. Ng, R. C. Thompson and S. P. Eisenberg (1991)
Interleukin-1 receptor antagonist binds to the type II interleukin-1 receptor on B
cells and neutrophils. J. Biol. Chem. 266(30): 20311-20315.

Dun, N. J. and S. Minota (1981) Effects of substance P on neurones of the inferior
mesenteric ganglia of the guinea-pig. J. Physio]. 321:259—271.

184

Dunn, P. M. and H. P. Rang (1990) Bradykinin-induced depolarization of primary
afferent nerve terminal in the neonatal rat spinal cord in vitro. Br. J. Pharmacol.
100: 656-660.

Eastgate, J. A., J. A. Symons and G. W. Duff (1990) Identiﬁcation of an interleukin-1
beta binding protein in hmnan plasma. FEBS Lett. 260(2): 213-216.

Eisenberg, S. P., R. J. Evans, W. P. Arend, E. Verderber, M. T. Brewer, C. H. Hanntun
and R. C. Thompson (1990) Primary structure and functional expression from
complementary DNA of a human interleukin-1 receptor antagonist. Nature.

343(6256): 341-346.
Erspamer, V. (1981) The tachykinin peptide family. Trends Neurosci. 4:267—269.

Faussone-Pellegrini, M. S. D. Pantalone and C. Cortesini (1990) Smooth muscle cells,
interstitial cells of Cajal and myenteric plexus interrelationships in the human
colon. Acta. Anat. Base]. 139: 31-34.

Ferreira, S. H., B. B. Lorenzetti, F. Q. Cunha and S. Poole (1993) Bradykinin release of
TNF-a plays a key role in the development of inﬂammatory hyperalgesia. Agents
Actions. Special Conference Issue. 38: C7-C9.

Floyd, K., V. E. Hick and J. F. B. Morrison (1976) Mechanosensitive afferent units in the
hypogastric nerve of the cat. J. Physio]. 259:457—471.

Floyd, K., V. E. Hick, J. Koley and J. F. B. Morrison (1977) The effects of bradykinin on
afferent units in intra-abdominal sympathetic nerve trunks. Q. J. Exp. Physio].
62: 19-25.

F reidin, M., M. V. Bennett and J. A. Kessler (1992) Cultured sympathetic neurons
synthesize and release the cytokine interleukin 1 beta. Proc. Nat]. Acad. Sci.
USA. 89 (21): 10440-10443.

Garry, R. C. (1933) The nervous control of the caudal region of the large bowel in the cat.
J. Physio]. Landon. 77:422-431.

Garry, R. C. (1934) The movement of the large intestine. Physio]. Rev. 14:103-132.

Gallagher, J. P. and P. Shinnick-Gallagher. (1986) Excitatory transmission in _
parasympathetic ganglia. In: Autonomic and Enteric Ganglia. New York:
Plenum: 341-351.

Gammon C. M., A. C. Allen and P. Morell (1989) Bradykinin stimulates
phosphoinositide hydrolysis and mobilization of arachidonic acid in dorsal root
ganglion neurons. J. neurachem. 53(1): 95-101.

185

Gauldie, J. (1991) Acute and chronic inﬂammation. In Effects of immune cells and
inﬂammation on smooth muscle and enteric nerves. W. J. Snape and S. M.
Collins Eds. 1-10. CRC Press, Inc. Boca Raton.

Gery, I. and B. H. Waksman (1972) Potentiation of the T-lymphocyte response to
nritogens 11. The cellular source of potentiating mediator(s). J. Exp. Med. 136:
143-1 55.

Gibbins, I. L., J. B. F umess and M. Costa (1987) Pathway-speciﬁc patterns of the co-
existence of substance P, calcitonin gene-related peptide, cholecystokinin and
dynorphin in neurons of the dorsal root ganglia of the guinea-pig. Cell Tissue
Res. 248: 417-437.

Gibbins, I. L., J. B. Furness and M. Costa, 1. MacIntyre, C. J. Hillyard and S. Girgis
(1985) Co-localization of calcitonin gene-related peptide-like immunoreactivity
with substance P in cutaneous, vascular and visceral sensory neurons of guinea
pigs. Neurosci Lett. 57: 125-130.

Gillespie, J. S. and J. D. Maxwell (1971) Adrenergic innervation of the sphincteric and
non-sphincteric smooth muscle in the rat intestine. J. Histachem. Cytochem.

19: 676-681.

Giulian, D., R. L. Allen, T. J. Baker and Y. Tomozawa (1986) Brain peptides and glial
growth I. Glia-promoting factors as regulators of gliogenesis in the developing
and injured central nervous system. J. Cell. Biol. 102(3): 803-811.

Goodman, E. C. and L. L. Iversen (1986) Calcitonin gene-related peptide: novel
neuropeptide. Life Sci. 38: 2169-2178.

Goyal, R. K., J. N. Sengupta and J. K. Saha (1992) Properties of esophageal
mechanosensitive receptors. Advances in the Innervation of the Gastrointestinal
Tract. G. E. Halle et a]. editors. 523-537.

Granowitz, E. V., B. D. Clark, J. Mancilla and C. A. Dinarello (1991) Interleukin-1
receptor antagonist competitively inhibits the binding of interleukin-1 to the type
II interleukin-1 receptor. J. Biol. Chem. 266(22): 14147-14150.

Griesbacher, T. and F. Lembeck (1987) Actions of bradykinin antagonists on bradykinin-
induced plasma extravasation, prostaglandin E release, nociceptor stimulation
and contraction of the iris sphincter muscle of the rabbit. Br. J. Pharmacol. 92:

333-340.

Grifﬁth III,W. H., J. P. Gallagher and P. Shinnick-Gallagher (1980) An intracellular
investigation of cat vesical pelvic ganglia. J. Neurophysiol. 43:343-354.

186

Grundy, D. (1985) Gastrointestinal Motility. The Integration of Physiological
Mechanisms. Lancaster, UK: MTP.

Grundy, D. and T. Scratcherd (1989) Sensory afferents from the gastrointestinal tract.
Handbook of Physiology. The Gastrointestinal system I. 593-620.

Guilbaud G., D. Lebars and J. M. Besson (1976) Bradykinin as a tool in
neurophysiological studies of pain mechanisms. In: Advances in pain research
and therapy, edited by J. J. Bonica and D. Albe-Fassard. New York: Raven. p.
67-73.

Guzman, F ., D. Braun and R. K. S. Lim (1962) Visceral pain and pseudoaffective
response to intra-arterial injection of bradykinin and other algesic agents. Arch.
Int. Pharmacodyn. T her. 136: 353-384.

Habler, H. J ., W. Janig and M. Koltzenburg (1988) A novel type of unmyelinated
chemosensitive nociceptor in the acutely inﬂamed urinary bladder. Agents
Actions. 25: 219-221.

Habler, H. J ., W. Janig and M. Koltzenburg (1989) Activation of chemosesitive C-ﬁbers
by an acute inﬂammation of the urinary bladder of the cat. J. Physiol. (Land).
415: 65.

Habler, H. J ., W. Janig and M. Koltzenburg (1990) Activation of Unmyelinated afferent
ﬁbers by mechanical stimuli and inﬂammation of the urinary bladder in the cat. .1.
Physio]. Land. 425: 545-562.

Habler, H. J ., W. Janig, M. Koltzenburg and S. B. McMahon (1990) A quantitative study
of the central projection patterns of unmyenlinated ventral root afferents in the
cat. J. Physio]. 422:265-287.

Hall, A. K., P. J. Barnes, L. A. Meldrum and H. J. Maclagau. (1989) Facilitation by
tachykinins of neurotransmission in guinea-pig pulmonary parasympathetic
nerves. Br. J. Pharmacol. 97:274-280.

Handwerker, H. O. (1976) Inﬂuence of algogenic substances abd prostaglandins on the
discharges of unmyenlinated cutaneous nerve ﬁbers identiﬁed as nociceptors. In:
Pain Research and Therapy, edited by J. J. Bonic and D. Albe-Fessard. New
York: Raven, vol. 1, p. 41-45.

Handwerker, H. O. (1980) Pain producing substances. In: Pain and Society edited by H.
W. Kosterlitz and L. Y. Terenius. Weinheim, F .R.G: VCH p. 325-338.

187
Hannum, C. H., C. J. Wilcox, W. P. Arend, F. G. Joslin, D. J. Dripps, P. L. Heimdal, L.
G. Armes, A. Sommer, S. P. Eisenberg and R. C. Thompson (1990) Interleukin-1
receptor antagonist activity of a human interleukin-1 inhibitor. Nature. 343: 336-
340.

Hardy, J. D., H. G. Wolff and H. Goodell (1950) Experimental evidence on the nature of
cutaneous hyperalgesia. J. Clinic. Invest. 29:115-140.

Hartman, D. and J. Krier (1984) Synaptic and antidromic potentials of visceral neurons in
ganglia of the lumbar sympathetic chain of the cat. J. physio]. 350:423-428.

Haupt, P., W. Jéinig and W. Kohler (1983) Response pattern of visceral afferent ﬁbers,
supplying the colon, upon chemical and mechanical stimuli. Pﬂugers Arch.
398:41-47.

Hertog, A. D., A. Nelemans and J. V. den Akker (1988) The multiple action of
bradykinin on smooth muscle of guinea-pig taenia caeci. Eur. J. Pharmacol.
151: 357-363.

Heyman, I. and H. P. Rang (1985) Depolarizing responses to capsaicin in a subpopulation
of rat dorsal root ganglion cells. Neurosci lett. 56:69-75.

Holtzer, P. (1988) Local effector functions of capsaicin-sensitive sensory nerve endings:
involvement of tachykinins, calcitonin gene-relared peptide and other
neuropeptides. Neuroscience. 24 (3): 739-768.

Holtzer, P. (1991) Capsaicin: cellular targets, mechanism of action, and selectivity for the
sensory neurons. Pharmacological Reviews. 43 (2): 143-196.

Hayes, A. D., and P. Barber (1981) Degenerawtion of axons in the ureteric and duodenal
nerve plexuses of the adult rat following in vivo treatment with capsaicin.
Neurosci. Lett. 25: 19-24.

Hurst, S. and S. M. Collins (1993) Interleukin-113 modulation of norepinephrine release
from ratrnyenteric nerves. Am. J. Physio]. 264 (Gastrointes. Liver Physiol. 27):
G30-G35.

Iggo, A. (195 7) Gastrointestinal tension receptors with unmyelinated ﬁbers in the vagus
of the cat. J. Exp. Physio]. 42:130-143.

J acobovvitz, D. (1965) Histochemical studies of the autonomic innervation of the gut. J.
Pharmacol. and Exp. Ther. 149:358-364.

J ancso, N. (1960) Role of the nerve terminals in the mechanism of inﬂammatory
reactions. Bull Millard Fillmore Hosp. 7:53-77.

188

Jancso, N. (1968) Desensitization with capsaicin and related acylarnides as a tool for
studying the function of pain receptors. In: Pharmacology of Pain-Proceedings of
Third International Congress on Pharmacology, ed. by R. K. S. Lim, p. 33-55,
Pergamon Press.

Jancso, N. and C. A. Maggi (1987) Distribution of capsiacin-sensitive urinary bladder
afferents in the rat spinal cord. Brain Res. 418:371-376.

Jancso, G., S. Karcsu, E. Kiraly, A. Szebeni, L. Toth, E. Bacsy, F. J06 and A. Parducz
(1984) Neurotoxin induced nerve cell degeneration: possible involvement of
calcium. Brain Res. 295: 211-216.

Janig, W. and M. Koltzenburg (1989) The neural basis of consciously perceived
sensations from the gut. In M. V. Singer and H. Goebell (Eds.). Nerves and the
Gastrointestinal Tract. MTP Press, Lancaster: 183-197.

J anig, W. and M. Koltzenburg (1991) Receptive properties of sacral primary afferent
neurons supplying the colon. J. Neurophysiol. 65:1067-1077.

J anig, W. and J. F. B. Morrison (1986) Functional properties of spinal visceral afferents
supplying abdominal and pelvic organs, with special emphasis on visceral
nociception. Progress in Brain Research. Elsevier Science Publishers B. V.
(Biomedical Division) 67:87-114.

Jessen, H. and L. Thuneberg (1991) Interstitial cells of Cajal and Auerbachs plexus. A
scanning electron microscopical study of guinea-pig small intestine. J.
Submicrasc. Cyto]. Pathol. 23: 195-212.

Jordan, F. L. and W. E. Thomas (1988) Brain macrophages: questions of origin and
interrelationship. Brain Res. 472(2): 165-178.

Juan, H. and F. Lembeck (1974) Action of peptides and other algesic agents on
paravascular pain receptors of the isolated perﬁrsed rabbit ear. Naunyn-
Schmiedebergs Arch. Pharmacol. 283: 151-164.

Julé, Y. and J. H. Szurszewski (1983) Electrophysiology of neurones of the inferior
mesenteric ganglion of the cat. J. Physio]. 344:277-292.

Kaczrnarek, L. K. (1987) The role of protein kinase C in the regulation of ion channels
and neurotransmitter release. Trends Neurosci. 10: 30-34.

Kampschnridt, R. F. (1981) Leucocyte endogenous mediator/endogenous pyrogen. In:
Powanda M. C., Canonico, P. G., eds. The physiologic and metabolic responses
of the host. Amsterdam: Elsevier/North-Holland: 55-74.

189

Katori, M., Y. Hori, K. Uchida, Tanaka and Y. Harada (1986) Different modes of
interaction of bradykinin with prostaglandings in pain and acute inﬂammation.
Adv. Exp. Med. Biol. 198 (B): 393-398.

Kaufman, M. P., D. G. Baker, H. M. Coleridge and J. C. Coleridge (1980a) Stimulation
by bradykinin of afferent vagal C-ﬁbers with chemosensitive endings in the heart
and aorta of dog. Cir. Res. 46: 476-484.

Kaufman, M. P., J. C. Coleridge, H. M. Coleridge and D. G. Baker (1980b) Bradykinin
stimulates afferent vagal C-ﬁbers in the intrapulmonary airways of dogs. J. Appl.
Physio]. 48: 511-517.

Keele, C. A. and D. Armstrong (1964) Substance P producing pain and itch. London:
Arnold, 1964.

Kennedy, C. and J. Krier (1987) 5-opiod receptors mediate inhibition of fast excitatory
postsynaptic potetrrials in cat parasympathetic colonic ganglia. Br. J. Pharmac.
92: 43 7-443.

King, B. F. and J. H. Szurszewski (1984) Mechanoreceptor pathways from the distal
colon to the autonomic nervous system in the guinea-pig. J. Physio]. Land 350:93-108.

Kirsner, J. B. and R. G. Shorter (1988) Inﬂammatory bowel disease. 3rd edit. Lea and
Febiger. Philadelphia.

Koch, T. R., A. Sonnenberg and J. A. Carney (1991) Gut neuropeptides and the
pathophysiology of inﬂammatory bowel disease. In Eﬂects of immune cells and
inﬂammation on smooth muscle and enteric nerves. W. J. Snape and S. M.
Collins, Eds. 169-180 CRC Press, Inc. Boca Raton.

Kreulen, D. L. , T. C. Muir and J. H. Szurszewski (1983) Peripheral sympathetic
pathways to gastroduodenal region of the guinea pig. Am. J. Physio]. 245
(Gastrointest. Liver Physiol. 8): G369-G375.

Kreulen, D. L. and Peters, S. (1986) Non-cholinergic transmission in a sympathetic
ganglion of the guinea-pig elicited by colon distension. J. Physio]. 374, 315-334.

Kreulen, D. L. and J. H. Szurszewski (1979a) Reﬂex pathways in the abdominal
prevertebral ganglia: evidence for a colo-colonic inhibitory reﬂex. J. Physio].
295:21 -32.

Kreulen, D. L. and J. H. Szurszewski (1979b) Nerve pathways in celiac plexus of the
guinea pig. Am. J. Physiol. 237(1):E90-E97.

190

Krier, J. and D. A. Hartman (1984) Electrical properties and synaptic connections to
neurons in parasympathetic colonic ganglia of the cat. Am. J. Physio]. 247:G52-
G61.

Krier, J. and J. H. Szurszewski (1982) Effect of substance P on colonic mechanoreceptors
motility and sympathetic neurons. Am. J. Physiol. 243: G259-G267.

Kumazawa, T. and K. Mizumura (1980) Chemical responses of polymodal receptors of
the scrotal contents of dogs. J. Physio]. Land 299:219-231.

Kuntz, A. and G. Saccomanna (1944) Reﬂex inhibition of intestinal motility mediated
through decentralized prevertebral ganglia. J. Neurophysiol. 7: 163-170.

Kuriyama, K., T. Hori, T. Mori and T. Nakashima (1990) Action of interferon a and
interleukin-113 on the glucose-responsive neurons in the ventromedial
hypothalamus. Brain res. Bull. 24: 803-810.

Langley, J. N. (1921) The Autonomic Nervous System, Part I, Heifer publications,
Cambridge.

Langley, J. N. and H. K. Anderson (1896a) The innervation of the pelvic and adjoining
visera. Part I. The lower portion of the intestine. J. Physiol. Land 18:67-105.

Langley, J. N. and H. K. Anderson (1896b) The innervation of the pelvic and adjoining
visera. Part II. The bladder. J. Physio]. Land 19:71-84.

Langley, J. N. and H. K. Anderson (1896c) The innervation of the pelvic and
adjoining visera. Part V. Position of the nerve cells on the course of the efferent
nerve ﬁbers. J. Physio]. Land 19:131-139.

Lawson, H. (1934) The role of the inferior mesenteric ganglia in the diphasic response of
the colon to sympathetic stimuli. Am. J. Physio]. 109: 257-273.

Lawson, H. and J. P. Holt (1937) The control of the large intestine by the decentralized
inferior mesenteric ganglia. Am. J. Physiol. 118: 780-785.

Lee, Y., K. Takami, Y. Kawai, S. Girdis, C. J. Hillyard, I. Macintyre, P. C. Emson and
M. Tohyama (1985) Distribution of the calcitonin gene-related peptide in the rat
peripheral nervous system with reference to its coexistence with substance P.
Neuroscience. 15: 1227-1237. '

Li, Z., K. Inenage, S. Kawano, H. Kannan and H. Yamashita (1992) Interleukin-113
directly excites hypothalamic supraoptic neurons in rats in vitro. Neurorepart. 3:
91 ~93.

191
Lichtrnan, J. W. (1977) The reorganization of synaptic connections in the rat

submandibular ganglion during post-natal development. J. Physiol. London. 273:
l 5 5-1 77.

Ligurnsky, L., P. L. Simon, F. Karmeli and D. Rachemilewitz (1990) Role of interleukin-
1 in inﬂammatory bowel disease-enhanced production during active disease. Gut.
31: 686-689.

Lim, R. K. S., C. N. Liu, F. Guzman and F. Braun (1962) Visceral receptors concerned in
visceral pain and psudoaffective response to intraarterial injection of bradykinin
and other algesic agents. J. Comp. Neurol. 118: 269-293.

Lin, J. and J. Krier (1993a) Excitatory inputs from distal colon/rectum to neurons in
pelvic plexus ganglia of male guinea-pig. Abstracts, Society for Neuroscience.
19: 1523.

Lin, J. and J. Krier (1993b) Central and peripheral synaptic inputs to neurons in guinea-
pig pelvic plexus ganglia. J. Gastrointestinal Motility. 5 (3):200.

Lin, J. and J. Krier (1994) Peripheral excitatory reﬂexes to colon and urinary bladder
involve mechanosensitive neurons in male guinea-pig pelvic plexus ganglia.
Abstract, Society for Neurosciences. 20: 1 12, .

Lin, J. and J. Krier (1995) Human recombinant interleukin-113 inhibits nicotinic
transmission in neurons of pelvic plexus ganglia of male guinea-pigs. Abstract,
Submitted to Gastraenterolagy.

Maggi, C. A. (1990) Tachykinin receptors in the airways and lung: what should we
block? Pharmacol. Res. 22 (5): 527-540.

Main, C., P. Blennerhassett and S. M. Collins (1993) Human recombinant interleukin 113
suppresses acetylcholine release ﬁom rat myenteric plexus. Gastroenteralogy.
104: 1648-1654.

Majno, G. (1985) Inﬂammatory mediators: Where are they going? IN; Higgs G. A.,
Williams, T. H. eds. Infalmmatory mediators. Basingstoke: MacMillan: 1-6.

March S. J., C. E. Stansfeeld, D. A. Brown, R. Davey, and D. McCarthy (1987) The
mechanism of action of capaicin on sensory C-type neurons and their axons in
vitro. Neoroscience 23: 275-290. '

Martin, H. A., A. T. Basbaum, G. C. Kwiat, E. J. Goetzl and J. D. Levine (1987)
leukotriene and prostaglandin sensitization of cutaneous high threshold C- and A-
delta mechanonociceptors in the hairy skin of rat hindlirnbs. neuroscience 22:
651-659.

192

Martin, A. R. and G. Pilar (1963) Transmission through the ciliary ganglion of the chick.
J. Physiol. London. 168: 464-475.

Marzuk, P. (1985) Biofeedback for gastrointestinal disorders: a review of the literature.
Ann. Intern. Med 103: 240-244.

McCarthy, D. O., M. J. Kluger and A. J. Vander (1986) Effect of centrally administered
interleukin-1 and endotoxin on food intake of fasted rats. Physiol. Behav. 36:
745-749.

McGehee, D. S., M. F. Gay and G. S. Oxford (1992) Involvement of the nitric oxide-
cyclic GMP pathway in the desensitization of bradykinin responses of cultured rat
sensory neurons. Neuron. 9(2): 315-324.

McLachlan, E. M. (1977) Interaction between sympathetic and parasympathetic pathways
in the pelvic plexus of the guinea pig. Proceedings of The Australian
Physiological and Pharmacological Society. 8: 15P.

McLellan, A. M. and H. Goodell (1943) Pain from the bladder, ureter, and kidney pelvic.
Research publications of the association of nervous and mental disorders. 23:
252-262.

McRorie, J ., J. Krier and T. Adams (1991) Morphology and projections of myenteric
neurons to colonic ﬁber bundles of the cat. J. Autonomic Nervous system. 32:
205-216.

Mense, S. (1977) Nervous control of skeletal muscle following chemical noxious
stimulation. J. Physiol. Land 267: 75-88.

Miller, R. J. (1987) Bradykinin highlights the role of phospholipid metabolism in the
control of nerve excitability. Trends Neurosci. 10: 226-228.

Mizel, s. B. (1989) The interleukins. FASEB J. 302): 2379-2388.

Mizumura, K., J. Sato and T. Kumazawa (1987) Effects of prostaglandins and other
putative chemical intermediaries on the activity of canine testicular polymodal
receptors studied in vitro. Pﬂu‘egers Arch. 408: 565-572.

Miztunura, K., M. Minagawa, Y. Tsujii and T. Kumazawa (1990) The effects of
bradykini agonists and antagonists on visceral polymodal receptor activities.
Pain. 40: 221-227.

193

Mrosovsky, N., L. A. Molony, C. A. Conn and M. J. Kluger (1989) Anorexic effects of
interleukin 1 in the rat. Am. J. Physio]. 257 (Regulatory Integrative Cam.
Physiol. 26): R1315-R1321.

Myers, A. C. and B. J. Undem (1991) Functional interactions between capsaicin-sensitive
and cholinergic nerves in the guinea pig bronchus. J. Pharmacol. Exp. Ther. 259:
104-109.

Myers, A. C., B. J. Undem and D. Weinreich (1990) Electrophysiological properties of
neurons in guinea pig bronchial parasympathetic ganglia. Am. J. Physiol. 259:
L403 -L409.

Myers, A. C., B. J. Undem and D. Weinreich (1991) Inﬂuence of antigen on membrane
properties of guinea pig bronchial ganglion neurons. J. App]. Physio]. 71 (3):
970-976.

Myers, A. C. and B. J. Undem (1993) Electrophysiological effects of tachykinins and
capsaicin on guinea-pig bronchial parasympathetic ganglion neurones. J. Physio].
470: 665-679.

Nakashima, T., T. Hori, T. Mori, K. Kuriyama and K. Mizuno (1989) Recombinant
human interleukin-1 beta alters the activity of preoptic thermosensitive neurons in
vitro. Brain. Res. Bull. 23(3): 209-213.

Nathan, P. W. and M. C. Smith (1953) Spinal pathways subserving defecation and
sensation from the large bowel. J. Neurol. Neurosurg. Psychiatry. 16:245-256.

Nishi, S. (1986) Electrophysiological properties of sympsthetic neurons. InzAutonomic
and Enteric Ganglia. New York: Plenum: 79-106.

Nishimura, T., J. Krier and T. Akasu (1991a) Endothelin causes prolonged inhibition of
nicotinic transmission in feline colonic parasympathetic ganglia. Am. J. Physiol.
261 (Gastrointes. Liver Physiol. 24): G628-G633.

Nishimura, T., J. Krier, and T. Akasu (1991b) Endothelin modulates calcium channel
current in neurones of rabbit pelvic parasympathetic ganglia. Br. J. Pharmacol.
103: 1242-1250.

Nishimura, T., T. Akasu and J. Krier (1992) Guanosine 3', 5'-cyclic monophosphate
regulates calcium channels in neurones of rabbit vesical pelvic ganglia. J.
Physio]. 457: 559-574.

Nishimura, T., T. Akasu and J. Krier (1993) Do depolarizing membrane oscillations
regulate spontaneous discharge of action potentials in feline parasympathetic
neurones. Society for Neuroscience Abstracts. 19: 1523.

194

Nishimura, T., J. Krier and T. Akasu (1993) Effect of vasoactive intestinal contractor on
voltage-activeatied Ca2+ currents in feline parasympathetic neurones. Am. J.
Physio]. 265 (Gastrointest. Liver Physiol. 28) : G1158-G1168.

Nja, A. and D. Purves (1977) Speciﬁc innervation of guinea pig superior cervical
ganglion cells by preganglionic ﬁbers arising from different levels of the spinal
cord. J. Physio]. London. 264: 565-583.

Norberg, K. A. (1964) Adenergic innervation of the intestinal wall studied by ﬂurescence
microscopy. Int]. J. Neurapharmaco]. 3: 379-3 82.

Ohlsson, K., P. Bjork, M. Bergenfeldt, R. Hageman, R. C. Thompson (1990) Interleukin-
1 receptor antagonist reduces mortality from endotoxin shock. Nature.
348(6301): 550-552.

Ohno, T., T. Yajirna, T. Urano and K. Nakamura (1984) Interaction of prostaglandin E2
and bradykinin in the induction of afferent spanchnic nerve discharges in cat.
Jpn. J. Pharmacol. 34: 191-203.

Papka, R. E., J. B. Furness, N. C. Della, R. Murphy and M. Costa (1984) Time course of
effect of capsaicin on ultralstructure and histochemistry of substance P-
immunoreactive nerves associated with the cardiovascular system of the guinea
pigs. Neuroscience 12: 1277-1292.

Percy, W., J. Walsh and J. Krier (1988) Morphological and electrophysiological
properties of cat lumbar paravertebral neurons. J. Autonomic Nervous system.
24: 183-192.

Perri, V., O. Sacchi and C. Casella (1970) Electrical properties and synaptic connections
of the sympathetic neurons in the rat and guinea pig superior cervical ganglion.
Pﬂuegers Arch. 314: 40-54.

Petsche, U. E. Fleischer, F. Lembeck and H. O. Handwerker (1983) The effect of
capsaicin application to a peripheral nerve an impulse conduction in functionally
identiﬁed afferent nerve ﬁbres. Brain Res. 265: 233-240.

Plata-Salaman, C. R., Y. Oomura and Y. Kai (1988) Tumor necrosis factor and
interleukin-113: Suppression of food intake by direct action in the central nervous
system. Brain Res. 448: 106-114.

Plata-Salaman, C. R. and J. M. H. Ffrench-Mullen (1992) Interleukin-113 depresses
calcium currents in CA1 hippocampal neurones at pathophysiological
concentrations. Brain Res. Bull. 29: 221-223.

195

Plata-Salaman, C. R. and J. M. H. F french-Mullen (1993) Interleukin-2 modulates
calcium currents in dissociated hippocampal CA1 neurones. Neurareport.. 4:
579-581.

Plata-Salaman, c. R. and J. M. H. Ffrench-Mullen (1994) Interleukin-113 inhibits Ca2+
channel currents in hippocampal neurones through protein kinase C. Eur. J.
Pharmacol. 266: 1-10.

Rada, P., G. P. Mark, M. P. Vitek, R. M. Mangano, A. J. Blume, B. Beer and B. G.
Hoebel (1991) Interleukin-1 beta decreases acetylcholine measured by

nricrodialysis in the hippocampus of freely moving rats. Brain Res. 550: 287-
290.

Ramilo, O., M. M. Mustafa, J. Porter, X. Saez-Lloren, J. Mertsola, K. D. Olsen, J. P.
Luby, B. Beutler and G. H. McCracken Jr. (1990) Detection of interleukin 1 beta
but not tumor necrosis factor-alpha in cerebrospinal ﬂuid of children with aseptic
meningitis. Am. J. Dis. Child. 144(3): 349-352.

Regoli, D., N.-E. Rhaleb, N.-E. Dion and G. Drapeau (1990) New selective bradykinin
receptor antagonists and bradykinin B2 receptor characterization. Trends
Pharmacol. Sci. 11: 156-161.

Rosenfeld, M. G., J-J Mermod, S. G. Amara, L. W. Swanson, P. E. Sawchenko, J. Rivier
(1983) Production of a novel neuropeptide encoded by the calcitonin gene via
tissue-speciﬁc RNA processing. Nature 304: 129-135.

Saha, J. K., J. N. Sengupta and R. K. Goyal (1990) Effect of bradykinin on opossum
esophageal longitudinal smooth muscle: evidence for novel bradykinin receptors.
J. Pharmacol. Exp. Ther. 252: 1012-1020.

Sawda, M., N. Hara and T. Maeno (1990) Extracellular tumor necrosis factor induces a
decreased K+ conductance in an identiﬁed neuron of Aplysia Kurodai. Neurosci.
Lett. 115: 219-225.

Sawda, M., N. Hara and T. Maeno (1991) Ionic mechanism of the outward current
induced by extracellular ejection of interleukin-1 onto identiﬁed neurones of
Aplysia. Brain Res. 6545: 248-256.

Semba T. (1954) Intestino-intestinal inhibitory reﬂexes. Jap. J. Physio]. 4: 241-245.
Sharkey, K. A. (1992) Substance P and celcitonin gene-related peptide (CGRP) in
gastrointestinal inﬂammation. Annals of The New York Academy of Sciences.
664: 425-442.

196

Shuttleworth, C. W., K. M. Sanders and K. D. Keef (1993) Inhibition of nitric oxide
synthesis reveals non-cholinergic excitatory neurotransmission in the canine
proximal colon. Br. J. Pharmacol. 109: 739-747.

Sims, J. E., M. A. Gayle, J. L. Slack, M. R. Alderson, T. A. Bird, J. G. Giri, F. Colotta, A.
Mantovani and K. Shanebeck (1993) Interleukin 1 signaling occurs exclusively
via the type I receptor. Proc. Nat]. Acad Sci. U. S. A. 90(13): 6155-6159.

Sims, J. E., C. J. March, D. Cosman, M. B. Widmer, H. R. MacDonald, C. J. McMahan,
C. E. Grubin, J. M. Wignall, J. L. Jackson and S. M. Call (1988) cDNA
expression cloning of the IL-1 receptor, a member of the immunoglobulin
superfamily. Science. 241(4865): 585-589.

Skoﬁtsch, G., D. M. Jacobowitz (1985) Calcitonin gene-related peptide coexists with
substance P in capsaicin sensitive neurons and sensory ganglia of the rat.
Peptides. 6: 747-754.

Skok, V. I. (1973) Physiology of autonomic ganglia. First Edition by Igaku Shoin LTD.
Steranka, L. R., D. C. Manning, Ch. J. de Haas, J. W. Ferkany, S. A. Borosky, J.
R. Connor, J. R. Vavrek, J. M. Stewart, and S. H. Snyder (1988) Bradykinin as a
pain mediator: receptors are localized to sensory neurons, and antagonists have
analgesic actions. Pro. Natl. Acad Sci. U. S. A. 85: 3245-3249.

Stylianou, E., L. A. O’Neil, L. Rawlinson, M. R. Edbrooke, P. Woo and J. Saklatvala
(1992) Interleukin 1 induces NF-kappa B through its type I but not its type II
receptor in lymphocytes. J. Bio. Chem. 267(22):]5836—15841.

Sun, W. M., N. W. Read, A. Prior, J.-A. Daly, S.K. Cheah and D. Grundy. (1990)
Sensory and motor responses to rectal distention vary according to rate and
pattern of balloon inﬂation. Gastraenteralagy 99: 1008-1015.

Suzuki, T. (1992) Bradykinin-induced responses in hamster submandibular ganglion
cells. Bull. T akyo Dent. Cal]. 33(1): 25-28.

Szurszewski, J. H. and B. F. King. (1989) Physiology of prevertebral ganglia in mammals
with a special reference to inferior mesenteric ganglia. Handbook of Physiology.
The Gastrointestinal system I. 519-592.

Tack, J. (1992) Physiology of the Enteric Nervous System in the Gastric Antrum of the
Guinea-Pig. Leuven University Press, Belgium. '

Undem, B., A. C. Myers, H. Barthlow and D. Weinreich. (1990) Vagal innervation of the
guinea pig bronchus. J. Appl. Physiol. 69: 1336-1346.

197

Undem, B. aand D. Weinreich. (1993) Electrophysiological properties and
chemosensitivity of guinea pig nodose ganglion neurons in vitro. J. Auto. Nerv.
Syst. 44(1): 17-33.

Venkova, K. and J. Krier (1993) Stimulation of lumbar sympathetic nerves evokes

contractions of cat colon circular muscle mediated by ATP and noradrenaline. Br.
J. Pharmacol. 110:1260-1270.

Venkova, K. and J. Krier (1994) Contractions mediated by alpha l-adrenoceptors and P2-
purinoceptors in a cat colon circular muscle. Br. J. Pharmacol. 112:1237—1243.

Wallis, D. J. and R. A. North (1978) Synaptic input to cells of the rabbit superior cervical
ganglion. Pﬂugers Arch. 374: 145-152.

Warner, S. J ., K. R. Auger and P. Libby (1987) Human interleukin-1 induces interleukin-
1 gene expression in human vascular smooth muscle cells. J. Exp. Med 165:
13 16-13 1 1 .

Watkins, L. R., E. P. Wiertelak, L. E. Goehler, K. P. Smith, D. Martin and S. F. Maier
(1994) Characterization of cytokine-induced hyperalgesia. Brain Res. 654: 15-
26.

Weems, W. A. and J. H. Szurszewski (1977) An intracellular analysis of some intrinsic
factors controlling neural output from inferior mesenteric ganglion of guinea pigs.
J. Neurophysiol. 41:305-321.

Willard, A. L. (1992) Excitatory and neurotoxic actions of platelet-activating factor on rat
myenteric neurones in cell culture. Ann. N. Y. AcadSci. 664: 284-292.

Wilhelm, D. L. (1973) Chemical mediators. In the inﬂammatory Process. Edt.
Zweifach, B. W., L. Grant and R. T. McCluskey: 251-301. New York: Academic
Press.

Wolf, S. (1965) The stomach. Oxford University Press, New York.

Wood, J. D. (1981) Physiology of the enteric nervous system. In: Physiology of the
Gastrointestinal Tract (lst ed.), edited by L. R. Johnson. New York: Raven, Vol.
1 : 1 -3 8.

Wood, J. D. (1989) Electrical and synaptic behavior of enteric neurons. Hand-book of
physiology. The gastrointestinal system I. 465-517.

Wood, D. D.and S. L. Gaul (1974) Enhancement of the humoral response of T-cell-
depleted murine spleens by a factor derived from human monocytes in-vitro. J.
Immunol. 1 13: 925-932.

198

Wood, J. N., J. Winter, L. F. James, H. P. Rang, J. Yeats and S. Bevan (1988) Capsaicin-
induced ion ﬂuxes in dorsal root ganglion cells in culture. J. Neurosci. 8(9):
3208-3220.

Wozniak, W. and U. Skowronska (1967) Comparative anatomy of pelvic plexus in cat,
dog, rabbit, macaque and man. Anat. Anz. Bd. 120: 457-473.

Xi, X., J. X. Thomas, Jr., W. C. Randall and R. D. Wurster (1991) Intracellular
recordings from canine intracardiac ganglion cells. J. Autonomic Nervous System.
32: 177-182.

Yau, W. M. and M. L. Youther. (1982) Direct evidence for a release of acetylcholine
from the myenteric plexus of guinea-pig small intestine by substance P. Eur. J.
Pharmacol. 81: 665-668.

Yau, W. M., K. G. Mandel, J. A. Dorsett and M. L. Youther. (1992) Neurokinin3 receptor
regulation of acetylcholine release from myenteric plexus. Am. J. Physiol. 263
(Gastrointest. Liver Physio]. 26): G659-G664.

Yokota, R. and G. Bumstock. (1983) Synaptic organisation of the pelvic ganglion in the
guinea-pig. Cell Tissue Res. 232: 379- 397.

Zeitlin, I. J ., B. J. Johnson, S. Fasth, L. Hulten and C. C. Eaglesom (1978) The kinin
system and non-adrenergic, non-cholinergic (N AN C) nerves in the colon. Adv.
Biosci. 22:269-270.

Zemig, G., P. Holzer and F. Lembeck (1984) A study of the mode and site of action of
capsaicin in guinea pig heart and rat uterus. Naunyn-schmiedebergs Arch.
Pharmac. 326: 58-63.

nrcrtrcnr'r STt-It'TE UNIV'. LIBRQRIIES
illllWillill”IUlllllllllllllIllIINWIWIIHIWI
31293014203205