 

""%'V‘IA‘II'.|

   

-I-'-v-~u.--<.u...n,.

 

 

   

 

 

 

 

 

 

 

“NW-inn"...

 

 

 

- --u¢.."‘"‘,

 

 

 

 

IVEHSITY LIBRARIE

llll’lllllllll lllllll lllllllll

3 1293 00885 3271

ll

 

 

 

 

 

 

 

 

 

This is to certify that the

dissertation entitled

Prejunctional Modulation of Acetylcholine Release
from Horse Airway Cholinergic Nerves

presented by

Zhaowen Wang

has been accepted towards fulfillment
of the requirements for

 

 

PhD degree in Physiology

 

Major professor

Date 7/22/93

 

MSU is an Affirmative Action/Equal Opportunity Institution 0-12771

JL'P

PREJUNCTIONAL MODULATION OF ACETYLCHOLINE RELEASE
FROM HORSE AIRWAY CHOLINERGIC NERVES

By

Zhaowen Wang

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Physiology

1993

PREIUIN

ABSTRACT

PREJUNCTIONAL MODULATION OF ACETYLCHOLINE RELEASE FROM
HORSE AIRWAY CHOLINERGIC NERVES
By

Zhaowen Wang

The modulatory effects of some prejunctional receptors on acetylcholine (ACh)
release from equine airway cholinergic nerves were studied. Acetylcholine release was
induced by electrical ﬁeld stimulation (EFS) of isolated airway smooth muscle
preparations suspended in 2-ml tissues baths. Prejunctional modulatory effects were
evaluated either by direct measurements of ACh release with HPLC plus electrochemical
detection or by comparing the contractile responses of airway smooth muscle to EFS and
ACh. Acetylcholine release was frequency-, voltage-, and pulse duration-dependent.
Changes in muscle preload had no influence on ACh release. Muscarinic antagonists
augmented ACh release, indicating the presence of muscarinic autoreceptors. Because
the M3 antagonist hexahydrosiladifenidol inhibited muscle contraction more potently that
ACh release, autoreceptors are not M3. Autoreceptors are unlikely to be Ml because the
augmenting effect of the Ml-receptor antagonist pirenzepine on ACh release emerged at
a concentration 240 times its K, for M1 receptors. Evidence was inadequate to either
conﬁrm or deny M2 receptors as muscarinic autoreceptors. Exogenous PGEz was more

potent in inhibiting pony trachealis contractions induced by EFS than by ACh, whereas

cyclooxyger
to ACh, su
cholinergic
PGb not e
adrenoceptor
were antagon
of inhibiton'
heaves, a tn
nerves iIme]
Prejunctiona
and the eye?
Clonid'mc he
effect in I}

mechanism

cyclooxygenase inhibitors had no signiﬁcant effect on the response of either to EFS or
to ACh, suggesting that exogenous PGEZ but not endogenous prostanoids inhibits
cholinergic neurotransmission in pony trachea. In the horse, however, neither exogenous
PGFQ nor endogenous prostanoids had any inhibitory effect on ACh release. The a2-
adrenoceptor agonists UK 14,304 and clonidine inhibited ACh release, and their effects
were antagonized by the arr-antagonists idazoxan and yohimbine, indicating the presence
of inhibitory az-adrenoceptors on horse airway cholinergic nerves. In horses with
heaves, a type of chronic obstructive pulmonary disease, ACh release from cholinergic
nerves innervating the trachealis was not different from that in normal horses and
prejunctional muscarinic autoreceptors were functionally normal. The responses to PGE2
and the cyclooxygenase inhibitor were similar to those of normal horses. However,
clonidine had less inhibitory effect on ACh release in the trachea and was without an
effect in the bronchi, suggesting a dysfunction of the prejunctional az-inhibitory

mechanism in horses with heaves.

Dﬂdicat.

Dedicated to my wife Mingfu Yu and my son Sijie Wang for their love and support.

iv

 

I an
Their suppr
lfee
N. Edward
is an excell
interested i
greatly fror
Iarr.
for their {a
A s]
MaU‘EHEn
Wordprm
of Phl'Siolol
I a]:
committee,
They Spent]

problem thi

ACKNOWLEDGMENTS

I am indebted to many people who have helped me over the past few years.
Their support has contributed tremendously to the fulﬁllment of my research.

I feel lucky to have worked on my Ph.D. dissertation under the supervision of Dr.
N. Edward Robinson. My experience with him has been pleasant and rewarding. He
is an excellent scientist and a great teacher. He encouraged me to pursue what I am
interested in and provided me with excellent guidance. My future career will beneﬁt
greatly from the training that I have received in his lab.

I am grateful to Cathy Berney, Mingfu Yu, Richard Broadstone, and Sue Eberhart
for their technical assistance and intellectual input.

A special recognition is deserved for the contributions of Victoria Kingsbury,
MaryEllen Shea, and Margaret Hofmann who, by using their expertise, have helped me
in wordprocessing or making graphics. As the administrative assistant in the Department
of Physiology Ofﬁce of Graduate Studies, Sharon Shaft has done a lot of work for me.

I also want to express my gratitude to the other members of my guidance
committee, Drs. John Chimoskey, James Galligan, Lana Kaiser, and William Spielman.
They spent their valuable time examining my experimental design, helping me solve the

problems that arose during my research, and reading and commenting on my dissertation.

 

Owing t

pharmac

vi
Owing to the nature of my study, I often had to ask Dr. Galligan questions about

pharmacology. His answers have been very valuable to me.

TABLE OF CONTENTS

LIST OF TABLES ..................................... x1
LIST OF FIGURES .................................... xn
LIST OF ABBREVIATIONS .............................. xxx
LIST OF DRUGS USED AND THEIR ACTIONS .................. xxi
INTRODUCTION ..................................... 1
CHAPTER 1: Literature Review ............................ 4
Pulmonary efferent innervation ........................... 4
Introduction................ .................... 4
Parasympathetic (cholinergic) innervation .................. 4
Adrenergic (sympathetic) innervation ..................... 9
N onadrenergic noncholinergic innervation .................. 12
Prejunctional modulation of airway cholinergic neurotransmission
and its role in airway obstruction ......................... 15
Modulation by prejunctional muscarinic receptors ............. l6
Modulation by sympathetic nerves and circulating catecholamines . . . . 20
Modulation by prostanoids ........................... 24
Modulation by histamine ............................. 26
Modulation by adenosine and adenosine nucleotides ............ 28
Modulation by vasoactive intestinal peptide and nitric oxide ....... 30
Modulation by other factors ........................... 32
Heaves .......................................... 34
Etiology ....................................... 34
Pathology ...................................... 35
Pulmonary function ................................ 36
Mechanism of airway obstruction ....................... 38
Airway hyperresponsiveness ........................ 39
Pulmonary neutrophilia ........................... 4O
Dysfunction of inhibitory NAN C nerves ................. 40
Reduced airway mucosal PGE2 production ............... 41
Increased ACh release from airway cholinergic nerves ........ 41

vii

viii

CHAPTER 2: Acetylcholine Release from Horse Airway Cholinergic Nerves:
Effects of Stimulation Intensity and Muscle Preload ........

Introduction .......................................
Materials and methods ................................
ACh analysis ....................................
Drugs ........................................
Statistical analysis .................................
Results ..........................................
Discussion ........................................

CHAPTER 3: Muscarinic Autoreceptors on Horse Airway

Cholinergic Nerves ............................
Introduction .......................................
Materials and methods ................................

Tissue preparation and equilibration ......................
Protocol .......................................
ACh analysis ....................................
Drugs ........................................
Statistical analysis .................................
Results ..........................................
Discussion ........................................

CHAPTER 4: Exogenous but not Endogenous PGE2 Modulates

Pony Tracheal Smooth Muscle Contractions .............
Introduction .......................................
Materials and methods ................................

Protocol 1: Effect of exogenous PGE2 on contractions
induced by EF S and ACh ...........................
Protocol 2: Effects of cyclooxygenase inhibitors on airway
smooth muscle contraction ..........................
Measurement of PGE2 ..............................
Drugs ........................................
Statistical analysis .................................
Results ..........................................
Protocol 1: Effect of exogenous PGE2 on contractions
induced by EFS and ACh ...........................
Protocol 2: Effects of cyclooxygenase inhibitors on tracheal
smooth muscle contraction ..........................
Discussion ........................................

45

45
48
50
52
52
52
54

7O

7O
71
7 1
72
74
74
74
75
76

83
83

86

88
88
89
89

89

9O

ix

CHAPTER 5: PGFa Inhibits Acetylcholine Release from Cholinergic

Nerves in Canine but not Equine Airways .............. 100
Introduction ....................................... 100
Materials and methods ................................ 102

Preparation of tissues ............................... 102
Protocols ...................................... 103
ACh analysis .................................... 106
Drugs ........................................ 107
Statistical analysis ................................. 107
Results .......................................... 108
Dog airways .................................... 108
Horse airways ................................... 109
Discussion ........................................ 109

CHAPTER 6: Prejunctional az-Adrenoceptors Inhibit Acetylcholine

Release from Cholinergic Nerves in Equine Airways ....... 123
Introduction ....................................... 123
Materials and methods ................................ 124

Muscle tension study ............................... 125
Protocol 1: Effect of az-receptor agonists clonidine
and UK 14,304 on EFS-induced contraction ............. 126

Protocol 2: Effect of clonidine (105 M)
and UK 14,304 (10*5 M) on the contractile response

to exogenous ACh ............................. 127

Protocol 3: Effects of 011- or az-receptor antagonists
on the actions of clonidine or UK 14,304 ............... 127

Protocol 4: Effect of yohimbine on EFS-induced

smooth muscle contraction ........................ 127
Measurement of ACh release induced by EFS ................ 128
Drugs ........................................ 129
Statistical analysis ................................. 130
Results .......................................... 131
Muscle tension study ............................... 131
ACh measurement ................................. 131
Discussion ........................................ 132

CHAPTER 7: ACh release from Airway A Cholinergic Nerves in

Horses with Airway Obstruction (Heaves) .............. 145
Introduction ....................................... 145
Materials and methods ................................ 146

Animals ....................................... 146
Preparation of tissues ............................... 147
Protocols ...................................... 148

ACh analysis .................................... 151

X

Drugs ........................................ 152
Statistical analysis ................................. 152

Results .......................................... 153
Discussion ........................................ 153
CHAPTER 8: Summary and Conclusions ........................ 164

LIST OF TABLES

Table 2-1 Detailed protocols 61
Table 5-1 Detailed protocols for dog tissues 114

Table 5-2 Detailed protocols for horse tissues 115

xi

Figure 2-1

Figure 2-2

Figure 2—3

Figure 2-4

Figure 2-5

LIST OF FIGURES

A typical chromatogram showing the separation of acetylcho-
line (ACh) and choline in a sample. The mobile phase was 50
mM NazHPO4 and the ﬂow rate was 0.5 ml/min. Two
hundred pl bath liquid was injected into the HPLC system.
The ACh and choline peaks represent 8.6 and 92.4 pmol,
respectively.

Effect of electrical ﬁeld stimulation (EFS, 20 V, 4 Hz, 2 ms)
on acetylcholine release from horse airway cholinergic nerves
in the presence of 1045 M atropine and the effect of 10*5 M
tetrodotoxin ('I'I‘X) on the EFS-induced release. * signiﬁcantly
different from the other two treatment groups. There was no
signiﬁcant difference between the "no EFS " and "TTX"
groups.

Effect of frequency on electrical ﬁeld stimulation-induced
acetylcholine release from horse airway cholinergic nerves in
the presence and absence of 10*3 M atropine. Signiﬁcantly
different from 2, 4, 8 Hz (a), 4, 8, 16 Hz (b), or 8, 16 Hz (c).
Values in the presence and absence of atropine were signiﬁ-
cantly different at all frequencies.

Effect of voltage on electrical ﬁeld stimulation-induced acetyl-
choline release from horse airway cholinergic nerves in the
presence and absence of 10*3 M atropine. Signiﬁcantly
different from 10, 15, 20 V (a), 15, 20 V (b), or 20 V (c).
Values in the presence and absence of atropine were signiﬁ-
cantly different at all voltages.

Effect of pulse duration on electrical ﬁeld stimulation-induced
acetylcholine release from horse airway cholinergic nerves in
the presence and absence of 10*3 M atropine. Signiﬁcantly
different from 0.5 ms (a), 2, 3 ms (b), or 3 ms (c). Values in
the presence and absence of atropine were signiﬁcantly
different at all pulse durations.

xii

62

63

65

66

Figure 2-6

Figure 2-7

Figure 2-8

Figure 3-1

Figure 3-2

Figure 3-3

xiii

Effect of preload on electrical ﬁeld stimulation-induced acetyl-
choline (ACh) release from horse airway cholinergic nerves in
the presence and absence of 1043 M atropine. Preload alter-
ations had no inﬂuence on ACh release. Values in the
presence and absence of atropine were signiﬁcantly different at
all preloads.

Electrical ﬁeld stimulation-induced acetylcholine release from
horse airway cholinergic nerves over 5 stimulation periods in
the presence and absence of 10‘ M atropine. Signiﬁcantly
different from S1, S4, S5 (3), or 81 (b). Values in the
presence and absence of atropine were signiﬁcantly different at
all stimulation periods.

Effects of stimulation duration alterations and hexamethonium
on acetylcholine release from horse airway cholinergic nerves
in the presence of 10‘5 M atropine. If expressed as
pmol/g/min, ACh release was the same whether the tissue was
stimulated for 15 or 30 min. Hexamethonium had no inﬂuence
on ACh release.

Effects of muscarinic antagonists on acetylcholine (ACh)
release from horse airway cholinergic nerves. All antagonists
augmented ACh release. Signiﬁcant effects appeared at 10 nM
atropine, 1 uM hexahydrosiladifenidol (HHSiD), and 10 ”M
pirenzepine and AF-DX 116. The potency and maximal effect
of atropine were much greater than those of the other antago-
nists.

ACh release in response to EFS in the time control tissue and
in tissues treated with AF-DX 116 vehicle, HHSiD vehicle,
and 10’5 M atropine. Atropine augmented ACh release. AF-
DX 116 and HHSiD vehicles had no inﬂuence on ACh release.
ACh release was constant over the six stimulation periods.

ACh release in response to 105 M atropine administered before
(Group A) and after EFS (Group B). Two periods of EFS
were applied to each group. Tissue bath samples were
collected 15 min after each period of EFS. During period 1,
no muscarinic antagonist was administered. For period 2,
atropine was administered before EFS in group A and after
EFS but prior to sample collection in group B. Atropine did
not augment ACh release in Group B, suggesting that the
augmentation was not simply a result of displacement of ACh
molecules from the receptors of the tissue into the bath liquid.

67

68

69

79

80

81

Figure 3-4

Figure 4-1

Figure 4-2

Figure 4-3

Figure 4-4

Figure 4-5

xiv

Effect of the M3-selective antagonist hexahydrosiladifenidol
(HHSiD) on the tension of horse airway smooth muscle
precontracted by 10 uM ACh. HHSiD concentration-depen-
dently relaxed the muscle, 1 uM and 10 uM eliminated about
80% and 100% of the precontraction, respectively.

Effect of exogenous PGEz on trachealis contractions induced by
EFS (Graph a) and ACh (Graph b). n = 5. Single factor
randomized design AN OVA was used to compare force at each
frequency of stimulation and at each concentration of ACh.
* = signiﬁcantly different from the corresponding value in the
vehicle-treated group.

Effect of meclofenamate and aspirin on contractions induced by
EFS in trachealis. Single factor randomized design AN OVA
was used to compare force at each frequency of stimulation.
No statistically signiﬁcant differences were observed. n = 8
in aspirin vehicle, 105 and 104 M aspirin-treated groups. n =
7 in the rest.

Effect of meclofenamate and aspirin on contractions induced by
ACh (ACh) in trachealis. Single factor randomized design was
used to compare force at each concentration of ACh. No
statistically signiﬁcant differences were found. n = 6 in 10‘8
and 10“ M meclofenamate-treated groups. n = 7 in 1045 M
meclofenamate and 103 aspirin-treated groups. n = 8 in the
rest.

Effect of meclofenamate and aspirin on PGE2 production by
tracheal strips subjected to EF S. Single factor randomized
design AN OVA was used to compare PGE2 levels at different
concentrations of the two drugs and paired t-test was used to
compare values before and after EFS. * = signiﬁcantly
different from the corresponding control value. # = signiﬁ-
cant increase in PGE2 production after EFS. + = p values of
0.051, 0.052, and 0.055 for comparison of pre- and post-EFS
PGE2 production in the meclofenamate vehicle, 10“ M
meclofenamate and aspirin vehicle-treated tissues, respectively.
n = 5 in all groups.

Effect of meclofenamate and aspirin on PGE2 production by
tracheal tissues subjected ACh. Single factor randomized
design AN OVA was used to compare PGEz production at
different concentrations of the two inhibitors and paired t-test
to compare values before and after ACh. * = signiﬁcantly

82

95

96

97

98

Figure 5-1

Figure 5-2

Figure 5-3

Figure 5-4

XV

different from the corresponding control value. # = signiﬁ-
cant increase in PGE2 production after EFS. n = 4 in 10“3 M
meclofenamate-treated group. n = 5 in the rest.

Typical chromatograms showing the effects of PGE2 and
indomethacin on acetylcholine (ACh) release from cholinergic
nerves innervating dog trachea. Each pair of chromatograms
represents the ACh amount of two samples from the same
tissue bath. Because the two samples of each pair were
analyzed a few hours apart and the detector sensitivity tended

to decrease over time, the proportionality between ACh amount '

and the peak height was not constant.

Acetylcholine (ACh) release (pmol/g/min) from cholinergic
nerves innervating dog and horse airways. The values
represent the ACh release rate during period 1 in tissues used
to study the effect of indomethacin and its vehicle. n equals 12
and 10 for dog and horse tissues, respectively. *P < 0.05
compared with the dog value.

Effect of indomethacin on acetylcholine (ACh) release from
cholinergic nerves innervating dog trachea (A) and bronchi (B).
Four periods of electrical ﬁeld stimulation were applied to two
groups (n = 6 in each group) of tissue. No indomethacin was
administered during period 1. During periods 2 to 4, one
group was treated with indomethacin (3 x 10*5 M), the other
with its vehicle. Indomethacin augmented ACh release
signiﬁcantly. Although ACh release tended to increase
progressively following indomethacin in the trachea, there was
no signiﬁcant difference among periods 2 to 4. * P < 0.05
compared with the vehicle group.

Effect of PGE; on acetylcholine (ACh) release from cholinergic
nerves innervating dog trachea (A) and bronchi (B). Four
periods of electrical ﬁeld stimulation were applied to four
groups (n = 6 in each group) of tissue which had been
pretreated with indomethacin. No PGE2 was administered
during period 1. During periods 2 to 4, one group was treated
with PGE2 vehicle, whereas the other three were treated with
different concentrations of PGE2. PGE2 inhibited ACh release
concentration-dependently and the magnitude of inhibition did
not change over time. * P < 0.05 compared with the control

group.

116

117

118

119

iigure 5-5

Figure 5-6

Figure 5-7

Figure 6-1

Figure 6-2

xvi

Effect of indomethacin on acetylcholine (ACh) release from
cholinergic nerves innervating horse trachea (A) and bronchi
(B). Four periods of electrical ﬁeld stimulation were applied
to two groups (n = 5 in each group) of tissue. No indometha-
cin was administered during period 1. During periods 2 to 4,
one group was treated with indomethacin (3 x 10*5 M) whereas
the other with its vehicle. Indomethacin had no effect on ACh
release.

Effect of meclofenamate on acetylcholine (ACh) release from
cholinergic nerves innervating horse trachea (A) and bronchi
(B). Four periods of electrical ﬁeld stimulation were applied
to two groups (n = 5 in each group) of tissue. No meclofena-
mate was administered during period 1. During periods 2 to
4, one group was treated with meclofenamate (10‘ M) whereas
the other served as control. Meclofenamate had no effect on
ACh release.

Effect of PGE2 on acetylcholine (ACh) release from cholinergic
nerves innervating horse trachea (A) and bronchi (B). Four
periods of electrical ﬁeld stimulation were applied to two
groups (n = 5 in each group) of tissue that had been pretreated
with indomethacin. No PGE2 was given during period 1.
Periods 2 to 4 were in the presence of increasing concentra-
tions of either PGE2 or its vehicle. PGFQ was without effect
on ACh release except that 107 M PGE2 augmented ACh
release signiﬁcantly in the trachea. * P < 0.05 compared
with the vehicle group.

Inhibitory effect of clonidine on the contractile response of
equine trachealis to electrical ﬁeld stimulation (EFS, 20 V, 0.5
ms). Percent inhibition was calculated from the amplitude of
muscle contraction as [(pre-drug - post-drug)/pre-drug] x 100.
The inhibitory effect of clonidine was concentration- (P =
0.0012) and frequency-dependent (P = 0.0000).

Frequency-response curves of equine trachealis strips to
electrical ﬁeld stimulation (20 V, 0.5 ms). 0 Control; [I] 104’
M clonidine, n = 6; O 10*6 M clonidine plus 10'7 M yohirn-
bine, n = 5; A 1045 M clonidine plus 1045 M yohimbine, n =
4. The inhibition of EFS-induced smooth muscle contraction
by clonidine was concentration—dependently attenuated by
yohimbine. * Signiﬁcantly different from control, + signiﬁ-
cantly different from the 10" M clonidine.

120

121

122

137

138

Figure 6-3

Figure 6-4

Figure 6-5

Figure 6-6

Figure 6-7

Figure 6-8

xvii

Frequency-response curves of equine trachealis strips to
electrical ﬁeld stimulation (20 V, 0.5 ms). 0 control, 0 10‘
M clonidine alone, n = 6; D 10‘ M prazosin plus 10‘ M
clonidine, n = 4; A 10‘ M idazoxan plus 10‘ M clonidine, n
= 3. The clonidine-induced inhibition was prevented by
pretreating the tissues with idazoxan but not with prazosin.
* Signiﬁcantly different from control, + signiﬁcantly different
from the 10‘ M clonidine.

Frequency-response curves of equine trachealis strips to
electrical ﬁeld stimulation (20 V, 0.5 ms). 0 control; D 10‘
MUK14,304,n=4;<>107MUK14,304,n = 3;A10‘
M UK 14,304, 11 = 4; V 107 M idazoxan added 20 min before
the addition of 107 M UK 14,304, n = 3. * Signiﬁcantly
different from control, + signiﬁcantly different from the 107
M UK 14,304.

Effect of yohimbine (10‘ M) on the contractile response of
equine trachealis strips to electrical ﬁeld stimulation (EFS, 20
V, 0.5 ms) in the absence of guanethidine. Yohimbine
increased the response to 0.1 Hz and this increase was not
affected by pretreating the tissues with guanethidine (105 M;
data not shown). * Signiﬁcant effect of yohimbine.

Effects of az-receptor agonists and antagonists on acetylcholine
(ACh) release from equine trachealis strips in response to
electrical ﬁeld stimulation (EFS, 20 V, 0.5 ms, 0.5 Hz). *
Signiﬁcantly different from time control, + signiﬁcantly
different from UK 14,304 (A) or clonidine (B). A shows the
inhibition of ACh release by UK 14,304 and the antagonizing
effect of idazoxan. B shows the inhibition of ACh release by
clonidine and the antagonizing effect of yohimbine. C gives
the ACh release rate expressed as pmol/g/min at a clonidine
concentration of 10‘ M.

Effect of clonidine on acetylcholine release from equine
bronchial rings in response to electrical ﬁeld stimulation (EFS,
20 V, 1 ms, 2 Hz). * Signiﬁcantly different from time
control.

Effect of frequency (A; 20 V, 0.5 ms) and voltage (B; 0.5 Hz,
0.5 ms) on clonidine-induced inhibition of acetylcholine (ACh)
release from equine trachealis strips in response to electrical
ﬁeld stimulation. Data were normalized to an initial measure-
ment of ACh release (2 Hz, 20 V, 0.5 ms) obtained prior to

139

140

141

142

143

144

Figure 7-1

Figure 7-2

Figure 7-3

Figure 7-4

xviii

addition of drug. * Signiﬁcant difference between time control
and clonidine.

Typical chromatograms showing the effect of atropine on ACh
release from cholinergic nerves innervating horse trachea. The
sensitivity of the electrochemical detector was set at 100 nA
during the ﬁrst 5 min of each run and at 2 nA thereafter.
Twenty-ﬁve pl tissue bath liquid was injected for each sample.
Atropine concentration—dependently augmented ACh release.

Effect of atropine (10‘8—10s M) on ACh release from choliner-
gic nerves innervating the trachea (A) and bronchi (B) of
normal and heavey horses. Acetylcholine release rate during
the ﬁrst EFS period was regarded as 1, and the rate during
subsequent periods was calculated as multiples of the rate of
period 1. There was no difference in the augmenting effect of
atropine between normal and heavey horses.

Effect of clonidine (107-105 M) on ACh release from choliner-
gic nerves innervating the trachea (A) and bronchi (B) of
heavey horses. Clonidine inhibited ACh release concentration-
dependently in the trachea but was without an effect in the
bronchi.

Effects of PGE2 (10'9—10‘7 M) (A) and indomethacin (3 x 10‘
M) (B) on ACh release from cholinergic nerves innervating the
heavey horse bronchi. Neither PGE2 nor indomethacin showed
any effect on ACh release.

160

161

162

163

 

4-DAMP
a-MeHA
APplmax
ACh

AChE

EF S
EJPs

EPSPs

LIST OF ABBREVIATIONS

4—diphenylacetoxy-N-methylpipcridine
oz-methylhistamine

maximal change in intrapleural pressure
acetylcholine

acetylcholinesterase

adenosine diphosphate

adenosine monophosphate

adenosine triphosphate

cyclic adenosine monophosphate
dynamic compliance

cyclic guanosine monophosphate
carbocyclic thromboxane A2
dirnethylphenylpiperaziniurn

molar concentration of an antagonist required to reach 50% of its
maximal augmenting effect

electrical ﬁeld stimulation
excitatory junction potentials

excitatory postsynaptic potentials

xix

{ACU

L-IHSiD

NANC

NE

N0

PGES

PHI

PHM

VIP

XX

high-afﬁnity choline uptake
hexahydrosiladifenidol
high-performance liquid chromatography
monoamine oxidase
nonadrenergic noncholinergic
norepinephrine

nitric oxide

prostaglandins of the E series
peptide histidine isoleucine
peptide histidine methionine
pulmonary resistance

vasoactive intestinal peptide

I

5 v
1 D
J
_¢

'—

nk '-

l

LIST OF DRUGS USED AND THEIR ACTIONS

acetylcholine (ACh)
AF-DX 1 16
aspirin
atropine
clonidine
guanethidine
hexahydrosiladifenidol
hexamethonium
idazoxan
indomethacin
meclofenamate
neostigmine
PGE,
pirenzepine
prazosin
tetrodotoxin
UK 14,304

yohimbine

a non-selective muscarinic receptor agonist
a muscarinic Mz-receptor antagonist
a cyclooxygenase inhibitor
a non-selective muscarinic receptor antagonist
an az—adrenoceptor agonist
a sympathetic nerve blocker
a muscarinic M3-receptor antagonist
a ganglionic blocker
an ozZ-adrenoceptor antagonist
a cyclooxygenase inhibitor
a cyclooxygenase inhibitor
an acetylcholinesterase inhibitor
a cyclooxygenase product
a muscarinic Ml-receptor antagonist
an al-adrenoceptor antagonist
a neural blocker
an az-adrenoceptor agonist

an az-adrenoceptor antagonist

xxi

INTRODUCTION

The cholinergic innervation is the predominant neural pathway in mammalian
airways. Its activation releases the neurotransmitter acetylcholine (ACh) , which causes
airway smooth muscle contractions by acting on muscarinic receptors on the muscle cell
membranes. Acetylcholine release from the airway cholinergic nerves may be modulated
by the activation of a number of receptors located on the nerve terminals. These
prejunctional receptors probably play an important role in the regulation of airway
cholinergic neurotransmission. There is evidence to suggest that certain prejunctional
receptors are dysfunctional in humans with asthma and in guinea pigs infected with
parainﬂuenza virus or challenged by antigen. An abnormality of prejunctional receptors
may lead to increased ACh release and, therefore, be a mechanism of the airway
obstruction or hyperresponsiveness present under these conditions.

Most of our knowledge about prejunctional modulation of airway cholinergic
neurotransmission comes from comparison of the effect of a factor on the contractile
response of airway smooth muscle to exogenous ACh and to nerve activation that releases
endogenous ACh. A greater effect on the response to nerve activation than to exogenous
ACh would be interpreted as prejunctional modulation. To use this approach, it is
necessary to assume that exogenous and endogenous ACh behave similarly. However,

because the amount of neurally released ACh is much less than the exogenously

2

administered ACh, contractions induced by nerve activation are more likely to be
inﬂuenced by modulators of muscle contractions than those induced by exogenous ACh.
Besides, various neurotransmitters may be released from all kinds of nerves during
electrical ﬁeld stimulation (EF S) and may therefore affect the response to EFS but not
to exogenous ACh. In addition, if a factor has an effect on gap junctions, it may
inﬂuence the response to EFS but not to exogenous ACh. Furthermore, if a potential
neuromodulator has a direct effect on airway smooth muscle, it would be difﬁcult to
differentiate its prejunctional effect from the postjunctional effect. For these reasons, the
comparison of responses to EF S and ACh may not always provide the correct conclusion
about ACh release. Knowledge about prejunctional modulation would be greatly
extended by direct measurement of ACh release from airway cholinergic nerves.

Our laboratory has a long-term interest in studying heaves, a type of chronic
obstructive pulmonary disease that affects horses and ponies. Affected equids develop
airway obstruction when stabled and fed hay and enter remission at pasture. During
airway obstruction, horses have increased pulmonary resistance (R), increased maximal
change in intrapleural pressure (APplm) during tidal breathing, and decreased dynamic
compliance (Cm). The disease bears several major similarities to human asthma such
as allergic etiology, spontaneous occurrence, and bronchiolitis. Therefore, information
about the mechanism of this disease may help us to understand the mechanism of human
asthma.

In heavey horses, intravenous administration of muscarinic antagonists such as
atr0pine alleviate the airway obstruction markedly, indicating the involvement of a

muscarinic mechanism. In tissue baths, isolated airway smooth muscle preparations from

3
heavey horses are hyporesponsive to exogenous ACh but hyperresponsive to EF S

compared with those of the control horses, suggesting that ACh release from airway
cholinergic nerves is probably increased in the heavey horse. One possible mechanism
for this increased ACh release is altered prejunctional modulation.

During my Ph.D. dissertation work, I succeeded in adapting an ACh analysis
technique to measure ACh release from the horse airway cholinergic nerves. Using this
technique and the more traditional approach of comparing the responses to EFS and
ACh, I studied the prejunctional modulatory effect of muscarinic autoreceptors, a2-
adrenoceptors, and prostanoid receptors on ACh release from airway cholinergic nerves
in control and heavey horses. The dissertation is presented as a series of chapters that
can be read independently. Chapter 1 is a literature review on airway innervation,
prejunctional modulation of airway cholinergic neurotransmission, and heaves. Chapters
2-7 are descriptions of my experiments, and Chapter 8 is a summary of the major results

and conclusions.

CHAPTER 1

LITERATURE REVIEW

Pulmonary efferent innervation
Introduction
Three neural pathways may be found innervating the lungs: cholinergic,
adrenergic, and nonadrenergic noncholinergic. These neural pathways form two plexuses
in the lung: peribronchial and periarterial plexuses. The peribronchial plexus further
divides into extrachondrial and subchondrial plexuses according to their relation to the
cartilage (Laitinen and Laitinen 1991). This chapter discusses the structure and function

of pulmonary efferent innervation. Where possible, studies on equine airways have been

emphasized.

Parasympathetic (cholinergic) innervation

The parasympathetic pathway originates at the vagal nuclei of the brain stem and
passes to the lung in the vagus (Barnes 1986). These efferent (preganglionic or
presynaptic) ﬁbers synapse with postganglionic neurons in ganglia located adjacent to
effector cells. The preganglionic nerve terminals release ACh, which acts on nicotinic
receptors on the plasma membrane of the postganglionic neurons. The axons of the

postganglionic neurons innervate airway smooth muscle, submucous glands, and,

U

possibly.
muscle c

It
in all sp
(AChE) 5

terminal

5

possibly, blood vessels. Activation of airway cholinergic nerves causes airway smooth
muscle contraction, mucus secretion, and vasodilation.

In airways, the parasympathetic innervation is the predominant neural pathway
in all species. Results obtained with histochemical staining of acetylcholinesterase
(AChE) suggest that cholinergic nerves are distributed in airways from trachea down to
terminal bronchioles (Partanen et al. 1982; Sheppard et al. 1983). The density of
cholinergic innervation is greatest in proximal airways and diminishes peripherally
(Barnes 1986); so does the density of muscarinic receptors on the airway smooth muscle
cell membrane (Barnes et al. 1983). Stimulation of the parasympathetic nerves causes
constriction from the trachea to the small airways with resting diameters of 1 mm but
less signiﬁcant contraction in airways smaller than 0.5 mm in diameter (Nadel et al.
1971).

By taking advantage of the universal distribution of AChE in neurons, Baker et
al. (1986) have provided a detailed picture about the architecture of nerves and ganglia
in ferret trachea by the use of AChE histochemistry. The most distinctive feature of the
neuronal architecture of the trachea is the longitudinal nerve trunks located at the border
of the trachealis muscle and cartilaginous rings. Branches of the longitudinal nerve
trunks divide and anastomose, thus forming three plexuses: one superﬁcial plexus on
each side of the dorsal tracheal membrane and one deep muscle plexus. There are
numerous ganglia containing one to four ganglion cell bodies located in the superﬁcial
plexuses and fewer ganglia containing 10—38 cell bodies dispersed along the longitudinal

nerve trunks.

 

and pos
from v2
varicosi
line act:
Minette
phospho
(PIPa) [C
from int
May S

[116 prege

CCHS' Alt

6

The autonomic neuroeffector junction is not a "synapse" where there are both pre-
and post-synaptic specialized structures, but a "junction" where transmitter is released
from varicosities in extensive branching terminal nerve ﬁbers. The distance between
varicosities and effector cells varies from 20 nm to 2 um (Burnstock 1988). Acetylcho-
line acts on M, receptors on the smooth muscle cell membrane (Brichant et al. 1990;
Minette and Barnes 1990; Yu et al. 1992a) and activates the membrane-bound enzyme
phospholipase C, which catalyzes the hydrolysis of phosphatidylinositol 4,5-bisphosphate
(PIPZ) to inositol 1,4,5-triphosphate (1P3) and diacylglycerol (DAG). 1P3 releases Ca2+
from intracellular stores and induces airway smooth muscle contration. Contraction of
airway smOOth muscle appears to be independent of changes in membrane potential or
the presence of extracellular Ca2+ (Hughes et al. 1990).

Acetylcholine is formed from acetyl coenzyme A and choline under the catalysis
of choline acetyltransferase (Tucek 1988) and stored in vesicles in the cholinergic nerve
terminals. When the neuron is stimulated, it may deve10p an action potential that Opens
voltage-dependent Ca2+ channels. Inﬂux of Ca2+ causes the vesicles to fuse with the
plasma membrane of the nerve terminals and release their contents (Kandel 1985). The
released ACh can act on nicotinic or muscarinic receptors on the effector cells and the
action is quickly terminated due to the breakdown of ACh to choline and acetate by
cholinesterase.

Acetyl coenzyme A is an intermediary product in the metabolism of carbohy-
drates, fatty acids, and several animo acids and is therefore present in all types of animal
cells. Although there is evidence that choline can be synthesized de novo in nerve

terminals (Blusztajn and Wurtrnan 1981), it is not clear to what extent, if any, this

 

 

 

 

choline
of AC]
uptake
uptake
perhaps
substrat
released
(
body an
ﬁber. ,6
are fouh

Concentr

be demc
COHIIacri
choline”
gangleD
bCCause ;
(DMPp)
of the PC
ngangh

Ir.
blocked i

7

choline contributes to the intraterminal pool of free choline and the subsequent synthesis
of ACh. Choline can be brought into the cytoplasm through a high-affmity choline
uptake (HACU) system. The process is energy-, sodium-, and voltage-dependent. The
uptake is enhanced when cholinergic terminals are depolarized. The HACU system is
perhaps the main source of choline for the synthesis of releasable ACh. Most of the
substrate for this system comes from recapture of choline produced by the hydrolysis of
released ACh (Ducis 1988).

Choline acetyltransferase and AChE are perhaps synthesized in the nerve-cell
body and are carried to the nerve endings by the axoplasmic current streaming down each
ﬁber. At the endings they accumulate. Therefore, although the two enzymes and ACh
are found in signiﬁcant amounts along the whole length of every cholinergic axon, the
concentration is highest in the nerve terminals (MacIntosh 1959).

The presence of a functional cholinergic innervation in airway smooth muscle can
be demonstrated by either direct vagal stimulation or EFS of isolated tissues. If the
contraction can be blocked by tetrodoxin and atropine, it is due to activation of
cholinergic nerves. Contractions induced by vagal stimulation may be abolished by the
ganglionic blocker hexamethonium, whereas those induced by EFS are unaffected
because EFS directly activates the postganglionic neurons. Dimethylphenylpiperazinium
(DMPP) induces airway smooth muscle contraction by activating the nicotinic receptors
of the postganglionic neurons. Therefore, the effect of this drug is independent of the
preganglionic neurons.

In the horse, EFS induces airway smooth muscle contractions, which can be

blocked by tetrodotoxin or atropine, indicating the presence of cholinergic innervation.

The cholinerg'
aresring outsi
et al. in press
inhibitory syn
alter blockade
humkno
mnetion is pr

There
because admi
al. 1988). H
sWiles such .

al. 1965; Wit

8

The cholinergic innervation is present from the trachea down to peripheral airways with
a resting outside diameter of 4-5 mm (Broadstone et al. 1991; LeBlanc et al. 1991; Yu
et al. in press b). No experiments have been conducted with smaller airways. Although
inhibitory sympathetic and NAN C innervation can also be demonstrated in the trachea
after blockade of the cholinergic function, the contractile response to EFS is not affected
by blockers of the inhibitory nerves (Yu et al. in press b), indicating that the cholinergic
function is predominant.

There appears to be no basal cholinergic tone in the airways of normal equids
because administration of atropine has no inﬂuence on airway resistance (Broadstone et
al. 1988). However, there may be a certain degree of cholinergic tone in some other
species such as humans (Vincent et al. 1970; de Troyer et al. 1979) and cats (Olsen et
al. 1965; Widdicombe et al. 1961). An increase in airway cholinergic activity could be
an important mechanism in the development of airway obstruction and hyperresponsive-
ness (De Jongste et al. 1991). This may result from three mechanisms: 1) increased
central efferent ﬁring; 2) reﬂex activation of cholinergic pathway due to stimulation of
sensory receptors; 3) abnormality in prejunctional receptors that regulate ACh release.

Ganglia, clusters of neuronal cell bodies, are not just simple relay stations
between the pre- and post-ganglionic neurons. Their functions also may include the
following:

Filtering function—Not all the action potentials of the preganglionic neurons are
relayed to the postganglionic neurons. In the airway ganglia, there is a type of neuron
called the AH cell, which is characterized by a single action potential followed by a large

and long-lasting after-hyperpolarization. This post-action potential after-hyperpolarization

 

effectively prev
activation of pr
Site of l
to neurons in 1]
input may infl:
al. observed
cholinergic n
muscarinic re
[hf-’56 recepto
Barnes 1989c
A siir.
may Serve as
STUPathetic

llO EVidencg

9

effectively prevents some of the excitatory postsynaptic potentials (EPSPs) induced by
activation of presynaptic neurons from developing into action potentials (Coburn 1987).

Site of neuromodulation—Adrenergic nerve terminals are seen in close apposition
to neurons in the ganglia (Jacobowitz et al. 1973), raising the possibility that sympathetic
input may inﬂuence airway cholinergic activity. In support of this possibility, Baker et
al. observed that the adrenergic nerve neurotransmitter norepinephrine inhibits
cholinergic neurotransmission in airway ganglia (Baker et al. 1983). Excitatory
muscarinic receptors (M1 subtype) may be also present in the ganglia, and activation of
these receptors facilitates airway cholinergic neurotransmission (Bloom et al. 1988;
Barnes 1989c).

A simple integration center—The complexity of airway ganglia suggests that they
may serve as simple integration centers for processing inputs from cholinergic excitatory,
sympathetic and nonadrenergic inhibitory, and afferent nerves (Barnes 1986). There is
no evidence yet that peripheral airway ganglion neurons may generate signals
independent of input from the central nervous system (Coburn 1987), although such is
the case in the myenteric plexus (Wood 1984), where central input has only a modulatory

function.

Adrenergic (sympathetic) innervation

The sympathetic nerve supply to the lung originates from the lateral horn of the
upper six thoracic segments of spinal cord. The preganglionic ﬁbers leave the spinal
cord through the ventral roots and synapse in the middle and inferior cervical ganglia and

the upper four thoracic paravertebral ganglia. Postganglionic ﬁbers then enter the lungs

at the hilum ant
Bent and Lev:
nicotinic recer
postganglionic

The po:
submucuous gl
causes vasoco
Vasoconstrictit
of airway $11104
smoorh muscle
nicePIOTS, but
lBanners 19893
dcmODStrated i
Only Under Cer

The sy
by tiTOSine h}-
DOpam'me 84'.
HeTilting is Stc

10

at the hilum and intermingle with the cholinergic nerves to form plexuses (Barnes 1986;
Berne and Levy 1988a). The preganglionic nerve terminals release ACh, which acts on
nicotinic receptors on the plasma membrane of the postganglionic neurons. The
postganglionic nerve terminals release norepinephrine (NE).

The postganglionic sympathetic nerve terminals innervate blood vessels, airway
submucuous glands, and airway smooth muscle. Activation of the adrenergic nerves
causes vasoconstriction, mucus secretion, and airway smooth muscle relaxation.
Vasoconstriction is due to the activation of cit-adrenergic receptors, whereas relaxation
of airway smooth muscle results from binding of N E to B-receptors. In the dog tracheal
smooth muscle, it appears that relaxation to exogenous B-agonists is mediated by 32-
receptors, but relaxation to sympathetic nerve stimulation is mediated by til-receptors
(Barnes 1989a). oz-adrenoceptors mediating smooth muscle contractions have also been
demonstrated in the airways of some species. However, their function can be observed
only under certain conditions such as following 6-adrenergic blockade (Leff et a1. 1986).

The synthesis of NE is a multiple-step process. Tyrosine is converted to L-dopa
by tyrosine hydroxylase. L-dopa is converted to dopamine by a speciﬁc decarboxylase.
Dopamine B—hydroxylase converts dopamine to NE (Berne and Levy 1988b). Norepi—
nephrine is stored in vesicles in the nerve terminal and released upon nerve stimulation.
The released N E acts on a— or B-receptors on the effector cells. Its action is quickly
terminated due to three mechanisms: 1) the majority of released NE is removed by
reuptake into the nerves, where it is either reincorporated into vesicular stores or
degraded by monoamine oxidase (MAO); 2) some NE is taken up by smooth muscle cells

and inactivated by intracellular MAO or catechol-O-methyltransferase; 3) some NE leaks

away into the ci
kidney.

The dis
generally spars
abundant adren
goat. sheep, pi;
inthebronchia
in horse airwa
ul‘ll‘ir trachea
WIVES have
(Jacobowitz e
activity

Adren
1” main. it
function of C
in the airwa:
directly. In
Inllscle is s

COHUHCIIOnS

11

away into the circulation and is inactivated by the same enzymes, largely in the liver or
kidney.

The distribution of adrenergic nerve terminals in airway smooth muscle is
generally sparse; however, there is considerable variation between species. There are
abundant adrenergic ﬁbers in the bronchial muscles of the cat (Dahlstrom et al. 1966),
goat, sheep, pig, calf (Mann 1971), and dog (Knight et al. 1981), whereas few are found
in the bronchial muscles of the rabbit (Mann 1971) and man (Pack and Richardson 1984).
In horse airways, the inhibitory function of sympathetic nerves can be demonstrated in
upper trachea but not in lower trachea and bronchi (Yu et al. in press b). Adrenergic
nerves have been found in close association with airway parasympathetic ganglia
(Jacobowitz et al. 197 3), which is perhaps important for modulation of the cholinergic
activity.

Adrenergic innervation is the major neural pathway of blood vessels in the lungs.
In airways, it is not as important as in the blood vessles. Adrenergic nerves oppose the
function of cholinergic nerves. In species with abundant direct adrenergic innervation
in the airway smooth muscle, the adrenergic nerves may inhibit muscle contractions
directly. In species where direct innervation of adrenergic nerves to airway smooth
muscle is scanty, adrenergic nerves may still modulate airway smooth muscle

contractions by acting on airway cholinergic nerves.

 

14.». Ann".

Nonadrenergic

1n the at
either adrenergi
Both inhibitory

The pre
synapse in the
parasympatheti
neurons of the
bl" aeting on m‘
[ﬁlm of the p
fr0111 neurons
V3$0aCtive ime
nerves, may a

Inhibitc
Titties includ
{Mitchell (:1 31
Pigs (Mylo: et

Innervati0n in

12

Nonadrenergic noncholinergic innervation

In the airways, there are neurally mediated responses that cannot be abolished by
either adrenergic or cholinergic blockers, indicating the presence of N AN C innervation.
Both inhibitory and excitatory NAN C innervations may be demonstrated in the airways.

The preganglionic ﬁbers of inhibitory NAN C innervation run in the vagus and
synapse in the peripheral airway ganglia with postganglionic neurons. Similar to the
parasympathetic and sympathetic nervous system, the neurotransmitter of preganglionic
neurons of the inhibitory NAN C system is ACh, which activates postganglionic neurons
by acting on nicotinic receptors in the ganglia. It is not known whether the neurotrans-
mitter of the postganglionic neurons is released from specialized inhibitory neurons or
from neurons that also contain ACh, although ultrastructural studies suggest that
vasoactive intestinal peptide (VIP), a possible neurotransmitter of the inhibitory NAN C
nerves, may coexist in the same nerve terminals as ACh (Laitinen et al. 1985b).

Inhibitory NAN C innervation has been demonstrated in the airways of several
species including humans (Richardson and Beland 1976; Taylor et al. 1984), pigs
(Mitchell et al. 1990), horses (Broadstone et al. 1991; Yu et al. in press b), and guinea
pigs (Taylor et al. 1984; Grundstrom et al. 1981b). The importance of inhibitory NANC
innervation in the control of airway smooth muscle tension varies between species. In
human (Richardson and Beland 1976), pig (Mitchell et al. 1990), and horse (Yu et al.
in press b) airways, it is the primary inhibitory nervous system. In contrast, dog airways
lack an NANC innervation (Russell 1980).

There is no general agreement about the neurotransmitters of the inhibitory

NANC nerves. Nitric oxide (NO) (Yu et al. in press b; Kannan and Johnson 1992; Li

 

and Rand 1‘.
Mmﬂﬂ
mﬂkw

Vw
wmnm
kaMu
mmmmz
ofuachea
terrodotor
from ner‘
Sm00thr

They are

thee

thei

Z9

13
and Rand 1991; Tucker et al. 1990) and neuropeptides such as VIP, peptide histidine

isoleucine (PHI), and peptide histidine methionine (PHM) (Barnes 1986) are among
possible candidates.

Vasoactive intestinal peptide is a 28 amino acid peptide (Barnes 1986). It is a
potent relaxant of airway smooth muscle (Palmer et al. 1986). Vasoactive intestinal
peptide has been localized to neurons and nerve terminals in airway smooth muscle and
submucosal glands (Laitinen et al. 1985b; Dey et al. 1981). Electrical ﬁeld stimulation
of tracheal preparations releases VIP into the bathing medium, and this is blocked by
tetrodotoxin (Cameron et al. 1983; Matsuzaki et al. 1980), indicating that VIP is released
from nerves. Peptide histidine isoleucine and PHM are also potent relaxants of airway
smooth muscle (Palmer et al. 1986; Lundberg et al. 1984a; Christoﬁdes et al. 1984).
They are structurally related to VIP.

Nitric oxide is widely known as the endothelium—derived relaxing factor in blood
vessels. It has a powerful ability to relax smooth muscle. Nitric oxide appears to be the
neurotransmitter or mediator of the inhibitory N AN C nerves in the airways of pigs
(Kannan and Johnson 1992), horses (Y u et al. in press b), and guinea pigs (Li and Rand
1991; Tucker et al. 1990). Conﬂicting data have been reported in cat airways: in one
study conducted with isolated trachealis, nitric oxide was proposed as the primary, if not
the only, mediator (Fisher et al. 1993), but in another study, nitric oxide did not mediate
the inhibitory NANC relaxation in vivo or in vitro (Diamond et al. 1992).

Vasoactive intestinal peptide and N O relax smooth muscle through different
effector pathways. Nitric oxide stimulates soluble guanylate cyclase to increase cyclic

guanosine monophosphate (cGMP), whereas VIP stimulates adenylate cyclase to promote

 

he accumul
which the cy
The accumu
concentratio
In he
with histamf
relaxation d
The degree
generation.
relaxation,
artirtine. '
aI1d centra
Press b).
A I
Some 0the
of “0311211
inhibitor}.
horses in
be deten

QuSe 01’

I
SpeCies.

Rich as

14

the accumulation of cyclic adenosine monophosphate (CAMP). The mechanisrrrs by
which the cyclic nucleotides produce smooth muscle relaxation are not fully understood.
The accumulation of either cAMP or cGMP is thought to reduce intracellular Ca2+
concentration (Said 1992).

In horse airway tissues incubated with atropine and propranolol and precontracted
with histamine, EF S produces frequency-dependent relaxation and the magnitude of the
relaxation decreases from trachea to central bronchi and is absent in peripheral airways.
The degree of relaxation in bronchi is not simply a function of bronchial size or
generation. NG-nitro-L-arginine, a nitric oxide synthase inhibitor, eliminates the
relaxation. ‘ This effect is reversed by L-arginine, the NO precursor, but not by D-
arginine. These results suggests that inhibitory N AN C nerves supply only the trachea
and central bronchi and that NO mediates inhibitory NAN C function (Yu et al. in
press b).

A dysfunction of the inhibitory NAN C nerves is probably related to asthma and
some other types of airway obstruction. Immunoreactive VIP is present in lung tissue
of normal humans but absent in that of asthmatics (Ollerenshaw et al. 1989). Functional
inhibitory NAN C innervation can be demonstrated in the central bronchi of normal
horses but not of horses with airway obstruction (Broadstone et al. 1991). It remains to
be determined if the dysfunction of inhibitory NAN C nerves in the heavey horses is a
cause or a result of airway inﬂammation and obstruction.

Excitatory NAN C innervation has also been demonstrated in the airways of a few
species. The neurotransmitter of the excitatory N AN C nerves is probably a tachykinin

such as substance P or neurokinin A. Substance P appears to be the most likely

 

 

ennui
brooch
Ander.
by sub
Ether
in the

Minn

Upon

throug
nerve
(Lunt
ofex
REE“
the p

tach)

dOm
1986

ObSh

15

candidate. For example, EFS of guinea pig bronchi produces a component of
bronchoconstriction that cannot be prevented by atropine (Lundberg et al. 1983c;
Andersson and Grundstrom 1983; Grundstrom et al. 1981). This response is mimicked
by substance P and antagonized by substance P antagonists, suggesting that substance P
is the neurotransmitter of the excitatory NAN C nerves. Subtance P is localized to nerves
in the airways of several species (Lundberg et al. 1984b; Polak and Bloom 1982;
Wharton et a1. 1979). The nerves that contain substance P appear to be afferent nerves
(Barnes 1986). Damage to airway epithelium may expose sensory nerve terminals which,
upon stimulation by inﬂammatory mediators, may lead to the release of substance P
through an axon reﬂex (Barnes 1986). Capsaicin releases substance P from sensory
nerves. Acute treatment with capsaicin induces airway smooth muscle contraction
(Lundberg et al. 1983b), whereas chronic treatment with capsaicin eliminates the function
of excitatory NAN C nerves due to the depletion of substance P (Lundberg et al. 1983a).
Removal of epithelium increases the constrictor effect of substance P, probably due to
the predominant localization of neural endopeptidase, the major degrading enzyme for

tachykinins, in airway epithelium (Barnes 1991).

Prej unctional modulation of airway cholinergic neurotransmission and its role in
airway obstruction

Control of airway smooth muscle tension in humans and most animal species is
dominated by excitatory neural inputs transmitted by cholinergic motor nerves (Barnes
1986). The fact that muscarinic antagonists such as atropine frequently alleviate airway

obstruction in some airway diseases suggests that activation of muscarinic receptors by

ACh release
of airway I
cholinergic
prostanoids
oxide. So:
other types
airway ob:

Pr:
deending
transmitte
heterorec
rt=Ct3ptors
nomencl

dETmed

ch01lnc

regular

Mode

th

9?

IQI

16

ACh released from airway cholinergic nerves plays an important role in the development
of airway obstruction (De J ongste et al. 1991). Acetylcholine release from airway
cholinergic nerves may be modulated by a number of factors, including ACh,
prostanoids, catecholamines, histamine, adenosine, serotonin, neuropeptides, and nitric
oxide. Some of these factors are often found at abnormal levels in asthma and some
other types of airway obstruction; therefore, they may contribute to the development of
airway obstruction by augmenting ACh release.

Prejunctional receptors are frequenctly classiﬁed as auto- or heteroreceptors
depending on the origin of the modulator. Autoreceptors respond to the neuron’s own
transmitter. They are involved in negative or positive feedback modulation, whereas
heteroreceptors respond to agonists other than the neuron’s own transmitter. Hetero-
receptors are further divided according to the agonists with which they interact. In some
nomenclatures, the term "homoreceptor" is also included. However, it is not uniformly
deﬁned (Vizi et al. 1991; Kalsner 1990a).

In this review, I summarize the current knowledge about modulation of airway
cholinergic neurotransmission and discuss the possible relationship between defective

regulation of airway cholinergic neurotransnrission and airway obstruction.

Modulation by prejunctional muscarinic receptors

Muscarinic receptors are not homogenous. As many as ﬁve different receptor
proteins have been cloned. By the use of selective muscarinic antagonists, the existence
of at least three functionally different subtypes has been recognized. Muscarinic recep-

tors with a high afﬁnity for pirenzepine are called Ml receptors, those with a high

afﬁnity
high aft
siladifer
that, in
furrcrion
receptor
ACh rel

l
mmxu
choline
nonSper

results

17
afﬁnity for methoctramine and AF-DX 116 are called M2 receptors, and those with a

high afﬁnity for 4-diphenyl acetoxy N-methyl piperidine (4-DAMP) and hexahydro-
siladifenidol are called M3 receptors (Minette and Barnes 1990). Most evidence suggests
that, in the airways, functional M3 receptors are present on smooth muscle cells, while
functional M, and M2 receptors are present on prejunctional sites. Activation of M3
receptors leads to muscle contraction, while activation of M, and M2 receptors modulates
ACh release from cholinergic nerves (Minette and Barnes 1990; Barnes 1989b).

M, receptors may be present in airway parasympathetic ganglia. In atopic
subjects, pirenzepine inhibits bronchoconstriction induced through reﬂex activation of
cholinergic nerves by sulfur dioxide but not that induced by methacholine, whereas the
nonspeciﬁc muscarinic antagonist ipratropium was able to block both responses. These
results suggest that pirenzepine is not acting directly on airway smooth muscle but on
some part of the cholinergic reﬂex pathway (Lammers et al. 1989). This is likely to be
the parasympathetic ganglia. Autoradiographic studies demonstrated a high density of
muscarinic receptors in human airway parasympathetic ganglia (van Koppen et al. 1988).
In in vitro studies with rabbit airways, pirenzepine has been shown to be more effective
in blocking contractions induced by pre- than post-ganglionic vagal nerve stimulation,
suggesting also that the M, receptors are most likely on parasympathetic ganglia (Bloom
et al. 1988).

Acetylcholine may exert a negative feedback effect on its own release by acting
on the autoreceptors on the airway cholinergic nerves. The muscarinic agonists
oxotremorine, carbachol, and pilocarpine concentration-dependently inhibit the

electrically evoked ACh release in guinea pig trachea (D’Agostino et al. 1990). The

 

methoctr.
al. 19922
cholinerg
potentiat:
SliIIllllallt
injection
to gallarn
gallamint
ACh, the
0f the air

Obtained

18

non-speciﬁc muscarinic antagonist atropine (Baker and Brown 1991; D’Agostino et al.
1990) and antagonists with relatively high afﬁnity for the M2 receptors, such as
methoctramine (Baker and Brown 1991), AF-DX116 (Kilbinger et al. 1991; Loenders et
al. 1992a) and AQ-RA741 (Kilbinger et a1. 1991), facilitate ACh release from airway
cholinergic nerves in guinea pig, dog, and rabbit trachea. In guinea pigs, gallamine
potentiates the airway constriction induced by either preganglionic or postganglionic
stimulation, but it does not inﬂuence the airway constriction produced by intravenous
injection of ACh. The selective M2 receptor agonist pilocarpine has the opposite effect
to gallamine. The inhibitory effect of pilocarpine is antagonized by gallamine. Because
gallamine and pilocarpine affect the responses to postganglionic stimulation but not to
ACh, the inhibitory muscarinic receptors must be located on the postganglionic neurons
of the airway parasympathetic nerves (Faulkner et al. 1986). Similar results have been
obtained with feline bronchial smooth muscle (Blaber et al. 1985).

Not all the evidence is in favor of the notion that muscarinic autoreceptors on
airway cholinergic nerves belong to the M2 subtype. Maclagan and Faulkner (1989) and
Kagaya et al. (1992) were unable to demonstrate the presence of prejunctional M2
receptors in rabbit and canine airways by comparing the effects of muscarinic receptor
agents on bronchoconstriction induced by vagal stimulation and intravenous ACh. This
is in contrast to the evidence obtained by direct measurements of ACh release (Loenders
et al. 1992a; Baker and Brown 1991). Some investigators have proposed that M,
(Janssen and Daniel 1990b) or M, (Deckers et al. 1989b) rather than M2 receptors are

inhibitory muscarinic receptors on airway cholinergic nerves.

Alt
obstructior
bronchoco
subjects, 1
cholinergi
feedback
(1989) re;
effect on
The abse
protective
infection
M2 511b,),
resPorrsit
tions,
dysfunc,

I
Some otl
in guine
the Ileur

relarecj

l9

Alterations in muscarinic receptors could occur in diseases with airway
obstruction or hyperresponsiveness. It has been reported that pilocarpine inhibits reﬂex
bronchoconstriction induced by sulfur dioxide (S0,) in normal but not in asthmatic
subjects, suggesting that feedback inhibitory muscarinic receptors may be present on
cholinergic nerves in normal airways and that there may be a dysfunction of this
feedback mechanism in asthmatic airways (Minette et al. 1989). Ayala and Ahmed
(1989) reported that prior muscarinic stimulation with methacholine has a protective
effect on histamine-induced bronchoconstriction in normal but not asthmatic subjects.
The absence of this inhibitory effect in asthmatic patients may represent loss of a
protective muscarinic receptor mechanism. Fryer et al. (1990) reported that pulmonary
infection with parainﬂuenza virus reduces the agonist afﬁnity for muscarinic (probably
M2 subtype) receptors in guinea pig lung and this decreased agonist affmity may be
responsible for the vagally induced bronchoconstriction seen in viral respiratory infec-
tions. Fryer and Wills-Karp (1991) observed that ovalbumin challenge causes
dysfunction of M2 muscarinic receptors in guinea pig pulmonary parasympathetic nerves.

The reason why muscarinic autoreceptors may be dysfunctional in asthma and
some other types of airway hyperresponsiveness is not known. There is evidence that,
in guinea pigs, the dysfunction of muscarinic receptors during viral infection is due to
the neuraminidase activity of the virus and that induced by ovalbumin challenge may be
related to the eosinophil major basic protein, which is an endogenous allosteric M2
receptor antagonist (Jacoby et al. 1992). It is possible that inﬂammation may also

damage other subtypes of muscarinic receptors in the lung such as the M3 receptors on

airvva

recep

Mod

ar

20

airway smooth muscle cells. However, because these receptors are in excess ("spare

receptors"), little change can be detected.

Modulation by sympathetic nerves and circulating catecholamines

Airway cholinergic neurotransmission can be inhibited by activation of certain
types of prejunctional adrenergic receptors. This effect has been demonstrated in human,
dog, and guinea pig airways.

Inhibitory adrenoceptors on human airway cholinergic nerves probably belong to
the a, and/ or 62 subtypes. In experiments with human bronchi, Grundstrom and
Andersson (1985) observed that exogenous norepinephrine (in the presence of propranolol
and cocaine to block B-receptors and neuronal uptake of norepinephrine, respectively)
inhibits EFS-evoked bronchial contractions that are atropine-sensitive. The inhibition was
antagonized by the az-adrenoceptor antagonist yohimbine, whereas the a,-adrenoceptor
antagonist prazosin was ineffective. Contractions evoked by exogenous ACh were
unaffected by the addition of norepinephrine. The results suggest that human bronchial
cholinergic neurotransmission can be inhibited by stimulation of prejunctional a2-
adrenoceptors. Rhoden et al. ( 1988) reported that isoproterenol, epinephrine, and
norepinephrine (in that order of potency) produced concentration-dependent inhibition of
human bronchial contractions induced by either EF S or ACh, and the inhibitory effect
was 10-100 fold stronger on EFS-induced contractions. The inhibitory effects of
isoproterenol on responses to EFS were prevented by propranolol and ICI 118551 (a 8,-
antagonist) but not by betaxolol (a B,-antagonist). These experiments suggest modulation

of cholinergic neurotransmission by prejunctional Bz-receptors.

In do
cholinergic 1
or salbutam
(TIPS) of c
tions suffrt
or on car
propranol
and ICI l
B, and if
bronchi,
“Wind:
EFS in
cells It
of pre
WEN

in Ca
diI‘ec
um

Dro‘

111‘

21

In dog airways, inhibitory B receptors seem to be important in modulation of
cholinergic neurotransmission. Janssen and Daniel ( 1990a) reported that isoproterenol
or salbutamol inhibits the contractions and amplitude of excitatory junction potentials
(EIPs) of canine bronchial smooth muscle induced by EFS but, when used at concentra-
tions sufﬁcient to eliminate EJPs, has little or no effect on resting membrane potential
or on carbachol-induced depolarization. These inhibitory effects are blocked by
propranolol or timolol, as well as by the selective antagonists ICI 89,406 (B,-selective)
and ICI 118,551 (B,-selective). They conclude that catecholamines act on prejunctional
B, and B, receptors, leading to inhibition of cholinergic neurotransmission in canine
bronchi. Ito (1988) reported that procaterol, a B,-adrenoceptor stimulant, dose-
dependently reduces the amplitudes of the twitch contractions and the EJPs evoked by
EFS in the dog trachea with no effect on the postjunctional response of smooth muscle
cells to exogenous ACh. Pretreatment with ICI-118,551 reduced the inhibitory action
of procaterol on the amplitude of the twitch contractions. This study indicates that
prejunctional B, receptors are important in modulation of cholinergic neurotransmission
in canine trachea. But Martin and Collier (1986), using a radioenzymatic method to
directly measure ACh release, reported that ACh release from canine isolated airway is
not modulated by norepinephrine. Failure to observe an effect of norepinephrine is
probably due to the fact that norepinephrine is mainly an a-receptor agonist, whereas the
inhibitory adrenoceptors on canine airway cholinergic nerves belong to the B type.

In a study with guinea pig tracheal rings, Grundstrom et al. (1981a) reported that
in the presence of B—blocking drugs, norepinephrine and epinephrine dose-dependently

inhibited contractions induced by cholinergic nerve activation. By contrast, contractions

induced I
receptor.
indicate 1
by acring
trachea t

tracheal

22
induced by exogenous ACh were unaffected. In order to characterize the norepinephrine

receptor, the effects of a, and 0:, blockers were evaluated using Schild plot. The results
indicate that, in guinea pig trachea, norepinephrine inhibits cholinergic neurotransmission
by acting on prejunctional a, receptors. The results of two other studies with guinea pig
trachea tissue also suggest that 0:, receptors are the inhibitory receptors on guinea pig
tracheal cholinergic nerves (Jacobsson et al. 1991; Kamikawa and Shirno 1990).

There are two possible sources of the physiologic agonists for these prejunctional
inhibitory adrenoceptors. One source is NE released by airway sympathetic nerves. In
contrast to the dense parasympathetic nerve supply to airways in all species, sympathetic
innervation is generally sparse, but there is considerable variation between species. In
humans, few, if any, adrenergic ﬁbers have been demonstrated in smooth muscle of
intrapulrnonary airways. Although there appears to be no direct functional sympathetic
innervation of human airway smooth muscle, it is possible that adrenergic nerves may
inﬂuence bronchomotor tone indirectly, e. g. , by modulating cholinergic neurotrans-
mission. Adrenergic nerves have been demonstrated within human airway ganglia.
Norepinephrine inhibits the ﬁring of airway ganglion cells in cat and ferret. In dogs,
electrical stimulation of thoracic sympathetic nerves causes bronchodilation, which is
abolished by B-antagonists and unaffected by adrenalectomy. The magnitude of
bronchodilation induced by sympathetic nerve stimulation is dependent on the degree of
preexisting vagal tone (Barnes 1986). These lines of evidence suggest that NE released
by the sympathetic nerves may modulate airway cholinergic neurotransmission.

The other possible source is circulating catecholamines. Norepinephrine in

plasma is derived almost entirely from overspill of sympathetic nerve activity, while

cpincphri
no signif
within th
hormone
physiolog
Therefor
modulati
if cholin
"l
abnorm;
Subjecrs
CORStn'c
3-fold r
CI al.
B-adrer,
Effeq (
1990).
mUScle

meChar

23
epinephrine is secreted by the adrenal medulla. Infusions of NE in normal subjects have

no signiﬁcant effects on cardiovascular or airway function at concentrations that are
within the physiologic range, suggesting that NE does not function as a circulating
hormone in humans. By contrast, epinephrine has potent metabolic effects within its
physiologic range and is a potent bronchodilator in normal and asthmatic subjects.
Therefore, epinephrine is probably the catecholanrine hormone that is important in
modulating airway cholinergic neurotransmission, and such an effect may be more likely
if cholinergic neurotransmission is increased (Barnes 1986).

There is increasing evidence suggesting that the secretion of epinephrine may be
abnormal in asthma. Exercise causes a rise in circulating catecholamines in normal
subjects, but in asthmatics who perform an exercise test sufﬁcient to induce broncho-
constriction, there is a blunted rise in NE and no rise in epinephrine (compared with a
3-fold rise in normal subjects who perform the same task) (Barnes et al. 1981; Warren
et al. 1982). It has been demonstrated that antigen challenge results in marked
B-adrenoceptor desensitization in guinea pig trachea, as indicated by the reduced relaxing
effect of epinephrine and decreased binding of 125I-cyanopindolol (Daffonchio et al.
1990). Although this observation is mainly a reﬂection of B receptors on airway smooth
muscle, it is possible that prejunctional B-receptors could be impaired through a similar

mechanism.

Modulti

ProStaj
import

DCUFO

(1989

from

addit

ex:

1'31

24
Modulation by prostanoids

Airway cholinergic neurotransmission may be modulated by prostanoids.
Prostaglandins of the E series (PGEs), particularly PGE,, appear to be the most
important prostanoids in this respect.

It is fairly well established that PGEs inhibit canine airway cholinergic
neurotransmission in spite of the existence of some conﬂicting data. Deckers et al.
(1989a) reported that the cyclooxygenase inhibitor indomethacin enhances ACh release
from canine bronchial tissue in response to EF S and this effect is reversed by the
addition of PGF,. Waters et al. (1984) observed that contractile responses of canine
tracheal smooth muscle to EFS diminished over a two-hour period of incubation in tissue
bath. Addition of indomethacin reversed this inhibition of contractile response.
Measured PGE, increased in the tisssue bath over two hours in control tissues. Addition
of PGE, to the tissue produced similar inhibition of contractile responses to EFS in a
concentration-dependent manner. In contrast, incubation alone, treatment with
indomethacin, or addition of PGE, had little, if any, effect on contraction induced by
exogenous ACh. Ito and Tajima (1981a) observed that the UPS induced by cholinergic
nerve activation in canine tracheal smooth muscle showed gradual and continuous
reduction in amplitude during prolonged exposure in Krebs solution (1-2 hours).
Indomethacin pretreatment prevented this reduction. Low concentrations (107-10*3 M)
of PGE, or PGE, markedly suppressed the amplitude of EJP in indomethacin-pretreated
tissue. However, Daniel et al. (1987) were unable to identify an inhibitory effect of

endogenous cyclooxygenase products on canine airway cholinergic neurotransmission.

 

Shore ct ;
release fr
Er

other spet

augments
greater er
prostanoit
bronchi, 1
UPS evoi
This obse
[11155011 in
such an e-

D.
Cholinergi
the eXClla‘
the release
Ai‘mimeti
but HOt I0
releaSe of
D: incl-3218‘
the resDon
HOWeVEr, .

25
Shore et al. (1987) observed only a trivial inhibitory effect of PGE, (10‘8 M) on ACh

release from cholinergic nerves innervating dog trachealis.

Evidence about the effect of PGE, on airway cholinergic neurotransmission in
other species is very limited. In the horse, meclofenamate, a cyclooxygenase inhibitor,
augments the EFS-induced contractions of trachealis strips and bronchial rings to a
greater extent than the exogenous ACh-induced contractions, suggesting that endogenous
prostanoids inhibit airway cholinergic neurotransmission (Yu et al. in press a). In human
bronchi, low concentrations of PGE, suppressed the amplitude of twitch contractions and
EJPs evoked by EF S but had no effect on the contractile response to exogenous ACh.
This observation suggests that PGE, has an inhibitory effect on cholinergic neurotrans-
misson in human bronchi. However, Black et al. (1989) were unable to demonstrate
such an effect of PGE, in human bronchial tissues.

There is only limited information about the effects of other prostanoids on airway
cholinergic neurotransmission. Exogenous carbocyclic thromboxane A, (cTXA,) inhibits
the excitatory neuro-effector transmission in the dog trachea, presumably by inhibiting
the release of ACh from the vagal nerve terminals (Inoue and Ito 1985). Thromboxane
A,-mimetic, U46619, enhanced the contractile response of canine bronchial rings to EFS
but not to methacholine (an ACh analogue) and may therefore increase the prejunctional
release of ACh (Chung et al. 1985). Tamaoki et al. (1987) observed that prostaglandin
D, increased the contractile response of canine bronchial rings to EFS but did not affect
the response to administered ACh, suggesting that PGD, potentiates ACh release.
However, Deckers et al. (1989a) did not detect a modulatory effect of PGD2 on ACh

output from canine bronchial tissue by direct measurement of ACh release. This

 

 

discrenancy H
An inhibitory
observed (Shc
Appro:
minutes to sev
(Robinson and
coughing and
adecrease in
asthmatics, hi
Uitinen et al.
(Orald et al. 1!
may SUbepit
M3187 and bi
“Th atOpic asn
melabolite 9a,
Cl'idence Sugge

TCSISIancC dUe l

Modulatim [7,, .

26

discrepancy may be due to the very low concentrations of PGD, applied in their study.
An inhibitory effect of PGI, on airway cholinergic neurotransmission has also been
observed (Shore et al. 1987; Inoue and Ito 1985).

Approximately 5-10 % of asthmatics experience profound bronchoconstriction 30
minutes to several hours after taking aspirin or other nonsteroidal antiinﬂammatory drugs
(Robinson and Holgate 1991). A few days after indomethacin injection, some dogs begin
coughing and wheezing (Ito and Tajima 1981b). Both of these effects might be due to
a decrease in certain inhibitory prostanoids following cyclooxygenase inhibition. In
asthmatics, histological lesions are common with airway epithelium (Dunnill 1960;
Laitinen et al. 1985a), which is thought to be an important source of PGE2 in the airways
(Ozaki ct al. 1987). In horses with heaves, a chronic obstructive pulmonary disease, the
airway subepithelial tissue produces signiﬁcantly less PGE, in response to stimulation by
A23187 and bradykinin than that of the control animals (Gray et al. 1992a). In humans
With atepic asthma, bronchial washings contained increased concentrations of PGD,, its
metabolite 9a, 116-PGF,, and TXB, (Robinson and Holgate 1991). These lines of

6Videllce suggest that alterations in prostanoid metabolism are likely to inﬂuence airway

feSiStance due to changes in ACh release.

MOdulation by histamine

Histamine is a biogenic amine that is found in large quantities in lung mast cells.
It is an important mediator of allergic bronchoconstriction and airway inﬂammation.
Plasma histamine level is elevated in athmatics. Histamine induces bronchoconstriction

through numerous mechanisms. When inhaled or intravenously injected, it induces

4 - I
.. .
i
I
J ins-gm;

bronchospasm at
mucus glycopror
Chung 1990).

Tm er
probably modulg
potentiates the tr
hi the radioimr
augmfinted by hi
Chlorphelliraminc
HTMCPIOYS, m;

In a110th:
MCHA) Caused ,
pig [lathe-a] tube
13' apolied ACh,
[Org did nor Pfe\

amagonist miOpe

 

 

"iterate, anti‘gOr
“Chinese CI al.
Cholinergic neur
inhibitory histan

In human airway

27

bronchospasm and airway nricrovascular leakage. In vitro, it increases the secretion of
mucus glycoproteins and has chemotactic activity for eosinophils (Chand et a1. 1990;
Chung 1990).

Three types of histamine receptors have been recognized: H,, H,, and H3. All
probably modulate cholinergic neurotransmission. In isolated canine trachea, histamine
potentiates the tracheal contraction induced by EF S. Direct measurement of ACh release
by the radioimmunoassay method showed that the ACh release induced by EF S is
augmented by histamine. The augmenting effect is blocked by the H, receptor antagonist
chlorpheniramine (T suchiya et al. 1990). These findings suggest that histamine, through
H,-receptors, may prejunctionally augment the release of ACh.

In another study, the histamine H3-receptor agonist (R)-a-methylhistamine (a-
MeHA) caused a dose-dependent inhibition of vagally mediated contraction of a guinea
pig tracheal tube preparation but did not alter tracheal contraction induced by exogenous-
ly applied ACh. Blockade of H,- and H,-histamine receptors, and 02- and B-adrenocep-
tors did not prevent the inhibitory effect of a-MeHA, whereas the speciﬁc H3-receptor
antagonist thioperamide prevented the effect of a-MeHA. In the presence of H,- and H,-
receptor antagonists, histamine also inhibited vagally mediated tracheal contraction
(Ichinose et al. 1989). These results suggest that prejunctional H3-receptors inhibit
cholinergic neurotransmission in the guinea pig airways. There is evidence that
inhibitory histamine H3-receptors are also present on postganglionic cholinergic nerves

in human airways (Ichinose and Barnes 1989).

ml

The effe
documented. B
release from the

Althougl
regulation of A(
of receptor ma;
between differer
precisely specul.

nerves under phj

MOdUIGIIOII b) a,

|
There are

are most Sensilir
("ATP recepm,»S
(Steinberg 1990;
AlmCePIOrs, Ori‘
adenosine. In u
by Imagining Wit
CAMP ICVels in
1990).

 

 

Numerou

0ther than

the air

 

28

The effect of H, receptors on cholinergic neurotransmission in the airways is not
documented. But there is evidence that H,-receptor activation results in depressed ACh
release from the superior cervical ganglion of the rat (Snow and Weinreich 1987).

Although all of the three types of histamine receptors are probably involved in the
regulation of ACh release from cholinergic nerves, the relative importance of each type
of receptor may vary between cholinergic nerves innervating different organs and
between different species. Because of the scantiness of evidence, it is too early to
precisely speculate on their role in controlling ACh release from airway cholinergic

nerves under physiological and pathological conditions.

Modulation by adenosine and adenosine nucleotides

There are two types of purinergic receptors. P, receptors ("adenosine receptors")
are most sensitive to adenosine and adenosine monophosphate (AMP). P, receptors
(" ATP receptors") are most sensitive to adenosine di- and triphosphate (ADP and ATP)
(Steinberg 1990; Burrrstock 1988). The P, receptors are further subclassiﬁed as A, and
A, receptors, originally based on the change in cAMP levels after receptor activation by
adenosine. In cells that possess A, receptors, adenosine decreases intracellular cAMP
by interacting with the Gi regulatory protein; conversely, adenosine increases intracellular
cAMP levels in cells bearing A, adenosine receptors, which stimulate Gs (Steinberg
1990).

Numerous lines of evidence indicate that adenosine modulates neurotransmission.
Most of the results were obtained from the central nervous system and peripheral tissues

other than the airways (Fredholm and Hedqvist 1980). There is only limited information

at

c}-

Cht

lint

29

about the modulatory effect of adenosine on airway cholinergic nerves. Aas et al.
reported that adenosine inhibits potassium-induced release of [3H] ACh from rat bronchial
smooth muscle. This effect is antagonized by the P, receptor antagonist 8-phenyl-
theophylline (Aas 1990; Aas and Fonnum 1986). The A, receptor agonist NECA
facilitates the ACh release and A, receptor agonist R-PIA does not affect it, indicating
that the inhibition of [’H]ACh release by adenosine in the airway smooth muscle has to
be due to stimulation of a prejunctional receptor different from the A, receptor (Aas
1990). It is unclear whether this inhibitory effect of adenosine on ACh release from
airway cholinergic nerves is also present in other animal species and human beings.

Adenosine and adenosine nucleotides are released from both nerve endings and
postjunctional structures as a consequence of nerve activity and the latter source seems
to be the more important. Like adenosine, adenosine nucleotides (ATP, ADP, and AMP)
also inhibit neurotransmission and they are probably as potent as adenosine (Fredholm
and Hedqvist 1980). Adenosine triphosphate is probably a transmitter coreleased with
NE from sympathetic nerves (Burnstock 1988). Evidence obtained in some non-airway
smooth muscles suggests that adenosine is probably an inhibitory neurotransmitter of the
NANC system (Small et al. 1990).

Very little is known about the mechanism by which adenosine inhibits transmitter
release. It seems clear that the presynaptic effects of adenosine are not mediated via
cyclic AMP (Fredholm and Hedqvist 1980).

The signiﬁcance of the inhibitory effect of adenosine on ACh release from airway
cholinergic nerves in airway hyperresponsiveness and disease is unknown. There are

hints that adenosine and purinoceptors may be of importance in the airways. Inhalation

 

c}

30

of aerosolized adenosine causes a concentration-related bronchoconsu'iction in asthmatic
patients but not in normal subjects (Cushley et al. 1983; Holgate et al. 1984; Marmo et
al. 1985). Theophylline, a competitive antagonist of cell surface purinoceptors in
concentrations within the therapeutic range of asthma therapy, inhibits bronchoconstric-
tion induced by adenosine (Cushley et al. 1984; Mann and Holgate 1985). Adenosine
also enhances antigen-induced bronchoconstriction and histamine release in rat isolated

lungs (Post et al. 1990).

Modulation by vasoactive intestinal peptide and nitric oxide

Both VIP and NO appear to be neurotransmitters or mediators of the inhibitory
N AN C nerves in the lung although their relative importance varies between species.
Vasoactive intestinal peptide and NO not only have a direct relaxant effect on airway
smooth muscle (Palmer et al. 1986; Munakata et al. 1990) but also may inhibit airway
cholinergic neurotransmission.

Vasoactive intestinal peptide inhibits cholinergic neurotransnrission in the airways
of ferrets, guinea pigs, and cats (Sekizawa et al. 1988; Martin et al. 1990; Xie et al.
1991). Sekizawa et al. (1988) observed that in isolated ferret tracheal segments, low
concentrations of VIP (up to 109 M) potentiate the response to EF S and higher
concentrations of VIP reduce the response. Contractions produced by ACh were not
affected by VIP at concentrations of 109 and 2 x 10'7 M. They concluded that VIP
modulates contractions produced by EFS via presynaptic cholinergic mechanisms.
Martin et al. (1990) reported that, in an isolated innervated guinea pig tracheal

preparation, exogenous VIP dose-dependently inhibits the contractile response to vagal

31
nerve stimulation (NS) and EFS. Vasoactive intestinal peptide (2.5 x 10‘, 2.5 x 10‘7)

inhibits contractions induced by NS signiﬁcantly more than EFS at 1 and 5 but not 20
Hz. In contrast, it does not signiﬁcantly reduce responses to ACh. They concluded that
in the isolated guinea pig tracheal preparation, VIP can modulate cholinergic neurotrans-
mission by actions that are predominantly on postganglionic nerves, but it has also a
small additional effect on ganglionic transmission. Xie et al. (1991) observed that VIP
antagonists [AC-Tyr‘, D-Phe2]-GRF(1-29)-NH, and [4-Cl-D-Phe‘, Leu”]-VIP dose-
dependently enhance the amplitude of EIPs evoked by a single stimulus in the eat but not
dog trachea. Neither VIP antagonist has any effect on the post-junctional response of
smooth musc1e cells to exogenously applied ACh in the dog or cat trachea. Their results
indicate that both VIP antagonists have a prejunctional action accelerating the excitatory
neuroeffector transmission, presumably by enhancing transmitter release from the vagus
nerves in the eat, but not in the dog trachea. An inhibitory effect of endogenous VIP on
airway cholinergic neurotranssmission has also been proposed based on the observation
that a-chymotrypsin, a proteolytic enzyme which degrades VIP, enhances the contractile
response of guinea pig tracheal smooth muscle to EFS without affecting the response to
exogenous ACh (Belvisi et al. 1993).

A modulatory effect of nitric oxide on airway cholinergic neurotransmission has
been demonstrated in guinea pigs. The nitric oxide synthase inhibitor L-NG-Nitro-
arginine methyl ester (L-NAME) enhances the contractile responses of guinea pig trachea
to EFS but not to exogenous ACh. The effects of L-NAME are reversed by L-arginine
but not D-arginine (Belvisi et al. 1991; Belvisi et al. 1993). These results suggest that

nitric oxide released by nerve stimulation inhibits cholinergic neurotransmission in guinea

32

pig trachea. This action of nitric oxide is in contrast to that observed in the rat basal
forebrain where nitric oxide enhances rather than inhibits the release of ACh from
neurons (Prast and Philippu 1992).

Inhibitory NAN C nerves have been demonstrated in vitro in the airways of several
species, including humans, horses, guinea pigs, pigs and cats. It appears to be the sole
inhibitory neural pathway in human airway smooth muscle (Barnes 1986). The
demonstration of the inhibitory effects of endogenous VIP and NO on airway cholinergic
neurotransmission suggests that VIP and NO may act as functional "brakes" on
cholinergic bronchoconstriction. In diseases with airway obstruction or hyperresponsive-
ness, the inhibitory NAN C innervation is frequently abnormal. For example, Broadstone
et al. (1991) reported that a functional NANC inhibitory innervation is present in the
bronchi of the control but not heavey horse. Ollerenshaw et al. (1989) reported that
immunoreactive VIP is present in lung tissue of normal humans but absent in that of
asthmatic patients. Miura et al. (1992) observed that the airway dilatory response to
activation of inhibitory N AN C nerves was signiﬁcantly inhibited in antigen-sensitized
cats. These observations suggest that an increased ACh release due to abnormalities in
the inhibitory NANC innervation is perhaps a mechanism in the development of airway

obstruction.

Modulation by other factors
Besides the above-mentioned factors, there are others that have also been reported
to be modulatory on airway cholinergic neurotransmission. Opioids (Belvisi et al. 1992),

5-hydroxytryptamine (Hahn et al. 1978), substance P (Tanaka and Grunstein 1990),

 

33

somatostatin (Sekizawa et al. 1989), and L-glutamate (Aas 1990) potentiate airway
cholinergic neurotransmission. Neuropeptide Y (Grundemar et al. 1988), L-aspartate,
D-glutamate, L-a-amino adipate (Aas 1990), and furosemide (a diuretic with bronchodi-
lating effect) (Y u et al. 1992b) inhibit airway cholinergic neurotransmission. The
modulatory effect of these factors is not discussed in detail in this review because the
evidence is very limited.

The majority of our information about modulation of airway cholinergic
neurotransmission has been gathered in the past ﬁve years. Most evidence is obtained
by comparing the contractile responses of airway smooth muscle to exogenous ACh and
to electrical activation of nerves, which releases endogenous ACh. As indicated in the
introduction of my dissertation, such an approach has its limitations. Direct measurement
of ACh release from airway cholinergic nerves is the most reliable way to monitor
changes in airway cholinergic activity. Although this had been considered technically
difﬁcult (Barnes 1991), a few groups of investigators have succeeded in measuring ACh
release over the last few years. The techniques used by these investigators to directly
measure ACh release include HPLC coupled with electrochemical detection (Deckers et
al. 1989a; Baker and Brown 1991), a radioenzymatic method (Martin and Collier 1986;
Shore et al. 1987), radioimmunoassay (Tsuchiya et al. 1990), and monitoring tritium
outﬂow from airway smooth muscle after preincubation with [3H]choline (D’Agostino et
a1. 1990; Kilbinger et al. 1991). Although each of these techniques has its advantages
and disadvantages, the HPLC plus electrochemical detection technique and the tritiated
technique seem to be gaining popularity in studies of ACh release for their overall

speciﬁcity, sensitivity, and simplicity. A judicious use of these techniques would greatly

ff. -
dam—:2.

expan‘

missio

Bea‘

 

34

expand our knowledge about prejunctional modulation of airway cholinergic neurotrans-

mission.

Heaves

In horses and ponies, there is a syndrome colloquially called "heaves" that is
characterized by recurrent airway obstruction. Clinical signs of heaves include dyspnea,
cough, exercise intolerance, and intermittent or chronic purulent nasal discharge. During
airway obstruction, animals have increased RL, increased APme during tidal breathing,
decreased Cm, and hypoxemia. Synonyms for heaves include chronic obstructive
pulmonary disease, recurrent airway obstruction, broken wind, chronic bronchitis,
chronic bronchiolitis, and equine pulmonary emphysema.

The occurrence of heaves is sporadic and the size of the population affected varies
widely in different regions. It is common in temperate parts of the world where horses
and ponies are stabled for a large part of their lifetime but rare in warmer parts where
the animals are kept on pasture year-round. Affected animals experience acute
exacerbations of heaves when stabled and fed poor-quality hay, and enter remission at

pasture. Normal equids do not develop heaves in the same environment.

Etiology

Heaves apparently has an allergic etiology. Although several factors, such as
viral infection (Gerber 1973) and pneumotoxin ingestion (Breeze et al. 1978), have been
proposed as the cause of heaves, exposure to environmental antigens in the barn (Breeze

1979) is most frequently associated with this syndrome.

35
Lowell (1964) was probably the ﬁrst person to report a relationship between the

acute exacerbations of heaves and the feeding of hay. He suggested that the syndrome
was precipitated by the inhalation of "heaves producing factors" in hay. A variety of
allergens can be found in hay (Halliwell et al. 1979; Woods et al. 1993). However, the
two most commonly implicated antigens are Micropolyspora faeni and Aspergillus
fumigatus, and the greatest number of heavey cases are associated with M. faeni.
Antibodies against these two antigens are frequently detected in the serum of affected
horses (Lawson et al. 1979). Many affected horses have a positive skin test in response
to these antigens (McPherson et al. 1979a). M. faeni is also associated with farmer’s
lung in humans (Fink 1976), a form of hypersensitivity pneumonitis. Factors that
increase environmental contamination with these antigens, such as inadequate ventilation
or feeding poor-quality hay, are associated with an increased incidence of heaves
(McPherson et al. 1979b).

However, an allergic response to environmental antigens is unlikely to be the sole
cause of the disease. Positive skin tests and serum antibodies against M. faeni and A.
ﬁemigatus are not restricted to animals with heaves, while animals without positive skin
test or the antibodies develop clinical signs characteristic of heaves (Lawson et al. 1979;

McPherson et al. 1979a).

Pathology
The lungs of heavey horses euthanized during an acute exacerbation of the
syndrome hardly collapse upon opening of the thoracic cavity and have a much larger

volume than lungs of normal horses. These changes are probably results of gas-trapping

36

due to airway obstruction. Excessive mucus is often present throughout the bronchial
tree.

Information about pulmonary histological changes in heavey horses is limited.
Published descriptions of the pulmonary lesions vary considerably because heaves is not
a speciﬁc disease identity but a syndrome that occurs in a variety of diseases. Major
morphologic changes, which are most apparent in airways less than 2 mm in diameter,
include chronic bronchiolitis marked by epithelial hyperplasia and metaplasia; plugs of
mucus and/or neutrophils in the lumina of small airways; peribronchiolar ﬁbrosis and
cellular inﬁltration by lymphocytes, plasma cells and mast cells; acinar overinﬂation;
eosinophil inﬁltration of bronchiolar walls and lumina; and alveolar emphysema (Breeze
1979; Thurlbeck and Lowell 1964). These lesions are not all found in every case, nor
are they always present to the same degree.

Examination of the bronchoalveolar lavage ﬂuid reveals that there is a neutrophil
inﬂux into the airways during acute exacerbation of heaves. Exposure of normal animals

to the barn environment does not change airway cell populations (Derksen et al. 1985c).

Pulmonary ﬁmction

Pulmonary resistance, APplmax during tidal breathing, and Cdyn are the three most
commonly used parameters for determining lung function in heaves. Heavey horses in
clinical remission have a RL similar to that of normal horses. When horses are brought
from pasture to a barn, airway obstruction occurs in heavey but not normal horses.

During airway obstruction, RL and APle increase, and CM decreases. These

37

parameters of lung fuction return to remission levels in 1-2 weeks when heavey horses
are brought back to pasture (Derksen et al. 1985a).

The increase in APplum is a result of both a decrease in intrapleural pressure
during inspiration and an increase in intrapleural pressure during expiration. The
increased APplm is necessary to generate airﬂow through the partially obstructed
airways. Gray et al. (1992a) reported that APplm (mean :I: SE) is 6.3 j: 0.4 and 7.6
:1; 0.3 cm H,O in normal and heavey horses kept at pasture, respectively. When horses
are stabled, APplm rises to 36.1 ;|: 4.9 cm H,O in heavey horses but remains constant
in normal horses (6.7 :1: 0.4 cm H,O).

Pulmonary resistance is the sum of airway resistance and lung-tissue resistance,
and is determined from the simultaneous measurements of the ﬂow and the pressure
difference between the airway opening and the pleural cavity. Because the cross-
sectional area of airways increases as the airways bifurcate, more than 80% of the airway
resistance is localized in the trachea and larger bronchial branches, whereas the resistance
of the bronchioles constitute less than 20% of the total resistance (Macklem and Mead
1967). Therefore, the R, value reﬂects mainly the resistance of larger airways. Results
of a previous study indicate that, during acute exacerbation of heaves, the R, is 2.42 :1:
0.26 cm H,O/L/s, which is more than twice the value during clinical remission (0.91 :l:
0.09 cm H,O/L/s). The R, does not change signiﬁcantly when normal horses are
transferred from pasture to a barn (0.63 d: 0.11 versus 0.87 :I: 0.10 cm H,O/L/s) (Gray
et al. 1992a).

Lung compliance is deﬁned as the lung volume change per unit of pressure

change across the lung. It can be determined either during breath holding (static

III: 5.5-1.

38

compliance) or during a normal breathing cycle (Cm). Static compliance deﬁnes the
static volume-pressure relationships of the lungs whereas Cd," is the ratio of tidal volume
to the APpr occurring between the beginning and end of a tidal volume. In normal
lungs, static and dynamic compliance should have the same value at normal respiratory
frequencies. In lungs with small airway obstruction, Cd,“ is less than static compliance,
particularly at rapid respiratory frequencies (Robinson 1979). Dynamic compliance
decreases in heavey horses during periods of airway obstruction (1.23 j; 0.18 versus
0.31 :t 0.06 L/cm H,O during remission and airway obstruction, respectively) (Gray et
al. 1992a).

During acute exacerbations of heaves, diffuse airway obstruction also leads to
ventilation and perfusion mismatching, which mainly impedes oxygen uptake in the lung
and has little inﬂuence on C0, removal due to the difference in the characteristics of the
O, and CO, dissociation curves. As a result, heavey horses are hypoxemic with only

mild hypercapnia or no hypercapnia (Gray et al. 1992a).

Mechanism of airway obstruction

The airway obstruction of heavey horses is due to a combination of excessive
mucus and exudates in the airway lumen, edema of the airway wall, and bronchospasm.
The proportion of airway obstruction contributed by bronchospasm varies widely not only
between different horses but also between different days in the same horse. Neverthe-
less, bronchospasm is generally responsible for a major proportion of the obstruction as
indicated by the marked and immediate reduction in R, caused by bronchodilators

(Broadstone et al. 1988; Broadstone et a1. 1992). The mechanism of bronchospasm in

39

heavey horses is unclear. Some factors that are probably related to the bronchospasm
are discussed below:

Airway hyperresponsiveness—Airway responsiveness is the ease with which a
stimulus induces airway narrowing. During airway obstruction, airways of heavey horses
and ponies are hyperresponsive to a variety of stimuli including aerosol and intravenous
histamine, aerosol methacholine, aerosol citric acid, and aerosol water (Armstrong et a1.
1986; Derksen et al. 1985a; Derksen et al. 1985b; Obel and Schmiterlow 1948). This
nonspeciﬁc hyperresponsiveness wanes when animals are brought back to the pasture.
The hyperresponsiveness makes it possible for a small amount of mediators or neuro-
transmitters to induce airway obstruction. In vitro, however, airway smooth muscle
preparations from heavey horses are hypo- rather than hyperresponsive to exogenous
ACh compared with those from control horses (Broadstone et al. 1991). Therefore, the
in vivo hyperresponsiveness of heavey horse airways probably results from changes other
than that of the smooth muscle cells. Some of the possible mechanisms for the in vivo
hyperresponsiveness follow. 1) The neurohumoral environment may facilitate airway
smooth muscle contraction in the lungs of heavey horses. 2) There may be an increased
thickness of airway wall due to edema or inﬁltration of inﬂammatory cells. When
airway smooth muscle contracts, the outside diameter of the airway decreases, but the
wall cross-sectional area does not change. As a result, the lumen of the airway narrows
to a much greater extent when airway walls thickened (Moreno et al. 1986). 3) The lung
parenchyma tethers the airways so that the airways do not close. This phenomenon,
called "interdependence, " pulls open the airways as the lung inﬂates (Ding et al. 1987).

Peribronchial edema and inﬁltration of inﬂammatory cells may unlink the

40

interdependence between the airway and the surrounding parenchyma (Macklem 1989)
and, therefore, increase the responsiveness of airway smooth muscle to broncho-
constrictors.

Pulmonary neutrophilia-In ponies with a history of heaves, barn exposure results
in increased neutrophils in bronchoalveolar lavage ﬂuid. This phenomenon does not
occur in normal ponies (Derksen et al. 1985c). Evidence obtained in other species
indicates that neutrophils produce mediators, such as platelet activating factor (Borgent
and Samuelsson 1979), prostaglandins (Doldstein et al. 1977), free radicals (Kanazawa
et al. 1991) and leukotrienes (Wang et al. 1986), that can induce bronchoconstriction.
Furthermore, the oxygen radicals produced by neutrophils may lead to airway
hyperresponsiveness (Lansing et al. 1993). Therefore, pulmonary neutrophilia could be
a contributing factor to the development of airway obstruction in heavey horses.
However, although aerosol challenge with M. faeni induces pulmonary neutrophilia in
both control ponies and ponies with a history of heaves, it affects pulmonary function
only in the latter (Derksen et al. 1987). The results suggest that it is not merely the
presence of neutrophils that alters lung function. What is probably more important is the
state of activation of the neutrophils and the mediators being released (Robinson 1989).
In asthmatics, pulmonary neutrophils are probably in an activated state (Kay 1988).

Dysfunction of inhibitory NANC nerves—Inhibitory N AN C innervation is the most
powerful inhibitory neural pathway in normal horse airways (Y u et al. in press b). There
is evidence that this inhibitory nervous system is dysfunctional in heavey horse airways:
the inhibitory effect of inhibitory NAN C nerves on airway smooth muscle tension can

be demonstrated in the central bronchi of normal but not of heavey horses (Broadstone

41
et al. 1991). A dysfunction of the inhibitory NANC nerves may disturb the balance

between excitatory and inhibitory innervations and favor the development of airway
obstruction. However, functional inhibitory NAN C nerves can be demonstrated only in
the trachea and central bronchi of normal horses (Y u et al. in press b). Therefore, it is
difﬁcult to explain the diffuse airway obstruction of heavey horses simply by an
dysfunction of inhibitory N AN C .

Nitric oxide is the neurotransmitter or a necessary mediator of inhibitory NAN C
nerves in horse airways. A dysfunction of inhibitory NAN C nerves may be a reﬂection
of a general deﬁciency of NO, which normally is widely distributed in the lung. Since
NO has an antiinﬂammatory effect (Kubes et al. 1991), a deﬁciency of NO may favor
the development of airway inﬂammation and obstruction.

Reduced airway mucosal PGE, production—P613, is the major prostanoid released
by tracheal epithelium of normal horses (Gray et al. 1992b). In heavey horses, airway
mucosal PGF, production is reduced (Gray et al. 1992a). PGE, has a direct inhibitory
effect on airway smooth muscle contractions (Wang et al. 1992). Furthermore, PGE,
inhibits ACh release from cholinergic nerves in canine airways (Deckers et al. 1989;
Shore et al. 1987). Therefore, a lack of PGE, production could be a mechanism of
airway obstruction in heavey horses.

Increased ACh release from airway cholinergic nerves—Blockade of muscarinic
receptors by intravenous administration of atropine has no effect on airway resistance in
the control horse or in the heavey horse during clinical remission. However, atropine
markedly reduces airway resistance in heavey horses during acute exacerbations of

heaves (Broadstone et al. 1988), indicating that a muscarinic mechanism is involved in

42

the airway obstruction. There are at least three possible mechanisms for the increased
_ response to muscarinic receptor blockade: 1) reﬂex activation of the cholinergic pathway
due to stimulation of sensory receptors in the airways; 2) increased sensitivity of airway
smooth muscle to ACh; and 3) increased ACh release due to altered prejunctional
regulation of airway cholinergic neurotransmission. Derksen et al. observed that, in
ovalbumin-sensitized ponies, ovalbumin challenge of one lung does not lead to airway
obstruction of the contralateral lung (Derksen et al. 1982). This observation suggests
that reﬂex mechanism is not important in the development of ovalbumin-induced equine
airway obstruction. Since heaves is also an allergic response, reﬂex mechanism is
probably also not important in the development of airway obstruction in heaves. During
acute exacerbation of heaves, the airways show a nonspeciﬁc hyperrresponsiveness,
suggesting that an increased sensitivity to ACh is at least in part responsible for the
response to atropine. In vitro, however, isolated airway smooth muscle preparations
from heavey horses are hyperresponsive to EFS but hyporesponsive to exogenous ACh
compared with those from control horses (Broadstone et al. 1991), suggesting that ACh
release from airway cholinergic nerves is increased. Therefore, increased ACh release
from cholinergic nerves is probably also responsible for the enhanced response to
atropine. The increased ACh release is most likely due to changes in prejunctional
regulation of airway cholinergic neurotransmission.

Muscarinic autoreceptors are potent inhibitory receptors on airway cholinergic
nerves. A dysfunction of these receptors may result in airway obstruction or hyper-
responsiveness due to increased ACh release. This possibility is strongly supported by

the ﬁndings that the muscarinic autoreceptors are dysfunctional in asthmatic patients

43
(Minette et al. 1989) and in guinea pigs infected with parainﬂuenza virus (Fryer et al.

1990) or challenged by ovalbumin (Fryer et al. 1991). It remains to be determined if
muscarinic autoreceptors are present on equine airway cholinergic nerves and if they are
dysfunctional in horses with heaves.

In canine airways, administration of exogenous PGF, inhibits ACh release,
whereas inhibition of endogenous prostanoid production by a cyclooxygenase inhibitor
augments ACh release (Deckers et al. 1989; Shore et al. 1987). It is unknown if PGE,
receptors are present on equine airway cholinergic nerve terminals and what role they
play. If inhibitory prejunctional PGE, receptors do exist and contribute to normal
regulation of airway cholinergic neurotransmission, the decreased airway mucosal PGE,
production in heavey horses (Gray et al. 1992a) may lead to increased ACh release.

The presence of inhibitory a,-adrenoceptors on cholinergic nerves has been
demonstrated in the airways of humans (Grundstrt'im et al. 1981a) and guinea pigs
(Grundstrom et al. 1981a; Thompson et al. 1990). It remains to be determined if they
play a physiological role in the regulation of airway cholinergic neurotransmission.
Possible physiological stimulants include NE released locally by sympathetic nerves and
circulating catecholamines. The observation that adrenergic and cholinergic nerves are
in close association in guinea pig trachea furnishes the anatomical possibility for the
interaction between adrenergic and cholinergic nerves (Jones et al. 1980). Catecholamine
levels are normally low in the plasma but increase sharply during exercise. It is unclear
if the physiological levels of circulating catecholamines are high enough to modulate
airway cholinergic neurotransnrission. In ponies with heaves, intravenous administration

of xylazine, an a, receptor agonist, causes bronchodilation and this effect is reduced by

44

the a,-receptor antagonist yohimbine (Broadstone et al. 1992). Although a,-receptors
also occur on the airway smooth muscle cells of some species (Barnes et al. 1982;
Takaganagi et al. 1990), their activation causes smooth muscle contraction. Therefore,
the bronchodilation induced by xylazine is probably due to activation of prejunctional a,-
receptors on airway cholinergic nerves. However, in vivo conditions are complicated.
The bronchodilation induced by xylazine may well be due to its tranquilizing effect in
the central nervous system. For these reasons, it is necessary to assess the importance
of prejunctional a,-adrenoceptors on equine airway cholinergic neurotransmission in
vitro.

My dissertation research investigated prejunctional modulation of equine airway
cholinergic neurotransmission. Most of the research was directed towards the
understanding of the physiological regulation of equine airway cholinergic neurotrans-
mission by the prejunctional muscarinic autoreceptors, 0:, adrenoceptors, and PGE,
receptors. Some experiments were also conducted with airway smooth muscle
preparations from heavey horses to determine if the development of airway obstruction
is related to a dysfunction of these prejunctional receptors. My experiments are
presented in the following chapters. Chapter 2 describes the technique of measuring ACh
release from airway cholinergic nerves, Chapter 3 concerns muscarinic autoreceptors on
horse airway cholinergic nerves, Chapters 4 and 5 describe the effect of PGE2 on airway
cholinergic neurotransmission, Chapter 6 investigates modulation of ACh release from
airway cholinergic nerves by inhibitory 0:, adrenoceptors, Chapter 7 concerns ACh
release from airway cholinergic nerves in horses with airway obstruction, and Chapter

8 summarizes the major results and conclusions.

CHAPTER 2
ACETYLCHOLINE RELEASE
FROM HORSE AIRWAY CHOLINERGIC N ERVES:

EFFECTS OF STIMULATION INTENSITY AND MUSCLE PRELOAD

Introduction

Knowledge about modulation Of airway cholinergic neurotransmission has usually
been inferred by comparing the effect of a factor on the contractile responses of airway
smooth muscle to EF S and to administration of ACh. Electrical ﬁeld stimulation
activates nerves within the preparation and releases neurotransmitters, including ACh,
whereas administration of ACh causes contraction by activation of muscarinic receptors
on the muscle. A greater effect on the contractile response to EFS than to ACh indicates
modulation of neurotransmission. Such indirect studies have suggested that airway
cholinergic neurotransmission may be modulated by a number of factors, such as
muscarinic autoreceptors (Fryer and Maclagan 1984; Minette and Barnes 1988),
prostanoids (Waters et al. 1984; Wang et al. 1992), histamine (Ichinose and Barnes 1989;
Ichinose et al. 1989), substance P (Tanaka and Grunstein 1990), VIP (Martin et al. 1990;
Sekizawa et al. 1988), catecholamines (GrundstrOm and Andersson 1985; Rhoden et al.

1988), and Opioids (Belvisi et al. 1992).

45

46

Interpretation of the results from indirect studies is often difﬁcult and could even
be misleading for the following two reasons: 1) During EFS, sympathetic, and NAN C
nerves may be activated as well as parasympathetic nerves. Although the function of
sympathetic nerves can be removed by drugs such as guanethidine, there are no speciﬁc
and powerful blockers for the various neurotransmitters of the NAN C nerves. These
latter neurotransmitters such as substance P, VIP, and NO may modulate the contractile
response to activation of cholinergic nerves at both pre- and postjunctional sites. 2) A
potential modulator of the airway cholinergic nerves frequently also has a postjunctional
effect. When both the frequency-response and ACh dose-response curves are affected,
it is difﬁcult to differentiate the prejunctional effect from the postjunctional effect.

To clearly demonstrate the neuromodulatory effect, it is essential to directly
measure ACh release from airway cholinergic nerves. This has been considered
technically difﬁcult (Barnes 1991), but recently a few laboratories have succeeded in
measuring EFS-induced ACh release (Aas 1990; Baker et al. 1992; D’Agostino et al.
1990; Deckers et al. 1989a; Martin and Collier 1986; Tsuchiya et al. 1990). The
prejunctional effect of neuromodulators is present or most apparent only within a certain
range of frequency (Duckles and Budai 1990; Knoll and Vizi 1971), voltage (Cherubirri
et al. 1985), and pulse duration (Kalsner 1990b). However, the majority of studies have
been conducted with a Single set of EFS parameters. There has been no systematic study
of the effect of Stimulation parameters on ACh release from airway cholinergic nerves.
Basic information about the effect of alterations in stimulation parameters such as

frequency, voltage, and pulse duration on ACh release not only indicates the sensitivity

Iﬂlivdedur .1,-

47

of the technique to detect changes in ACh output but also provides guidelines for
selecting EFS parameters in future Studies.

In some previous ACh release studies (Baker et al. 1992; D’Agostino et al. 1990;
Kilbinger et al. 1991), the airway smooth muscle strips were kept isometric for the
purpose of simultaneous tension measurements. Since a potential prejunctional modulator
often also has a direct postjunctional effect, the release of ACh from the control and
treated tissue might have been studied under a different muscle tone. Active contractions
or passive stretch of airway smooth muscle can release mediators such as PGE,
(Grodzinska et al. 1975; Orehek et al. 1973; Tiirker and Zengil 1976) that can
themselves prejunctionally modulate neurotransmission. Therefore, it seemed reasonable
to determine if alterations of muscle preload inﬂuence ACh release.

In the present study, I used high-performance liquid chromatography (HPLC) plus
electrochemical detection to measure ACh release from horse airway cholinergic nerves
in order to determine the effect of alterations in stimulation parameters and muscle
preload on ACh release. In this type of study, it is necessary to prevent ACh breakdown
by treating the tissues with a cholinesterase inhibitor. The resultant accumulation of
ACh may inhibit further ACh release due to the autoinhibitory effect of released ACh on
prejunctional muscarinic autoreceptors. This autoinhibitory effect may interfere with the
prejunctional regulatory effect of other modulators. For example, Deckers et al. (1989a)
reported that the potentiating effect of the cyclooxygenase inhibitor indomethacin on ACh
release from canine bronchial tissue can be demonstrated only in the presence of atropine
but not in its absence. Because future studies may often require the administration of

atropine in order to study the effect of other prejunctional modulators, it was necessary

48

to conduct the present study both in the presence and absence of atropine. In addition,
administration of atropine allowed us to determine if muscarinic autoreceptors are present
on cholinergic nerve terminals.

Horse airway smooth muscle was used for two reasons. First, the horse trachea
provides a large amount of smooth muscle for many tissue baths. This removes the
interanimal variability that occurs when it is necessary to use the trachea of one animal
per bath in order to obtain detectable ACh release. More importantly, our laboratory has
a long-term interest in studying heaves, a type of chronic obstructive pulmonary disease
that affects horses and bears similarities to human asthma. Results of previous studies
indicate that the airway obstruction is partly due to increased ACh release from the
airway cholinergic nerve terminals (Broadstone et al. 1991). However, the evidence is
indirect and it is unclear why ACh release may be elevated in heavey horses. It was
therefore necessary to establish a technique to directly measure ACh release from horse

airway cholinergic nerves.

Materials and methods

Experiments were approved by the All-University Committee on Animal Use and
Care of Michigan State University. Five horses, 438.9 1 10.9 kg, approximately 5.8
i 0.8 years old, were used in this study. Other investigators also utilized tissues for a
variety of investigations. Horses had no history of heaves, showed no clinical signs of
respiratory tract disorder and, post-mortem, their lungs and airways were normal in gross
appearance. Immediately after euthanasia by intravenous injection of an overdose of

pentobarbital sodium, the thoracic cavity was opened. The trachea was quickly removed

49
and immersed in Krebs-Henseleit solution (composition in mM: NaCl 118.4, N aHCO3

25.0, dextrose 11.7, KCl 4.7, CaCl, ° 2H,O 2.6, MgSO4 - 7H,O 1.19, KH,PO4 1.16)
that had been bubbled with 95 % O,/5 % C0,. Tracheal smooth muscle Strips with
epithelium intact were prepared from the tracheal segment between the 21st and 30th
cartilaginous rings above the carina. The muscle was cut with a template along the ﬁber
direction, and four strips (each measuring 2 x 30 mm) were cut each time. These four
strips were tied together at four places with silk and cut into two bundles of four muscle
strips, 15 mm long, each with the silk tied at both of their ends. Twelve such strip
bundles were prepared from the trachea of each animal. They were bubbled in Krebs-
Henseleit solution with 95 % O,/5 % CO, in a 30-ml tissue bath at 38°C for approximately
2 hours in order to remove the excessive foam formed during bubbling. The bundles
were then suspended in twelve 2-ml tissue baths (Radnoti Glass Technology, Inc. ,
Monrovia, CA). Each bath had a pair of parallel platinum wire electrodes built against
the wall in the vertical direction. One end of the bundle was ﬁxed to the bottom of the
bath with a glass tissue holder, and the suture on the other end was passed over a pulley
and attached to weights that maintained a passive tension. Tissues were equilibrated for
a further hour in these baths. During equilibration, the bathing solution was changed at
10—15 min intervals.

After equilibration, tissues were incubated with Krebs-Henseleit solution
containing 10'5 M neostigmine (a cholinesterase inhibitor to prevent the breakdown of
released ACh) with or without 10‘ M atropine for 30 min. Electrical ﬁeld stimulation
(20 V, 4 Hz, 2 ms, except otherwise speciﬁed) was then applied to each tissue for 30

min. Tissues were stimulated 3-5 times with a 30-min resting period between

50

consecutive stimuli. Tissue bath solution was collected after each EFS or at the end of
a 30—min resting period. Before the start and at the end of each EFS, the bath solution
was replaced with Krebs-Henseleit solution along with appropriate concentrations of
drugs. Each time the bath solution was changed, the tissues were rinsed three times with
Krebs-Henseleit solution containing 105 M neostigmine. All bundles, except for the two
bundles used for the muscle preload study, were studied under a preload of 8 g, since
previous studies had determined that the Optimal resting tension for each of the four
strips in the bundle is 2 g (Yu et al. 1992a). Electrical impulses were produced by a
stimulator (S88, Grass Instrument Co. , Quincy, MA) and passed through a stimulus
power booster (Stimu-Splitter II, Med-Lab Instruments, Loveland, CO). The electrical
impulses consisted of square waves. When the applied voltage was 20 volts, the current
going through each bath was 331.3 :1: 4.3 mA at 38°C as measured with the stimulus
power booster. The speciﬁc protocol for each strip bundle is shown in Table 2-1.

The wet weight of strip bundles determined at the end of the experiment after

blotting with ﬁlter paper averaged 235.3 1 5.4 mg (n = 60).

ACh analysis

High-performance liquid chromatography coupled with electrochemical detection
was used to measure the amount of ACh in the tissue bath liquid. The HPLC system
consisted of an Epson computer, a pump (Model 620, Perkin-Elmer, Norwalk, CT), an
autosampler (ISS 100C, Perkin-Elmer), an electrochemical detector (LC-4C, Bioanalytical
System, Inc. , West Lafayette, IN), an ACh and choline analytical column (Bioanalytical

System Inc.), and an enzyme reaction column with immobilized AChE and choline

51
oxidase (Bioanalytical System, Inc.). The software program used to control the system

was Perkin-Elmer’s LC analyst (version 2.01). The principle of the technique was that:
l) ACh and choline were separated in the analytical column; 2) ACh was hydrolyzed to
acetate and choline by AChE, and choline was oxidized to betaine and H,O, by choline
oxidase; and 3) H,O, was then oxidized at 0.5 V applied potential. The current
generated was proportional to the amount of H,O, and therefore to the amount of ACh
and choline. The lower limit of ACh determination was 1 pmol per injection. If 200
pl bath liquid was injected into the column each time, the lowest concentration of ACh
that could be correctly determined by this technique would be 5 x 109 M. The detector
response was linearly proportional to the amount of ACh over the range of 1 to 100
pmol. The mobile phase was 30 mM or 50 mM Na,I-IPO4 (pH = 8.5) and the ﬂow rate
was 0.5 or 1 ml/min. A typical chromatogram is shown in Fig. 2-1.

Collected samples were ﬁltered through 0.2 p nylon membrane Acrodiscs
(Gelman Sciences, Ann Arbor, MI) and stored in sample vials at 4°C if they were to be
analyzed within 12 hours or at -15°C if they were to be stored overnight. Two hundred
pl was injected into the column for both samples and calibration standards. An external
standard of 20 pmol ACh and choline was used for calibration. The standard was inject-
ed every 6 samples and the ACh amount in the samples was calculated based on bracket-
ed calibration. Approximately 6 samples were loaded onto the autosampler each time
and efforts were made to ensure that samples from the same bath remained in the
autosampler for about the same length of time, because ACh content in the samples
tended to decrease over time at room temperature. Analysis of samples began as soon

as they were collected and ﬁnished within 24 hours.

52‘

Drugs

Neostigmine methylsulfate, atropine, and hexamethonium (Sigma Chemical Co. ,
St. Louis, M0.) were diluted with water on the day of experiment. Tetrodotoxin
(Sigma) stock solution (1 mg tetrodotoxin in 5 m1 Krebs-Henseleit solution and 5 ml
acetic acid) stored at 4°C was diluted with water on the day of experiment. N eostigmine
and atropine were directly mixed into Krebs-Henseleit solution. Hexamethonium and
tetrodotoxin solutions were pipetted into the tissue bath in 20 pl volume (1% of the bath

volume). The concentrations of all drugs were expressed as their ﬁnal bath concentra-

tions.

Statistical analysis

Single factor repeated design AN OVA followed by the Tukey HSD test was used
to compare ACh release from the same tissues over different stimulation periods. An
unpaired t-test was used to compare the release between tissues with and without
atropine. All values are expressed as mean j: SE. P < 0.05 was considered

statistically signiﬁcant.

Results
Among all the samples (n = 210), the ACh content of only three samples was
below measurable range. Of the three samples, two were collected when EFS was not
applied and one was collected after a 5 volts-EFS period in the absence of atropine.
Sample Stability was checked at room temperature (23°C), refrigerator temperature

(4°C), and after freezing (-15°C). At room temperature, ACh concentration decreased

53
13.9 i 1.9% (means 5.6 pmol/200 pl versus 4.9 pmol/200 pl, p < 0.05, paired t-test,

n = 6) in eight hours. Refrigeration for 18 hours or freezing for 10 days did not result
in a decrease in ACh amount, the means being 3.8 pmol/200 pl versus 3.7 pmol/200 pl
for refrigeration (n = 5) and 6.9 pmol/200 pl versus 6.8 pmol/200 pl for freezing
(n = 8).

In the absence of EFS, the spontaneous ACh release was small. The dramatic
elevation of ACh release induced by EFS was abolished by tetrodotoxin (Fig. 2-2). The
release of ACh was frequency- and voltage-dependent. The frequency response curve
was sigmoid, with release reaching a plateau at 8 and 2 Hz in the absence and presence
of atropine respectively (Fig. 2-3). Acetylcholine release increased with increasing
voltage. In the absence of atropine, there was no difference in release at 15 and 20 volts,
whereas, in the presence of atropine, a plateau in release was achieved at 10 volts
(Fig. 2-4).

In the absence of atropine, increases in pulse duration caused a slight but
signiﬁcant increase in ACh release. In the presence of atropine, there was no clear
relationship between pulse duration and ACh release (Fig. 2-5).

Alterations in preload had no inﬂuence on ACh release (Fig. 2-6). When the
same tissue was stimulated ﬁve times, the amount of ACh released each time was fairly
constant (Fig. 2-7). Although the release of ACh was signiﬁcantly higher in the second
than in some of the other stimulation periods, the magnitude of the difference was small
in comparison to the total amount of released ACh.

Examination of the curves for frequency, voltage, pulse duration, preload-

response, and time-control (Figs. 2-3 to 2-7) shows that atropine signiﬁcantly and

54
dramatically augmented ACh release. In the time-control tissues stimulated at 4 Hz, 20

volts, 2 ms pulse duration, the release of ACh was approximately 10 and 45 pmol/g/min
in the absence and presence of atropine, respectively (Fig. 2-7).

Fifteen min of EFS released the same amount of ACh as 30 min EFS when the
ACh amount was expressed as pmol/g/min, suggesting that ACh release rate was constant
during the 30-min period of stimulation. Ganglion blockade with hexamethonium did not

inﬂuence the ACh release (Fig. 2-8).

Discussion

The present study demonstrated that EFS-induced ACh release from horse
trachealis strips was frequency- and voltage-dependent. The neural origin of ACh was
indicated by the observation that tetrodotoxin abolished the EFS-induced release.

When the length of stimulation period is ﬁxed, as in the present study, nerves
receive more pulses during each stimulation period as the frequency of stimulation is
increased. At lower frequencies, the total release of ACh increased with frequency,
while at higher frequencies, total ACh release reached a plateau. At each frequency, the
ACh release was much greater in the presence than in the absence of atropine. The
mechanisms of the plateau at higher frequencies may be different in the absence and
presence of atropine. In the absence of atropine, the plateau was mainly due to the
autoinhibitory effect of released ACh on prejunctional muscarinic autoreceptors. The
potential ability of tissue to further release ACh was clearly indicated by the high output
of ACh in tissues treated with atropine. For example, at a frequency of 8 Hz, ACh

release was 17.1 :i; 4.4 and 41.9 :I: 10.6 pmol/g/min in the absence and presence of

55

atropine, respectively. There could be several possible mechanisms for the plateau of
ACh release in atropine-treated tissues. First, the concentration of atropine applied may
not have been high enough to block all the prejunctional autoreceptors. Second, the
maximal ability of the tissue to synthesize or release ACh may have been reached.
Third, other neurotransmitters may have been released during EFS. Horse trachea has
parasympathetic, adrenergic, and NAN C nervous systems (Broadstone et al. 1991).
N orepinephrine and the neuropeptides released by the NAN C system can all potentially
modulate ACh release. Conclusions cannot be reached based on the results of the present
study, which was not designed to investigate these mechanisms.

The shape of the voltage-response curve obtained in the present study bears
Similarity to that obtained in tension studies where EFS causes a voltage-dependent
contractile response of airway smooth muscle (Broadstone et al. 1991). Maximal ACh-
release and muscle tension both occur at 15-20 V. Blockade of autoinhibitory receptors
increased the amount of ACh released at each voltage and the maximal release appeared
at 10 rather than 15 V.

The increase in ACh release with increase in pulse duration observed when the
tissues unheated with atropine were stimulated at 20 V is similar to the pulse-duration-
dependent increase in N E release from sympathetic nerves stimulated supramaximally
(Kalsner 1990b). Since the individual action potential is an all or none event, the increase
in per pulse release has been attributed to the increased recruitment of terminal ﬁlaments
of autonomic nerves.

Airway smooth muscle contractions induced by a variety of stimuli trigger the

synthesis of prostaglandins, particularly PGE, (Grodzinska et al. 1975; Orehek et al.

56
1973; Tiirker and Zengil 1976). Synthesis is thought to be related to the mechanical

consequence of the contraction (Orehek et al. 1973). Stretch of muscle can also elicit
the release of PGE-like material from isolated cat trachea (Ti'rrker and Zengil 1976). It
has been repeatedly demonstrated that PGEs, particularly PGE2, have an inhibitory effect
on airway cholinergic neurotransmission (Waters et al. 1984; Deckers et al. 1989a). Our
results indicate that muscle preload alterations over the range of 2—20 g had no inﬂuence
on ACh release. The results obtained in the presence of atropine reﬂect the effect solely
of changes in passive tension, because postjunctional muscarinic receptors were blocked
and the muscle did not contract during EFS. In the absence of atropine, the muscle
contracted and developed active tension equal to the preloads. The fact that alterations
of muscle preload over the 2—20 g range did not change ACh release indicates that either
1) the muscle does not release PGE2 or other prejunctional modulators in response to
different preloads, or 2) the cholinergic nerves are insensitive to these modulators.
Therefore, if ACh release is studied under isometric conditions, the difference in tension
between control and treated tissues is not likely to affect the outcome of results. Further-
more, if the release is studied under isotonic conditions as in the present study, the
release would not be affected over a wide range of muscle preloads.

Starke (1987) deﬁnes presynaptic autoreceptors as receptors located on or close
to the axon terminals of a neuron, through which the neuron’s own transmitter can and,
under appropriate conditions, does modify transmitter biosynthesis or release. The
presence of prejunctional muscarinic autoreceptors on airway cholinergic nerves has been
evidenced for several species (Barnes 1989c; Fryer and Maclagan 1984; Minette and

Barnes 1988). Most evidence suggests that these muscarinic autoreceptors belong to the

57

M2 subtype. In horse trachealis strips, we have examined the presence of M2 receptors
by comparing the effect of M2 receptor antagonist gallamine on the contractile response
to EFS and exogenous ACh. The contractile response to BF S is due to activation of the
cholinergic nerves. Gallamine has no effect on the ACh dose-response curve. It also
does not potentiate the response to EFS. The results suggest that there are no functional
Mz-receptors on the prejunctional site (Y 11 et al. 1992a). However, in the present study,
ACh release was augmented by atropine in all the treatment groups. The augmentation
of ACh release by atropine was not due to displacement of ACh molecules from
muscarinic receptors of the tissue into the incubation solution (Chapter 3). Therefore,
the results indicate the presence of muscarinic autoreceptors on the airway cholinergic
nerve terminals. Since atropine is a non-speciﬁc muscarinic antagonist, it is impossible
to judge the muscarinic autoreceptor subtype based on this ﬁnding. It is yet to be under-
stood why our indirect and direct experiments have produced different results. One
possibility is that, in the presence of cholinesterase inhibitor, the accumulated ACh
exerted an enhanced negative feedback effect on the autoreceptors and therefore the
function of these receptors is easier to demonstrate when a receptor antagonist is
administered. Another possibility is that during EFS, the NANC nerves were activated,
and their neurotransmitters postjunctionally modulated the contractile response to EFS.
The enhanced contractile response to EFS due to blocking of autoreceptors might have
been canceled by the direct inhibitory effect of the neuropeptides on the airway smooth
muscle. Such an effect could only be reflected in indirect studies. The third possibility
is that the prejunctional muscarinic autoreceptors on cholinergic nerves innervating horse

trachea do not belong to the M2 subtype. In rabbit airways, similar circumstances have

58

been encountered: the presence of prejunctional muscarinic autoreceptors on cholinergic
nerves can be detected by direct measurements of ACh release (Loenders et al. 1992) but
not by comparing the bronchoconstricting response to vagal stimulation and exogenous
ACh (Maclagan and Faulkner 1989). These differences highlight the importance of
direct measurement of ACh release.

The magnitude of augmentation of ACh release after atropine observed in the
present study is much higher than that reported by D’Agostino et al. (1990), who used
the tritiated choline method. The difference could be due to the 10 times higher
concentration of atropine used in the present study. Other possibilities include failure to
label all ACh storage pools by [3H]choline, and an enhanced negative feedback effect in
our study due to the use of a cholinesterase inhibitor, which resulted in accumulation of
ACh.

The ACh release was the same when expressed as pmol/ g/ min whether the tissue
was stimulated for 15 or 30 min, suggesting that the ACh release rate was constant and
stable over the 30-min stimulation period. When the same tissue was stimulated for 5
consecutive times of 30 min each, the release of ACh was fairly consistent over different
stimulation periods. Although the ACh release was higher in the second than in the other
stimulation periods, the magnitude of difference was small relative to the total amount
of released ACh.

The cholinergic nerves in the isolated trachealis preparation may comprise the
preganglionic autonomic nerve terminals and the postganglionic parasympathetic neurons.
In tension studies, it has been found that, during EFS, postganglionic neuron excitation

is not dependent on activation of the preganglionic nerves (Chapter 4). In the present

59
study, ACh may be released from both the pre- and post-ganglionic neurons. However,

it is impossible to determine the proportions from each site. In ferret, it has been found
that there are only a small number of ganglia in the dorsal membrane of the trachea
(Baker et al. 1986). If this holds true in horse trachea, our trachealis preparations may
contain few ganglia and the ACh release from the presynaptic nerve terminals may
contribute only a trivial proportion to the total release. The fact that the ganglionic
blocker hexamethonium did not inﬂuence ACh release conﬁrms that ACh release from
the postganglionic nerve terminals is independent of activation of preganglionic neurons.
Several techniques have been used to measure ACh release from airway
cholinergic nerves. Incubation with [3H]choline followed by monitoring of tritium
outﬂow from airway smooth muscle (Aas 1990; D’Agostino et al. 1990; Kilbinger et al.
1991) is very sensitive and does not require the addition of an AChE inhibitor to prevent
ACh break down. However, all the neural compartments from which ACh is released
are not equally labelled with tritium and the tritium stores are easily exhausted (Beani et
al. 1984; Luz et al. 1983). The radioenzymatic method is complicated, time-consuming,
and not as sensitive as HPLC plus electrochemical detection (Martin and Collier 1986).
Radioimmunoassay is very sensitive, but the antibodies against ACh have cross reactions
with some other choline esters (Kawashima et al. 1980; Tsuchiya et al. 1990). Use of
HPLC plus electrochemical detection to measure ACh was originally developed by Potter
et al. (1983) to measure ACh in rat brain tissues. It is highly speciﬁc and sensitive. Its
high speciﬁcity can be attributed to several factors: 1) different chemical compounds have
different retention times when they pass through the analytical column; 2) the enzymes

in the reaction column have speciﬁcity; and 3) only the chemicals that can be oxidized

60
at less than or equal to the applied voltage can be detected. This technique has recently

been used to measure ACh release from airway cholinergic nerves of guinea pigs, dogs,
cattle, and rabbits (Baker et al. 1992; Baker and Brown 1991; Deckers et al. 1989a;
Loenders et al. 1992a). Our data indicate that the HPLC technique provides a sensitive
method to detect the effects of stimulus intensity and autoinhibitory feedback on ACh

release from horse airway parasympathetic nerves.

61

.292 803 035 8:323:35: .03 33.5 533350. he
8233:0030 05. 6 59¢ .3 33:3 0335.308 3 030—8 3333 3: Eu Ana ﬂ vv 3033308”. 0032 3 mum 6:383 me 00:83 05 5 033 530 33033
030—8 no< 6:383 30 00:00.03 05 S 09300: 6:383 5223 v3 53 0:3: 05 9 33.3 33 33:5 880%: < 8 ”30:30.30 :03 5 33:00.3
803 x 33308 3: 3:. c3 0:383 33008 53 05 A339 2 .85 33: me 833 0E 05 5 Av ”3320 33 3:28 53 0:. 0:5 :30 3.33038
803 0:383 A339 083. :3 .03 0:333:02 Am ”:3. on n 83:30 mum .3:— N u :23020 0013 .3: v u 30:850.: .> cm ﬂ 0959, .3508» 83.3.3
300nm AN ”mum 5223 332.05 :3. on :03 3.00:8 33 £033 . 2 53m .3 0368 .8... 05 300:0 Gum—V 333.53 20c 30300.0 .33 3.00:8 803 83:30 =< C ”082

 

 

 

 

 

 

 

 

 

 

 

 

 

 

viii] 5-14.! iJIJl'lljlll!
_ _ A E: N: no _ maNezvsoN m
. 2; >8 :3 a; EN.N:_V.>0N v z
‘ _
A xm: 5? E... as Em x3. 53 Em m 8 8. N > 2 N: .. 3: v 8. N a: v .> 8 m H
m ea. 8 3m 33 m S we _ > 2 N: N N: N as N .3: e .> 0N N H
_
h E... n: Em 3m oz 0 N we no > n N: _ N: _ E N .N: c .> 0N _ _
- _
A 3 3 z _ > __ z 3 z > =1 z > z _ > 3 29:3 _
a :3. on 08 0:383 95 3223 3 E .23 0562300: 53 539:0:— g
A 83233.3 _
_ - I,
N. _ : _ 2 3 ﬂ 3 N v n
u M a
a _ x: _ “
xm: a. :85: 253 _ 222.. .535: 8.3 3:383
. d 0:00.533 _ ‘

 

 

 

20883 8.30: .:-N 0.3:.

62

chohne

.‘u

 

 

 

 

 

 

 

 

 

 

 

 

 

.C
l U l
. .<
a A l \
\ " \ ,
.. 4 5. ~I“~““'~--....«----~---*'// \/ “xx N.
0 5 10 15
minutes

Fig. 2-1. A typical chromatogram showing the separation of acetylcholine (ACh) and
choline in a sample. The mobile phase was 50 mM NaZHPO4 and the ﬂow rate was 0.5
mllmin. Two hundred pl bath liquid was injected into the HPLC system. The ACh and

choline peaks represent 8.6 and 92.4 pmol, respectively.

63

70—

so— *

 

40-

 

pmol/g/mln

20*-

10— —1——

 

 

 

 

 

 

 

no EFS EFS TTX

Fig. 2-2. Effect of electrical ﬁeld stimulation (EFS, 20 V, 4 Hz, 2 ms) on acetylcholine
release from horse airway cholinergic nerves in the presence of 10‘ M atropine and the
effect of 10*5 M tetrodotoxin (TTX) on the EFS-induced release. * signiﬁcantly different
from the other two treatment groups. There was no signiﬁcant difference between the
"no EFS" and "TTX" groups.

. u 11

7O —
‘F wlth atropine

6° — 0 without atroplno

50_

40"

pmol/g/mln

 

 

Frequency (Hz)

Fig. 2-3 . Effect of frequency on electrical ﬁeld stimulation-induced acetylcholine release
from horse airway cholinergic nerves in the presence and absence of 10‘5 M atropine.
Signiﬁcantly different from 2, 4, 8 Hz (a), 4, 8, 16 Hz (b), or 8, 16 Hz (c). Values in
the presence and absence of atropine were signiﬁcantly different at all frequencies.

65

7O —
+ with atropine

60 - . without atropine

40—

 

30—

pmol/g/mln

20—

10—

 

 

 

l l l

5 1o 1 5 20
Voltage (volts)

Fig. 2-4. Effect of voltage on electrical ﬁeld stimulation-induced acetylcholine release
from horse airway cholinergic nerves in the presence and absence of 10‘ M atropine.
Signiﬁcantly different from 10, 15, 20 V (a), 15, 20 V (b), or 20 V (c). Values in the
presence and absence of atropine were signiﬁcantly different at all voltages.

66

 

 

 

7° — *with atropine
O

6 O _ without atropine a

50 —
.S
E 40 "
\
U!
E
0 3O -
E
Q

20 -

1 o — b C

o l l l l l
O 5 1 2 3

Pulse duration (ms)

Fig. 2-5. Effect of pulse duration on electrical ﬁeld stimulation-induced acetylcholine
I"ﬁlease from horse airway cholinergic nerves in the presence and absence of 10‘5 M
atropine. Signiﬁcantly different from 0.5 ms (a), 2, 3 ms (b), or 3 ms (c). Values in

«3 presence and absence of atropine were signiﬁcantly different at all pulse durations.

67

 

 

 

 

 

 

70 —
"- with atropine
60 __ .without atropine
50 r-
.E
E _
E, 40
'2
a 30 —
20 L
10 — e— + #f
l l l J
O 2 10 20

Muscle tension (9)

Fig. 2-6. Effect of preload on electrical ﬁeld stimulation-induced acetylcholine (ACh)
release from horse airway cholinergic nerves in the presence and absence of 1045 M
atropine. Preload alterations had no inﬂuence on ACh release. Values in the presence
and absence of atropine were signiﬁcantly different at all preloads.

68

 

 

 

70 P "' with atropine
O
60 __ without atropine
a
50 -
.E
E
E» 40 —
".8.
O- 30 -
20 —
b
o l l L l l l
1 2 3 4 5

Sample number

Fig. 2-7. Electrical ﬁeld stimulation-induced acetylcholine release from horse airway
cholinergic nerves over 5 stimulation periods in the presence and absence of 1045 M
atropine. Signiﬁcantly different from 81, S4, SS (a), or 81 (b). Values in the presence
and absence of atropine were signiﬁcantly different at all stimulation periods.

69

70'-

60—

 

 

 

40*—

 

 

 

pmol/g/mln

20F

10—

 

 

 

 

 

 

 

 

1 5 min 30 min hexamethonium

Fig. 2-8. Effects of stimulation duration alterations and hexamethonium on acetylcholine
release from horse airway cholinergic nerves in the presence of 10*5 M atropine. If
expressed as pmol/ g/ min, ACh release was the same whether the tissue was stimulated
for 15 or 30 min. Hexamethonium had no inﬂuence on ACh release.

CHAPTER 3
MUSCARINIC AUTORECEP’I‘ORS

0N HORSE AIRWAY CHOLINERGIC N ERVES

Introduction

The release of ACh from airway cholinergic nerves, the major bronchoconstrictor
neural pathway in all mammals, is regulated by prejunctional muscarinic autoreceptors.
Muscarinic autoreceptors are probably dysfunctional in asthmatic patients (Minette et al.
1989) and in guinea pigs infected with parainﬂuenza virus (Fryer et al. 1990) or
challenged by ovalbumin (Fryer and Wills-Karp 1991). This dysfunction may lead to
increased ACh release and airway obstruction.

Although ﬁve subtypes of muscarinic receptors have been cloned, only three, M1,
M2, and M3, can usually be identiﬁed by the use of selective muscarinic antagonists.
Most investigators conclude that the muscarinic autoreceptors on airway cholinergic
nerves belong to the M2 or a "Mz-like" subtype (Aas and Maclagan 1990; Brichant et al.
1990; Kilbinger et al. 1991; Minette and Barnes 1988). However, some have suggested
that they belong to the M1 (Janssen and Daniel 1990b), M3 (Deckers et al. 1989b), or
M2/M4 (Loenders et al. 1992b) subtype.

Information about prejunctional muscarinic autoreceptors on horse airway

cholinergic nerves is limited. Direct measurements of ACh release from airway

7O

14.1. ﬁml. I.“ 1 l... i .3

71

cholinergic nerves demonstrate that the nonselective muscarinic receptor antagonist
atropine dramatically augments ACh release (Chapter 2), indicating that inhibitory
muscarinic receptors are present. However, the selective M2 receptor antagonist
gallamine does not augment the response of horse trachealis to EFS, suggesting that there
are no functional inhibitory M2 receptors (Y u et al. 1992a). Therefore, the subtype of
autoreceptors remains to be determined.

Measurement of ACh release from airway cholinergic nerves is the most direct
and reliable method to study prejunctional modulation. The present study was conducted
to characterize the prejunctional muscarinic autoreceptors by studying the effects of

several competitive muscarinic antagonists on ACh release.

Materials and methods
Tissue preparation and equilibration

Trachealis tissues of 10 horses (body weight 405.8 1: 35.6 kg) were used for this
study. Horses had no clinical signs of respiratory disease for several weeks before they
were euthanized by injection of an overdose of pentobarbital sodium. Trachealis stn'p
bundles with epithelium were prepared from the 16th to 25th cartilaginous rings above
the carina using the method previously described (Chapter 2). Each bundle was
suspended in a 2-ml tissue bath that contained Krebs-Henseleit solution (composition in
mM: 118.4 NaCl, 25.0 NaHCO3, 11.7 dextrose, 4.7 KCl, 2.6 CaC12.2H20, 1.19
MgSO4.7H20, and 1.16 KH2P04) maintained at 38°C and bubbled with 95% 02 / 5%
C02. The tension applied to each bundle was well within the range of 2-20 grams,

which had been determined to have no influence on ACh release from horse airway

72

cholinergic nerves (Chapter 2). Tissues were equilibrated in the baths for 2 to 3 hours,
during which time the bath solution was replaced at 10-20 min intervals.

Each 2-ml tissue bath has a pair of built-in platinum wire electrodes. Electrical
impulses were produced by a stimulator and passed through a stimulus power booster as

described in Chapter 2.

Protocol

After equilibration, the tissues were incubated with 105 M neostigmine, a
cholinesterase inhibitor, and 10'5 M guanethidine for 60 min. Four trachealis strip
bundles from each animal were used to study the effects of atropine (non-selective,
1043-104 M), pirenzepine (Ml-selective, 104-103 M), AF-DX 116 (Mz-selective,
107-10'3 M), and hexahydrosiladifenidol (HHSiD, M3-selective, 104—104 M) on ACh
release, respectively. Electrical ﬁeld stimulation (20 V, 4 Hz, 2 ms) was applied to the
tissues for six 20-min periods with a 30-min interval between consecutive periods.
During period 1, the tissues were stimulated in the absence of muscarinic antagonists.
During subsequent periods, each bundle was stimulated in the presence of logarithmically
increasing concentrations of a muscarinic antagonist. The tissue was preincubated with
each concentration of an antagonist for 25-30 min. Bath liquid was collected for ACh
analysis 1—2 min after each EFS period. The tissue bath solution was replaced before
the start and at the end of each EFS period.

One bundle from each animal was stimulated for 6 periods in the absence of any
muscarinic antagonist and served as time control. Two bundles were used to examine

the effect of the vehicles of AF-DX 116 and HHSiD. To determine if ACh release could

73

be sustained at a high level following the administration of muscarinic antagonists, one
bundle was stimulated for 6 periods in the presence of 105 M atropine.

Two bundles were used to determine if the increase in ACh concentration in the
tissue bath following muscarinic antagonists was partly due to displacement of ACh
molecules from muscarinic receptors in the tissues into the bath solution. These two
bundles were stimulated for two periods. Bath liquid samples were collected 15 min
after each EFS period. During period 1, no muscarinic antagonist was administered.
During period 2, one bundle was stimulated in the presence of 105 M atropine after
incubation with it for 20—25 min, whereas the other bundle was stimulated in the absence
of atropine but incubated with 105 M atropine for 15 min after EFS prior to collecting
the sample.

At the end of experiment, tissues were blotted and weighed. The wet weight
averaged 0.197 :I: 0.006 g.

In order to compare the effect of HHSiD on tracheal smooth muscle contraction
with that on ACh release, two trachealis strips (10 x 2 mm) with epithelium were
suspended in tissue baths and equilibrated using the method described previously (Y u et
al. 1992a). The isometric force generated by the muscle strip was recorded on a
polygraph. After equilibration, 105 M ACh was added to the baths to precontract the
strips. When the effect of ACh had reached a plateau, HHSiD was added to one bath
at concentrations beginning from 10'8 M in log increments until the tension had returned

to baseline. The other strip was treated with HHSiD solvent and served as time control.

74
ACh analysis
ACh content was determined by HPLC coupled with electrochemical detection

(Chapter 2).

Drugs

Neostigmine methylsulfate, guanethidine monosulfate, atropine sulfate, and
pirenzepine dihydrochloride (Sigma Chemical, St. Louis, M0) were dissolved and diluted
with either water or Krebs-Henseleit solution on the day of experiment. Frozen stock
solutions of HHSiD hydrochloride (Research Biochemicals Inc. , Natick, MA) and AF -
DX 116 (courtesy Boehringer Ingelheim, Ridgeﬁeld, CT) were diluted with water before

use. The concentrations of all drugs are expressed as their ﬁnal bath concentrations.

Statistical analysis

ACh release was expressed as pmol/g/min. One way analysis of variance
(AN OVA) for repeated measures followed by Newman-Keuls test was used to compare
the difference in ACh release rate before and after the treatment with the muscarinic
antagonist, and to compare the maximal effects and the negative logarithm of EC,0 (the
molar concentration of an antagonist required to reach 50% of its maximal augmenting
effect) of different muscarinic antagonists. All values are expressed as mean i SE.

P < 0.05 was considered statistically signiﬁcant.

75
Results

All samples contained a measurable amount of ACh. All the muscarinic
antagonists augmented ACh release concentration-dependently (Fig. 3-1). Atropine was
the most potent antagonist in augmenting ACh release. A signiﬁcant increase appeared
at 10‘8 M with atropine, 1045 M with HHSiD, and 105 M with pirenzepine and AF-DX
116 . However, the magnitude of augmentation by 105 M pirenzepine and AF-DX 116
was larger than that by 10”8 M atropine. Therefore, 3 x 10" M might be the approximate
concentration required for pirenzepine and AF -DX 116 to produce an effect comparable
to that of 10“3 M atropine. The release rate decreased at the highest applied concentra-
tions of all muscarinic antagonists except for pirenzepine. The maximal effect
(pmol/g/min, n = 5) of atropine (78.3 :I; 8.8) was signiﬁcantly greater than that of
pirenzepine (50.3 i 7.1) and AF—DX 116 (50.4 i 7.0). The -log EC,0 (n = 5) of
atropine (7.4 j; 0.3) was signiﬁcantly greater than that of pirenzepine (5.5 :l: 0.2) and
AF-DX 116 (5.7 :l: 0.2). The maximal effect and ECso of HHSiD could not be
determined because ACh release did not reach a plateau following HHSiD administration.

In the absence of muscarinic antagonists, ACh release was constant over the six
stimulation periods (11 = 5). The vehicles of AF-DX 116 and HHSiD had no inﬂuence
on ACh release (I) = 4). The augmenting effect of 105 M atropine was consistent over
the six stimulation periods (n = 4) (Fig. 3-2).

Incubation of tissues with atropine during EFS augmented ACh release (11 = 3).
However, incubation of tissues with atropine after EFS but before sample collection had

no effect on the measured amount of ACh (n = 4) (Fig. 3-3).

76

The tension increase following 10" M ACh was stable over the experimental
period. The selective M3 muscarinic receptor antagonist HHSiD concentration-
dependently reduced the tension of the ACh-precontracted tissues. HHSiD reduced the
tension to about 20% of the precontracted level at 1045 M and relaxed the muscle

completely at 105 M (Fig. 3-4).

Discussion

In the present study, all four muscarinic antagonists augmented ACh release.
Augmentation was not simply a result of displacement of ACh molecules from the
receptors of the tissue into the bath liquid because incubation with atropine after EFS but
prior to sample collection had no inﬂuence on the measured amount of ACh. During the
6th EFS period that coincided with the highest concentration of each antagonist, the
augmenting effect of some antagonists decreased. This was a result of nonspeciﬁc
antagonist effects rather than of depletion of ACh, because the high ACh release rate in

response to 105 M atropine was maintained over six EFS periods.
The increase in ACh release in response to atropine conﬁrms the presence but
does not identify the subtype of prejunctional muscarinic autoreceptors on horse airway
<=3l1()linergic nerves. Although all three selective muscarinic antagonists, including
pirenzepine (M1 selective), AF-DX 116 (M2 selective), and HHSiD (M3 selective),
augmented ACh release, none had the same potency or maximal effect as atropine. Even
alough the muscarinic autoreceptors may comprise a single subtype, the augmentation by
a l 1 antagonists was not unexpected because they are receptor-selective rather than

r"eceptor-speciﬁc.

77

M3 receptors are responsible for horse airway smooth muscle contraction in
response to ACh (Yu et al. 1992a). Therefore, to determine if the prejunctional
muscarinic autoreceptors are also M3, 1 compared the effects of HHSiD on airway
smooth muscle contraction and ACh release. If the prejunctional autoreceptors are also
M3, HHSiD should have a similar potency on both muscle contraction and ACh release.
The observation that HHSiD (1 pM) inhibited approximately 80% of the 10 uM ACh-
induced contraction but augmented ACh release only slightly indicates that the
prejunctional muscarinic autoreceptors are not M3. Using a similar approach, Loenders
et al. (1992b) reached the same conclusion about rabbit tracheal parasympathetic nerves.
By contrast, Deckers et al. (1989b) concluded that the prejunctional muscarinic
autoreceptors are M3 because the M3-selective muscarinic antagonist Miphenylacetoxy-
N-methylpiperidine (4-DAMP) augmented ACh release from canine bronchial segments.
I have made a similar observation in horse trachealis strips (unpublished data). Because
4-DAMP is a selective rather than speciﬁc M3 antagonist, its augmenting effect may be
eXplained by actions on other receptor subtypes.

M, receptors are generally regarded as excitatory rather than inhibitory receptors

on airway cholinergic neural pathways (Barnes 1989b,c). Pirenzepine has a higher
afﬁnity for M, receptors than other subtypes. Its Kd for M, receptors is approximately
1 2.6 nM based on the pA2 value (Mei et al. 1989). In the present study, ~3 uM
D irenzepine was required to augment ACh release signiﬁcantly. This concentration is
about 240 times its K4, suggesting that the prejunctional autoreceptors are unlikely to be

of the M, subtype.

78

The prejunctional muscarinic autoreceptors appear to be of the M, subtype in
several species. However, in the present study, the concentration-response curve
obtained with the Mz-selective antagonist AF-DX 116 not only looked dramatically
different from that of atropine but also superimposed on the concentration-response
Curves of the M, and M3 antagonists. Although the augmenting effect of AF-DX 116
Was not signiﬁcant until 10 ”M, 3 pM might be able to produce an effect comparable to
that of 10 nM atropine. Even the latter concentration of AF -DX 116 is about 21 times
of its Kd (141 nM calculated from pA2 in Mei et al. 1989) for M2 receptors, suggesting
that prejunctional muscarinic autoreceptors are not M2. However, this conclusion is
tempered by the observation that the concentration of atropine (10 nM) required to
augment ACh release was also about 8 fold greater than its Kd for M2 receptors (1.26 nM
calculated from the pA2 in Mei et al. 1989). The present study therefore does not allow
elimination of the possibility that the prejunctional autoreceptor is M2. However, in a
previous study the M2 antagonist gallamine did not augment the response of equine
tl'achealis to EFS, suggesting an absence of functional M2 receptors (Yu et al. 1992a).
It may be necessary to investigate if the autoreceptors on horse airway cholinergic nerves

belong to other muscarinic receptor subtypes such as the M4 or a novel uncharacterized
Subtype. M4 receptors share the same transduction mechanism as M2 receptors. Due to
axe lack of muscarinic antagonists with satisfactory selectivity for M4 receptors,
eJeamination of the possibility that the autoreceptor is M, must await the development of

new drugs.

79

 

 

 

100 F
+ atropine + pirenzepine “A” AF-DX 116 0 HHSiD
\
g '
‘e’
9.- 60 ‘
.9. _
L“.
g 40 r-
3
e _
5
< 20 _
l l l l l l l l
before drug 8 7 6 5 4 3

Antagonist concentration (-log M)

Fig. 3-1. Effects of muscarinic antagonists on acetylcholine (ACh) release from horse
ail'vvay cholinergic nerves. All antagonists augmented ACh release. Signiﬁcant effects
a1,.3133ared at 10 nM atropine, 1 pM hexahydrosiladifenidol (HHSiD), and 10 uM
plrenzepine and AF-DX 116. The potency and maximal effect of atropine were much
g1‘ﬁater than those of the other antagonists.

 

so.

1 00 _ + TC + Atropine (10‘5 M)

* AF—DX1 16 vehi. 0 HHSiD vehi.
eo — .. l l

40—-

 

 

 

 

 

 

 

ACh release (pmol/g/min)

 

 

Stimulation periods

I:ig. 3-2. ACh release in response to EFS in the time control tissue and in tissues treated
with AF-DX 116 vehicle, HHSiD vehicle, and 10’5 M atropine. Atropine augmented
ACh release. AF—DX 116 and HHSiD vehicles had no inﬂuence on ACh release. ACh
t"ﬁlease was constant over the six stimulation periods.

81

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

100 "—
" Dperiod 1 .period 2
,5 80— '7—
'g _
on
>- a
0 GOP
.2, \§
3'} 40— \
9.’
< 20— \
_l—_ I —l'_
h _L_
L. \ \
Group A (n=3) Group B (n=4)

F i g. 3-3. ACh release in response to 105 M atropine administered before (Group A) and
after EFS (Group B). Two periods of EFS were applied to each group. Tissue bath
Samples were collected 15 min after each period of EFS. During period 1, no muscarinic
antagonist was administered. For period 2, atropine was administered before EFS in
g roup A and after EFS but prior to sample collection in group B. Atropine did not
augment ACh release in Group B, suggesting that the augmentation was not simply a
ITS—Sun of displacement of ACh molecules from the receptors of the tissue into the bath
1 lquid.

82

 

 

   

 

 

10
C
-2 8°
8
g
g 60—
Q
E
:2.
‘—
O 40.
O
o\
q, +control
Q .
8 20— OHHsto
La. ”:5
l
O 8 7 6 5 3

HHSiD concentration (-log M)

Fig. 34. Effect of the M3-selective antagonist hexahydrosiladifenidol (rmsm) on the
ttension of horse airway smooth muscle precontracted by 10 ptM ACh. HHSiD

<To()Ir<:entration-dependently relaxed the muscle, 1 pM and 10 pM eliminated about 80%
and 100% of the precontraction, respectively.

In!) .53.?5‘Wm1‘k V

,\

CHAPTER 4
EXOGENOUS BUT NOT ENDOGENOUS PGEz MODULATES PONY

TRACHEAL SMOOTH MUSCLE CONTRACTIONS

Introduction
Some horses and ponies that are housed in barns and fed hay develop a type of
chronic obstructive pulmonary disease colloquially called "heaves. " Like asthma in
humans, heaves is characterized by periods of airway obstruction and hyperresponsive-
ness followed by periods of disease remission (Derksen et al. 1985a). The airway
obstruction in affected animals occurs mainly through stimulation of muscarinic receptors
because R, decreases and Cd,n increases following atropine administration (Broadstone
et al. 1988).
In vitro experiments utilizing isolated tracheal smooth muscle strips have shown
that, compared to controls, muscle from heavey horses is slightly hyporesponsive to
e=xogenous ACh, but hyperresponsive to EFS (Broadstone et al. 1991). One possible
eJtplanation for these observations is a defective regulation of cholinergic neurotrans-
Ihission in affected animals, such that nerve stimulation results in increased ACh release
at the neuromuscular junction.
The parasympathetic cholinergic innervation of the airways is modulated by a

I5111111ber of factors, including prostaglandins (Ito 1991). An inhibitory effect of PGE2 on

83

84

parasympathetic nerves in canine airways is well documented (Waters et al. 1984;
Nakanishi et al. 1976; Ito and Tajima 1981a; Ito et al. 1990; Daniel et al. 1987; Shore
et al. 1987; Deckers et al. 1989a). Epithelium is an important source of prostanoids in
the airways and, in the normal horse, PGE, is the major prostanoid released by tracheal
epithelium (Gray et al. 1992b). Furthermore, airway mucosal tissues of heavey horses
produced less PGFQ in response to stimulation by certain agonists than those of control
horses (Gray et al. 1992a). Therefore, it is possible that a lack of PGE2 in the heavey
horses may result in increased release of ACh from parasympathetic nerves. This would
result in bronchospasm and the clinical signs of airway obstruction typical of heaves.
Before determining if a lack of PGE2 is an important mechanism in the
pathogenesis of heaves, it is ﬁrst necessary to establish the role of PGE2 in the
modulation of airway parasympathetic nerves of normal animals. The purpose of this
study was 1) to determine if exogenous PGE2 can modulate the response of tracheal
Smooth muscle to EFS and ACh; and 2) to investigate, by means of cyclooxygenase

inhibition, if endogenous PGE2 modulates airway smooth muscle contraction.

Materials and methods
One female and 10 male ponies, weighing between 104 and 209 kg, were studied.
They had no history of respiratory disease, showed no clinical signs of respiratory tract
d isorder and, post-mortem, their lungs and airways were normal in gross appearance.
Immediately after euthanasia by intravenous injection of an overdose of pentobarbital
SOciium, the thoracic cavity was opened. The trachea was quickly removed and immersed

in Krebs-Henseleit solution (composition in Mm: NaCl 118.4, NaHCO3 25.0, Dextrose

«‘9, . x’iiivnrpc ‘

85
1 1.7, KCl 4.7, Cac1,.2H,o 2.6, Mgso, 1.19, KHZPO, 1.16), which had been bubbled

with 95% Oz and 5% C02. It was then transported a short distance to the laboratory
where it was gassed with 95 % Oz and 5 % CO2 continuously.
Tracheal smooth muscle strips with epithelium intact were prepared from the ﬁrst
2.0 tracheal rings above the carina. The muscle was cut with a template along the ﬁber
direction. Each strip measured 3.5 x 10 mm. The strips were suspended by 3-0 surgical
silk in 10-ml tissue baths containing Krebs-Henseleit solution bubbled with 95 % 02-5 %
CO2 and maintained at 38°C. One end of the tissue was tied to the hook at the lower
end of an electrode holder that was placed in the muscle bath. The other end was tied
to a force transducer (Grass FTO3) mounted on micromanipulators so that the tissue
length could be changed. A continuous recording of isometric force generation was
obtained by a polygraph (Grass Model 7D, 7E, 79D) connected to the force transducer.
Tissues were suspended between platinum electrodes for electrical stimulation. Electrical
impulses were produced by a Grass S88 stimulator and passed through a stimulus power
booster (Stimu-Splitter II, Med Lab Instruments, Fort Collins, Colorado). The electrical
illipulses consisted of square waves.
Tissues were equilibrated for approximately 90 min with an initial passive tension
of 2 g applied and maintained. In preliminary experiments, it was determined that 2
grams is the optimal resting tension for the trachealis strips. During the equilibration,
tissues were electrically stimulated for 2-3 min at 8-10 min intervals (15 v, 0.5 ms pulse
duration, 16 Hz) until the baseline was stable and the magnitude of response to the same
StiInulus was consistent. The bath solution was changed every 15 min. At the end of

equilibration, the Krebs-Henseleit solution was replaced by 127 mM KCl-substituted

an. 0.. 3"”. s

86

Krebs-Henseleit solution, and the contractile response to this solution was recorded.
Tissues were then rinsed with Krebs-Henseleit solution several times until the muscle

tension returned to baseline.

Protocol 1: Eﬁ'ect of exogenous PGE2 on contractions induced by EF S and ACh
Seven tracheal muscle strips from each pony were suspended between rectangular
electrodes placed parallel to the muscle. Meclofenamate (10“ M) was added to each
tissue bath. After a 40-min incubation period, guanethidine (10‘5 M) was added to the
bath. Sixty min after adding meclofenamate, PGE2 was added to six tissue baths and
ethyl alcohOl (vehicle) equivalent to the volume used to dissolve 10*s PGE2 was added to
a seventh bath to give a ﬁnal concentration of 0.1% of ethyl alcohol. One concentration
of PGE2 was added to each bath (10‘, 107' 10"“, 109, 10‘”, 10“ M). Twenty min after
adding PGE, or its vehicle, contractile responses to increasing frequencies (0.5, l, 2, 4,
8, 16, 32 Hz) of EFS (15 v, 0.5 ms pulse duration) were obtained. Each stimulus was
applied until a plateau in tension was achieved, approximately 2—3 min. The interval
between consecutive stimuli was 6 min.
Following EFS, ACh was added to the bath cumulatively at concentrations of 103,
1 0‘7, 10", 10‘s, 10“, and 10'3 M. When tension had plateaued following addition of a
concentration of ACh, the next concentration was added. The tissues were then removed
from the bath, blotted with ﬁlter paper, and weighed (wet weights).
An experiment was conducted to determine if the PGE2 concentration in the
muscle bath remained stable after addition of exogenous PGFq. One muscle strip was

incubated with 10" M meclofenamate for one hour and exogenous PGE was added to

87

the muscle bath. PGE2 samples were taken immediately and after 110 min, the

approximate time needed to conduct the experiments with EFS and ACh administration.

Protocol 2: Eﬁects of cyclooxygenase inhibitors on airway smooth muscle contraction

Two sets of eight tracheal strips were prepared from each pony. One set was
used to study the effect of cyclooxygenase blockade on the response to EFS, another set
to study the effect on the ACh dose-response curve. Tissues were suspended between
ring-shaped electrodes. Three concentrations of aspirin (103, 10“, 10’5 M) and
meclofenamate (10" 10‘, 108 M) and their solvents (methanol and distilled water,
respectively) were added separately to eight pairs of organ baths. The amount of solvent
administered to the control baths was equal to that used to dissolve the highest
concentration of inhibitors. Forty min after adding the inhibitors, guanethidine (105 M)
was added to each organ bath and 20 min later, 0.5 ml of bathing solution was taken
from each bath for measurement of PGE2 concentration and replaced with the same
volume of Krebs-Henseleit. As soon as the PGB2 samples had been taken, ACh was
added cumulatively (103' 10”, 10‘, 10's, 10", 103 M) to one set of strips and the tissues
of the other set were stimulated electrically as described under protocol 1.

Upon completion of the ACh administration or EFS, another 0.5 ml of bathing

solution was taken for measurement of PGE2. The tissues were then removed from the

bath, blotted, and weighed as in protocol 1.

88

Measurement of PGE2

Tissue bath samples were stored at -80°C and PGE2 was measured by radioimmu-
noassay in batches after all samples had been collected. The PGE2 antiserum was
obtained from Advanced Magnetics (Cambridge, MA), tritiated PGEz from Dupont—New
England Nuclear (Boston, MA), and authentic PGFQ standard from Cayman Chemical
Co. (Ann Arbor, MI). The crossreactivity of the antiserum at 50% B/Bo was PGE,
100%, PGE, 50%, PGA2 6%, PGan 1.3%, and all other eicosanoids 1% or less. The
lower limit of PGE2 detection was 3.05 pg! 100 p1. When concentrations were below the
limit of detection, the value of 3.05 pg/100 pl was used for statistical purposes. Tissue
production of PGE2 (ng/ g wet wt) was calculated from bath concentration (pg/ 100 pl),

bath volume, and tissue wet weight.

Drugs

The following drugs were used: acetylcholine chloride, aspirin (Sigma), sodium
monohydrate meclofenamate (courtesy of Parke-Davis Pharmaceutical Research Division,
Warner-Lambert Company), guanethidine monosulfate (courtesy of CIBA Pharmaceutical
Co.), PGE, (Cayman Chemical Co.). All drugs were dissolved in distilled water except
PGE,2 and aspirin. Aspirin was dissolved in methanol to make a solution of 101 M,
which was subsequently diluted with distilled water. The PGFq was dissolved in ethyl
alcohol to 103 M and stored at -20°C. The stock solution of PGE2 was diluted with
distilled water before use in each experiment. Drugs were added in a volume equal to
1 % of the total volume of solution in the muscle bath. Concentration of drugs was

eXpressed as the ﬁnal bath concentration.

89
Statistical analysis
Responses to EFS and ACh were calculated as a percentage of the maximum
contraction produced by 127 mM KCl. Data were analyzed by between-within/split-plot
AN OVA, single factor randomized design, and paired t-test as appropriate. P < 0.05
was considered statistically signiﬁcant. All values are expressed as mean ;t SE. n =

number of animals.

Results

Trachealis muscle contracted in a frequency-dependent manner in response to
EFS. All contractions were abolished by atropine (1045 M) and by tetrodotoxin (1045 M)
but not affected by hexamethonium (10'5 M), indicating that they resulted from
postganglionic cholinergic nerve stimulation and activation of muscarinic receptors. ACh

produced a dose-dependent trachealis contraction that was abolished by atropine.

Protocol 1: Eﬁect of exogenous PGE2 on contractions induced by EFS and ACh
Muscle contraction in response to EF S was inhibited in a dose-dependent manner
by PGE,. Concentrations of PGE2 10" M and above caused signiﬁcant inhibition. This
inhibitory effect was stronger at low than at higher frequencies. When PGE, concentra-
tion was equal to or less than 10'9 M, no inhibitory effect was observed (Fig. 4-1a).
Prostaglandin E, also inhibited ACh-induced contractions, but the minimum concentration
of PGE, needed to signiﬁcantly inhibit ACh-induced contractions was 10’7 M, higher than
that needed to inhibit EFS-induced contractions (Fig. 4-1b). The concentration of PGE,

in the muscle bath remained stable during the llO—min incubation period, 197.0 ng/ 100

90
pl immediately after exogenous PG]?q addition and 201.2 ng/100 pl at the end of

incubation.

Protocol 2: Eﬁ’ects of cyclooxygenase inhibitors on tracheal smooth muscle contraction

After the 60-min incubation period, concentrations of PGE2 in the organ bath
averaged 9.8 :l; 1.7 x 10'10 M (mean :1: se) in the absence of cyclooxygenase inhibitors.
Neither meclofenamate nor aspirin had any signiﬁcant effect on the frequency-response
curves or the ACh dose-response curves (Fig. 4-2, 4-3).

Meclofenamate (10*6 and 10“ M) inhibited PGE, production both before and after
EFS and ACh-induced contractions. At 10'8 M, the inhibitory effect of meclofenamate
was variable (Fig. 4-4a, 4-5a). Aspirin (104 and 10‘3 M) also inhibited PGE, production
but 105 M did not (Fig. 4-4b, 4-5b). The increase in PGE, levels that followed EFS was
suppressed by both 1045 and 10“ M meclofenamate but only by 103 M aspirin (Fig 4-4a,
4—4b). The tendency for PGFQ to increase following ACh was suppressed by 10‘ and

10“ M meclofenamate and by 10“ and 103 M aspirin (Fig. 4-5a, 4-5b).

Discussion

The purpose of the present study was to determine the role of PGE, in the
regulation of equine airway smooth muscle. Two approaches were used: the administra-
tion of exogenous PGE, and inhibition of endogenous PGE2 synthesis by cyclooxygenase
inhibitors.

Our results clearly show that exogenous PGE, inhibits tracheal smooth muscle

Contractions induced by both EFS and ACh. The action of PGE, on ACh-induced

91

contractions must be direct on smooth muscle, but the effect on EFS-induced contractions
could be directly on smooth muscle, indirectly through nerves, or both. The minimum
concentration of PGE, Med to inhibit contractions induced by EF S was 10" M. This
was less than the 107 or 10‘ M needed to inhibit contractions induced by ACh. These
observations suggest that the effect of PGFq on EFS-induced contraction was exerted in
part directly on muscle and in part through nerves. The equine trachea receives
parasympathetic, sympathetic, and NAN C innervation (Broadstone et al. 1991). In the
present experiments, the function of sympathetic nerves was blocked by the administra-
tion of guanethidine. We were unable to block the NAN C system, because the
neurotransmitter had not been identiﬁed at the time of this study and there is therefore
no blocker. It is most likely that the contractions we observed during EFS were due to
activation of postganglionic parasympathetic nerves, because contractions were eliminated
by both atropine (10‘) and tetrodotoxin (10‘) and were unaffected by hexamethonium.
The inhibitory effect of exogenous PGE,, on nerves must therefore have been exerted on
parasympathetic postganglionic nerves.

The inhibitory effect of PGE2 on parasympathetic nerves of dog airways is well
documented. As we observed in the pony, the parasympathetic nerves of dog airways
have a higher sensitivity to the inhibitory effect of PGE, than does airway smooth
muscle. The actual concentrations of PGE, needed to inhibit the nerves and muscle vary
widely in different studies. Ito et al. reported that when PGEz concentration is between
10“ and 109 M, it only inhibits EJPs. In doses over 10'8 M, PGE,2 also reduces muscle

tone and dose-dependently suppresses the amplitude of twitch contractions (Ito and

Tajima 1981a; Ito et al. 1990). Daniel et al. (1987) observed that 109 and 108 M PGE,

92
inhibits the contractile response to EFS as well as FJPs remarkably, but the inhibitory

effect on ACh-induced contractions is signiﬁcantly less. Waters et al. (1984) found that
3 x 10”3 M PGE, inhibits the contractile response to EFS signiﬁcantly but shows little
effect on ACh-induced contractions. Similar results have also been obtained by
Nakanishi et a1. (1976) but with much higher concentrations of PGE2 (2.8 x 10'7 and 2.8
x 10*5 M). In equine trachealis electrically stimulated using parameters that caused
muscle contraction by both a direct and a neurally mediated mechanism, Gill and
Kroeger (1990) observed that exogenous PGE, at 10‘ M inhibited the neural component
and at 107 M inhibited both the neural and myogenic components. The inhibitory PGE,
concentrations reported by Gill and Kroeger are identical to those of the present study.

From our results with exogenous PGE,, it is apparent that endogenous PGFq could
affect the frequency-response curve only at concentrations equal to or greater than 108
M. The concentration of PGE, in the muscle bath after a one-hour incubation period
averaged 9.8 x 1010 M and after EFS was 1.9 x 109 M in the absence of cyclooxygenase
inhibitors. These concentrations were well below those that inhibited the response to
EFS. Because PGE, is lipid soluble and can easily pass through cell membranes, it is
unlikely that the PGE, concentration in the muscle was much higher than that in the
organ bath. Therefore, even if production of all the endogenous PGE, was inhibited, a
signiﬁcant leftward shift of either the frequency-response curves or ACh dose-response
curves would not be expected. Moreover, the muscle contractions in response to BE S
should have been markedly inhibited if the PGIEL2 concentration inside the muscle strip
had exceeded 108 M. Based on the measured levels of PGE2 in the muscle bath and the

observed inhibitory effect of exogenous PGE,, it seems unlikely that endogenous PGFQ

93
has inhibitory effects in pony trachealis muscle. To further test this possibility, we

examined the effect of cyclooxygenase blockade on the muscle’s response to RF S and
ACh.

Indomethacin is the most frequently used cyclooxygenase inhibitor for this type
of experiment. We had planned to use indomethacin, but it frequently induced
spontaneous contractions of tracheal smooth muscle, an effect that appeared to increase
at higher concentrations. Because of the spontaneous contractions, it was difﬁcult to
measure the contractile responses to EF S and ACh accurately. Therefore, two other
cyclooxygenase inhibitors instead of indomethacin were used in this study.

Both meclofenamate and aspirin blocked PGE, synthesis. At the highest
concentrations of these drugs used, PGE2 concentration after a one-hour incubation period
was decreased from 9.8 x 10'10 M to 3.3 x 1010 M with 10‘3 M aspirin, to 3.0 x 1010 M
with 106 M meclofenamate, and to 1.7 x 10‘10 M with 10“ M meclofenamate. The value
of PGE2 after 10“ M meclofenamate is even an overestimate of the actual PGE2
concentration, because several PGE, samples were below the measurable range yet were
reported as the lowest detectable value. These same high concentrations of aspirin and
meclofenamate also blocked the increase in PGE,2 following EFS.

Despite the decrease in PGE, concentration in our organ baths, cyclooxygenase
inhibitors did not affect the contractile response of tracheal smooth muscle to either EFS
or ACh. These results are analogous to those of Daniel et al. (1987) obtained in canine
tracheal strips dissected free of epithelium. In their experiments, indomethacin had little
effect on the amplitude of EJPs elicited by electrical stimulation. By contrast, others,

in both in vitro (Waters et al. 1984; Shore et al. 1987) and in vivo (Bethel and McClure

1990)
respo:
cyclo<
that a
muscl
increa
contra
endog
1990)
rather
Howe
rESpm
CndOg

0f air

effect
adm,n
not in
and'r
WhEez
it “ml:

“353 a

94
1990) studies, reported that cyclooxygenase blockade enhances the canine trachealis

response to electrical stimulation. In the only other reported study of the effect of
cyclooxygenase blockade on equine trachealis, Gill and Kroeger used stimulus parameters
that were very different from those of the present study in that they had direct effects on
muscle as well as activating nerves. These investigators observed that indomethacin
increased the contractile response to BE S, especially when the neurogenic component of
contraction was blocked with atropine. These observations led them to conclude that
endogenous prostanoids exert modulatory effects in horse trachealis (Gill and Kroeger
1990). Because the stimulus parameters were so different and because we studied ponies
rather than horses, it is difﬁcult to compare our results with those of Gill and Kroeger.
However, the failure of both meclofenamate and aspirin to shift the EFS and ACh
response curves suggests that, under the conditions of the present experiments,
endogenous prostanoids do not exert a tonic inhibition on either parasympathetic nerves
or airway smooth muscle.

Our conclusion that endogenous prostanoids do not exert an important modulatory
effect on pony airway smooth muscle is supported by in vivo observations. Several days’
administration of the cyclooxygenase inhibitor ﬂunixin meglumine to normal ponies did
not induce airway obstruction (Gray et al. 1989). This is unlike the observation of Ito
and Tajima (1981b), who reported that dogs treated for 5 days with indomethacin began
wheezing and coughing. The lack of an important role for PGE2 in normal ponies makes
it unlikely that an absence of PGE2 is an important factor in the airway hyperresponsive-

ness and airway obstruction characteristic of heaves.

95

(a) 200 Force (% of response to 127 mM KCL)

_.. women + 10“ M + td‘° M —a— 10° M
->e- 10" M +107 M -A— 10‘ M

 

 

03'- hi

0.5 1 2
Frequency (Hz)

(b) Force (% of response to 127 mM KCL)
200 ’

—.— vehicle + 10'"M -*— 1610M -B- 100 M
-)(—1o‘°M +107M -A-10°M

 

 

 

Acetylcholine concentration (-log M)

Fig. 4—1 . Effect of exogenous PGE2 on trachealis contractions induced by EFS (Graph
a) and ACh (Graph b). n = 5. Single factor randomized design AN OVA was used to
compare force at each frequency of stimulation and at each concentration of ACh. * =
signiﬁcantly different from the corresponding value in the vehicle-treated group.

(a) 200 Force (% of response to 127 mM KCL)

150*-

100-

    

10'0 M meclofenamate
10“ M meclolenamate

10“ M mecroronamaia

 

COOUOI

l l

0.5 1 2 4 8 16 32

b

(b) 200 Force (‘36 of response to 127 mM KCL)

150

100

 

 

0 . t 1 r r r
0.5 1 2 4 8 16 32

Frequency (Hz)

Fig. 4-2. Effect of meclofenamate and aspirin on contractions induced by EFS in
trachealis. Single factor randomized design AN OVA was used to compare force at each
frequency of stimulation. No statistically signiﬁcant differences were observed. 11 = 8
in aspirin vehicle, 10‘5 and 10“ M aspirin-treated groups. n = 7 in the rest.

97

(a) 200 Force (‘1. oi response to 127 mM KCL)

* 104 M meclofenamate
-0- 104 M meclofenamate
‘A' 10‘4 M meclofenamate
-X- control

  
   

 

150-

100

 

oL

(b) Force (% of response to 127 mM KCL)

*10’5 M aspir'n
-0- 10" M aspirin
-A- 10.3 M aspirin
* vehicle

150-

 

100

 

 

Acetylcholine concentration (-Iog M)

Fig. 4»3. Effect of meclofenamate and aspirin on contractions induced by ACh (ACh)
in trachealis. Single factor randomized design was used to compare force at each
concentration of ACh. No statistically signiﬁcant differences were found. n = 6 in 10"
and 10" M meclofenamate-treated groups. n = 7 in 10‘ M meclofenamate and 103

aspirin-treated groups. n = 8 in the rest.

A PGE;(ng/g wet wt.)

 

 

98

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

200T + DPre EFS ZlPost EFS
150 - I
100 -
50 — / * *
* ir
/ T _L *
_I_
, // A , Fri-be
0 10 nM 1 MM 100 pM
Meclofenamate concentration
B PGE2 (ng/g wet wt.)
2°°'l El pre EFS E2 post EFS
1 so — #
+
100 — T /
/ / ..
so — _L / / *
/ / * *
// // /,
vehicle 10 pM 100 [1M 1 mM

Aapirln concentration

Fig. 44. Effect of meclofenamate and aspirin on PGFQ production by tracheal strips
subjected to EFS. Single factor randomized design AN OVA was used to compare PGE,,
levels at different concentrations of the two drugs and paired t-test was used to compare
values before and after EF S. * = signiﬁcantly different from the corresponding control
value. # = signiﬁcant increase in PGE2 production after EFS. + = p values of 0.051,
0.052, and 0.055 for comparison of pre- and post-EFS PGE, production in the
meclofenamate vehicle, 10‘8 M meclofenamate and aspirin vehicle-treated tissues,

respectively. n = 5 in all groups.

 

 

subli‘rCted
prOdUClio,

VaIUes be 1

 

 

 

99

A PGE2 (rig/g wet wt.)
l [3 pre ACh E post ACh

200— I
150~

100‘

 

#

T
T ** **
i2 ﬂew?

10 nM 1pM 100pM

 

 

r—'———i
°\\\\\\\

50-1

 

 

 

 

 

 

 

 

Meclofenamate concentration

P652 (ng/g wet wt.)

l D pre A011 12 post ACh
200 -

150i T T J

100‘

 

 

 

 

 

 

 

 

 

 

 

50;J_ _L/ T T
‘ 6.

.LI TT
//. //

vehicle 10 pM 100 pM 1 mM

 

 

 

 

 

 

 

 

 

 

Aspirin concentration

Fig. 4-5. Effect of meclofenamate and aspirin on PGE2 production by tracheal tissues
subjected ACh. Single factor randomized design AN OVA was used to compare PGE,
production at different concentrations of the two inhibitors and paired t-test to compare
values before and after ACh. * = signiﬁcantly different from the corresponding control
value. # = signiﬁcant increase in PGE2 production after EFS. n = 4 in 10" M
meclofenamate-treated group. n = 5 in the rest.

CHAPTER 5
PGE2 INHIBITS ACETYLCHOLINE RELEASE FROM CHOLINERGIC

NERVES IN CANINE BUT NOT EQUINE AIRWAYS

Introduction

Heaves is a type Of chronic Obstructive pulmonary disease in horses and ponies.
Affected animals develop airway Obstruction when stabled and fed hay and enter
remission at pasture. Atropine markedly reduces the increased airway resistance of
horses with heaves, indicating the involvement of muscarinic receptors in airway
Obstruction (Broadstone et al. 1988). A comparison Of the response Of airway smooth
muscle to ACh and EFS suggests that increased ACh release from airway cholinergic
nerves is at least in part responsible for the increased response to muscarinic blockade
in horses with heaves (Broadstone et al. 1991). The elevated release of ACh from
cholinergic nerves could be due to alterations in prejunctional regulation.

Prostanoids have an important modulatory effect on airway cholinergic
neurotransmission. An inhibitory effect of PGE2 on airway cholinergic neurotransmission
is well documented in dogs. Low concentrations of exogenous PGFq inhibit the
contractile response Of airway smooth muscle to EFS but have little or no effect on that
to exogenous ACh. Furthermore, inhibition of endogenous prostanoid production by

cyclooxygenase inhibitors potentiates the response tO EFS with little or no effect on the

100

101
response to ACh (Nakanishi et al. 1976; Waters et al. 1984). PGE, also inhibits the

amplitude Of EFS-induced EJPs (Inoue et al. 1984; Ito and Tajima 1981a). Direct
measurements of ACh release have shown that PGE, inhibits ACh release from airway
cholinergic nerves and that cyclooxygenase inhibition has the Opposite effect (Deckers
et al. 1989a; Shore et al. 1987).

The airway subepithelial tissues Of heavey horses have a reduced ability to release
PGE, when compared with those of control horses (Gray et al. 1992a). Based on the
above evidence from dogs, I hypothesized that PGE, can inhibit ACh release from the
cholinergic nerve terminals Of control horses and that this effect Of PGE2 is lacking or
attenuated in heavey horses. In order to test this hypothesis, it was ﬁrst necessary to
determine if PGE, has a modulatory effect on airway cholinergic neurotransmission in
control horses.

In the study that I described in Chapter 4, I concluded that exogenous PGE2 has
an inhibitory effect on cholinergic neurotransmission in pony trachea because it has a
greater inhibitory effect on the contractile response Of trachealis to EFS than to ACh, and
that endogenous prostanoids have no modulatory effect on cholinergic neurotransmission
in pony trachea because inhibition of cyclooxygenase has no inﬂuence on the contractile
response Of trachealis to EFS. However, this second conclusion is not in agreement with
results Obtained in horse airways. Yu et al. (in press a) Observed that cyclooxygenase
inhibition augmented the contractile response Of horse tracheal and, in particular,
bronchial smooth muscle preparations to EFS to a much greater extent than the response
to exogenous ACh, suggesting that endogenous prostanoids inhibit cholinergic

neurotransmission in the horse airways. In order to clarify the role Of PGE2 on equine

102

airway cholinergic neurotransmission, it became essential to directly measure ACh
release from airway cholinergic nerves.

In pilot studies conducted on several horses, I was unable to detect an effect of
either PGE, or cyclooxygenase inhibition on ACh release from airway cholinergic nerves
even though such an effect had been demonstrated in dogs. The purpose Of the present
experiment was therefore to compare the effect Of exogenous PGE, and endogenous
prostanoids on ACh release from horse and dog tracheal and bronchial cholinergic

DCI'VCS .

Materials and methods
Preparation of tissues

Seven dogs (three female, four male, body weight 18.8 :1: 1.2 kg) and ﬁve horses
(two female, three male, body weight 421.4 :1: 41.5 kg) were used for this study. The
horses had no history Of heaves. Immediately after euthanasia by intravenous injection
of an overdose of pentobarbital sodium, the thoracic cavity was Opened and the trachea
and lungs were removed. The lungs from all the animals were normal in gross
appearance. In dogs, the trachea and both lungs were immersed in Krebs-Henseleit
solution (composition in mM: NaCl 118.4, NaHCO3 25.0, dextrose 11.7, KCl 4.7,
CaClz-ZHZO 2.6, MgSO4-7H20 1.19, KHzPO4 1.16) that had been bubbled with 95%
02/5 % C02. In horses, the trachea between the twenty-ﬁrst and thirtieth cartilaginous
rings above the carina and a piece of lung tissue from the middle part Of one lung were

immersed in Krebs-Hemeleit solution. The tissues were transferred to the laboratory

103
within 10-15 min after dissection, where they were continually gassed with 95 % 02/5 %

C02.

Tracheal and bronchial smooth muscle preparations with epithelium were made
from the airways Of both species. Typically, six tracheal and six bronchial preparations
were prepared from each animal. In dogs, the trachea was cut into segments containing
two or three cartilage rings. The loose connective tissue on the serosal side was
dissected away. The dorsal tracheal membrane adjacent to the tips of the cartilage was
tied with silk suture. The tissue, which included mucosa, serosa and all tissues between,
was then out free from the cartilage. Bronchial rings (3-5 mm in outside diameter, 15
mm in length) were prepared from random locations Of both lungs by dissecting away
the surrounding lung parenchyma.

In horses, trachealis strip bundles were prepared using the method described
previously (Chapter 2). In addition, bronchial rings Of size similar to those Of the dogs

were prepared.

Protocols

The tissue preparations were then suspended in twelve 2-ml tissue baths (Radnoti
Glass Technology, Inc. , Monrovia, CA). Each bath had a pair of parallel platinum wire
electrodes built against the wall in the vertical direction. For tracheal strips or bundles,
one end of the preparation was ﬁxed to the tip of a glass tissue holder. The suture on
the other end was attached to a steel bar through a piece Of elastic. The tension applied
to the tissue was well within the range Of 2—20 grams, which has been determined to

have no inﬂuence on ACh release from airway cholinergic nerves (Chapter 2). The

104

bronchial tubes were gently guided on to the shorter arm Of an "L"-shaped glass tissue
holder. To prevent the bronchial tube from ﬂoating above the surface Of the bath
solution, a piece of silk suture was passed through a hole in the bronchial wall and tied
to the hook at the end of the tissue holder. Tissue baths contained Krebs-Henseleit
solution, which was maintained at 38°C and bubbled with 95 % 02/5 % C02.

Tissues were equilibrated in the baths for approximately two hours, during which
time the bath solution was replaced at 10-15 min intervals. This extended equilibration
period was necessary in order to reduce the amount Of foam formed during bubbling.
In a 2-ml bath, foam formation results in overﬂow of bath solution so that ACh release
cannot be accurately determined. After equilibration, tissues were used to study the
effect Of either endogenous prostanoids or exogenous PGE, on ACh release. The general
protocols are described below and the speciﬁc experimental procedures for each bath are
listed in Tables 5-1 and 5-2.

To determine the effect Of endogenous prostanoids on airway cholinergic
neurotransmission, tissues were incubated with atropine (3 x 107 M) and neostigmine
(1.5 x 1045 M) for 30 min before four periods Of 12-min EFS were applied. Period 1
was applied in the absence Of cyclooxygenase inhibitors, whereas periods 2-3 were in the
presence Of a cyclooxygenase inhibitor or its vehicle. A 60-min period was allowed
between period 1 and period 2 for incubation with a cyclooxygenase inhibitor or vehicle.
Intervals Of 30 min were allowed between subsequent stimulation periods. Indomethacin
(3 x 10‘5 M) was used as the cyclooxygenase inhibitor for dog tissues. For horses, it was
necessary tO compare the results with those Obtained from dogs as well as to closely

mimic the conditions of an earlier indirect study (Yu et al. in press a) in which

105

meclofenamate was used as a cyclooxygenase inhibitor. For these reasons, both
meclofenamate (1045 M) and indomethacin were used, and EFS with a set of different
parameters was applied to the meclofenamate-treated tissues.

Tissues used tO determine the effect of PGE, were preincubated with atropine (3
x 107 M), neostigmine (1.5 X 10‘5 M), and indomethacin (3 x 10“ M) for 120 min
before four 12-min periods Of EFS were applied at 30-min intervals. The ACh release
during period 1 served as control. After period 1, in addition to the three preincubation
drugs, either PGE, or its vehicle was added to each tissue bath. Because Shore et al.
(1987) had reported a time-dependency of the effect of PGE2 on ACh release in dog
airways, only one concentration Of PGE2 was added to each bath containing dog tissues,
and ACh release was measured following the three subsequent EFS periods. However,
analysis Of our dog data revealed no such time—dependency, and therefore PGE, was
added tO horse tissues in a cumulative way. The tissues were incubated with each
concentration Of PGE, or its vehicle for 20-25 min before they were electrically
stimulated in the presence of this concentration Of PGE2 or its vehicle. The concentration
Of PGE, vehicle used in the dog experiment was equivalent to that used to dissolve 107
M PGE,. In the horse experiment, vehicle concentration matched that used to dissolve
each concentration Of PGE,.

At the end Of each EF S period, 1.5 ml of the bath solution was collected from
each bath for ACh analysis. The tissue bath was ﬂushed with Krebs-Henseleit solution
containing atropine (3 x 107 M) and neostigmine (1.5 x 10*5 M) with or without indo-
methacin (3 x 10“3 M) for 2-3 times before and after each EFS period. Each time the

bath solution was replaced, various drugs were replenished at appropriate concentrations.

106

Electrical impulses were produced by a stimulator (S88, Grass Instrument CO. , Quincy,
MA) and passed through a stimulus power booster (Stimu-Splitter II, Med-Lab
Instruments, Loveland, CO). The electrical impulses consisted of square waves. The
parameters of EFS were 15 V, 5 Hz, and 2 ms for all tissues except those used to study
the effect of meclofenamate. In the latter case, the parameters were 15 V, 1 Hz, and 0.5
ms, which had been used in the previous- indirect study (Yu et al. in press a).

The wet weight of each tissue preparation was determined at the end Of the
experiment after blotting with ﬁlter paper. It averaged in mg 153.7 :l: 5.2 (n = 36),
151.8 :1: 5.3 (n = 36), 210.7 :I: 5.6 (n = 30), and 193.5 :1: 16.8 (n = 30) for dog

tracheal, dog bronchial, horse tracheal, and horse bronchial preparations, respectively.

ACh analysis

Bath liquid samples were ﬁrst ﬁltered through 0.2 p nylon membranes (Acrodiscs
13, Gelman Sciences, Ann Arbor, MI) and then analyzed by HPLC coupled with
electrochemical detection. The method was the same as that I described in Chapter 2
except for the following modiﬁcations: 1) A precolumn (Bioanalytical System, Inc. , West
Lafayette, IN) containing choline oxidase and catalase was added before the analytical
column to remove choline in the samples. This enabled me tO shorten the length of each
run without concern about the separation of ACh and choline. 2) The mobile phase was
50 mM NazHPO4 (pH = 8.5) and the ﬂow rate was 1 ml/min. 3) A calibration standard
Of 20 pmol ACh was injected every 4 samples and the ACh amount in the samples was
calculated based on the bracketed calibrations. 4) A new autosampler (ISS200, Perkin-

Elmer, Norwalk, CT) replaced the Old model.

107

Drugs

Neostigmine methylsulfate, atropine sulfate, and indomethacin (Sigma Chemical
CO. , St. Louis, MO.) solutions were prepared on the day Of experiment. Neostigmine
and atropine were dissolved in water to concentrations of 102 and 103 M, respectively.
Indomethacin was dissolved in solutions Of equimolar sodium carbonate (Malinckrodt,
Inc. , Paris, KY) tO a concentration Of 102 M. These solutions Of drugs were mixed into
a bulk volume Of Krebs-Henseleit solution to reach their ﬁnal concentrations. PGE2
(Cayman Chemical Company, Ann Arbor, MI) was ﬁrst dissolved with methanol (1
mg/ml) and subsequently diluted to 10.5 M with distilled water. The 105 M PGE, was
stored as a stock solution at -15°C and diluted with water before use. Meclofenamate
sodium monohydrate (courtesy of Parke-Davis Pharmaceutical Research Division, Ann
Arbor, MI) was dissolved in and diluted with water on the day Of experiment. The PGE2
and meclofenamate solutions were pipetted into the tissue bath in 20 pl volume (1% Of
the bath volume). The concentrations Of all drugs were expressed as their ﬁnal bath

concentrations.

Statistical analysis

TO determine the effects Of cyclooxygenase inhibition and exogenous PGE, on
ACh release, the ACh release in each bath during the ﬁrst EFS period was regarded as
100% , and the release during subsequent periods was expressed as a percentage Of this
value. Depending on the number Of treatment groups, either an unpaired t-test or one-
way AN OVA for randomized measures followed by Tukey’s HSD test was used to

determine the effects of PGE2 or cyclooxygenase inhibition. TO compare data between

108

the two species, ACh release was expressed as pmol/g/min and analyzed by unpaired t-

test. All values were expressed as mean i SE. P < 0.05 was considered statistically

signiﬁcant.

Results

All tissue preparations released a measurable amount Of ACh during each EF S
period. Typical chromatograms are shown in Figure 5-1. There was a signiﬁcant
difference in ACh release rate (pmol/g/min) between the two species. In tracheal
preparations, the ACh release rate in horse was about twice Of that in dog. In bronchial
preparations, the ACh release rate in dog was approximately three times that in the horse

(Fig. 5-2).

Dog airways

In the absence of cyclooxygenase inhibition, ACh release was fairly constant over
the four stimulation periods. Cyclooxygenase inhibition by indomethacin signiﬁcantly
augmented ACh release in both the tracheal (Fig. 5-3A) and bronchial (Fig. 5-3B)
preparations. The effect was greater in the trachea than in the bronchi. Although ACh
release tended to increase gradually following indomethacin administration in the trachea,
there was no signiﬁcant difference among periods 2 to 4 (Fig. 5-3A).

PGE, concentration—dependently inhibited ACh release from dog airway
cholinergic nerves and the effect was greater in the trachea (Fig. 5-4A) than in the
bronchi (Fig. 5-4B). There was no difference in ACh release values among the three

EFS periods subsequent to administration Of a dose Of PGE,.

109

Horse airways

In both trachea and bronchi, the release Of ACh was constant over the four
stimulation periods in the absence of cyclooxygenase inhibition. In contrast to the results
Obtained in dog tissues, neither indomethacin (Fig. 5-5) nor meclofermmate (Fig. 5-6)
signiﬁcantly inﬂuenced ACh release from either the tracheal or bronchial preparations
Of horses.

In the trachea, 109 and 10'8 M PGE2 did not affect ACh release. However, 107
M PGE, signiﬁcantly augmented ACh release (Fig. 5-7A). PGIE",2 had no effect on ACh

release from horse bronchial preparations (Fig. 5-7B).

Discussion

The major ﬁndings Of the present study were that exogenous PGE,2 inhibited ACh
release from canine airway cholinergic nerves and inhibition Of endogenous prostanoid
production by a cyclooxygenase inhibitor had the Opposite effect. These results could
not be reproduced in horses.

An inhibitory effect of exogenous PGE, and endogenous prostanoids on ACh
release from canine airway tissues has also been reported by Deckers et al. (1989a) and
Shore et al. (1987). Like me, they also Observed that cyclooxygenase inhibition by indo-
methacin augmented ACh release. With respect to the effect of exogenous PGE,, Shore
et al. measured ACh release induced by three repeated 15-min EFS periods in
indomethacin-pretreated tissue and Observed that 10‘8 M PGE, prevented the gradual
increase in ACh release on consecutive stimulation periods. This is similar to my

Observation except that Shore et al. Observed a greater time-dependency in the effect Of

1 10
PGFq. Because the effect Of PGE, was not Observed during the ﬁrst period Of EFS but

developed slowly over time, Shore et al. reasoned that PGE2 may affect the synthesis or
mobilization of ACh rather than the release process itself. In the present study, PGE,
(109- 10'7 M) inhibited ACh release in a concentration-dependent manner after 20-25 min
preincubation and the effect Of PGE, did not change over time. Therefore, my results
favor an effect of PGE2 on the release rather than the synthesis of ACh.

The mechanism Of action of PGE2 on ACh release from canine airway cholinergic
nerves is not fully elucidated. In sympathetic nerves, the mechanism whereby PGI-E‘,2
inhibits norepinephrine release is believed to be inhibition of Ca2+ inﬂux (Malik and
Sehic 1990), probably via the EP3 receptor (Racké et al. 1992). PGE, could be working
through a similar mechanism in canine airway cholinergic nerves.

My results with dog tissues indicate that both indomethacin and PGE2 had a
greater effect on ACh release in the trachea than in the bronchi. These Observations
suggest that the higher response Of tracheal tissue to PGE, and indomethacin is more
likely a result Of higher sensitivity to PGE2 rather than an increased PGFq concentration
in the tracheal tissue.

In the present study, I did not Observe either an inhibitory effect Of exogenous
PGE, or an augmenting effect Of cyclooxygenase inhibitors on ACh release from horse
airway cholinergic nerves. Although the results Obtained with PGE, and cyclooxygenase
inhibitors are in mutual agreement, they are contradictory to the conclusion Of a previous
study (Y 11 et al. in press a), which, based on comparisons Of the contractile response of
horse airway smooth muscle preparations to EFS and exogenous ACh, suggests that

endogenous prostanoids inhibit ACh release from horse airway cholinergic nerves.

111

Failure to Observe an inhibitory effect of exogenous PGE, and endogenous
prostanoids on ACh release from horse airway cholinergic nerves is unlikely to be due
to a lack Of sensitivity of my ACh measuring technique. The sensitivity Of my technique
is well exempliﬁed by the fact that: 1) under the same experimental conditions, I was
able tO conﬁrm the inhibitory effect Of prostanoids on ACh release from dog airway
cholinergic nerves; and 2) using the same technique, I have detected a clear concentra-
tion-dependent inhibition Of ACh release from horse airway cholinergic nerves by a,-
adrenoceptor agonists (Chapter 6).

For technical reasons, direct measurement of ACh release from airway cholinergic
nerves became possible only recently. Previously, most of our knowledge about
prejunctional modulation of airway cholinergic neurotransmission was inferred by
comparing the effect Of a potential neuromodulator on the contractile response Of airway
smooth muscle to EF S and exogenous muacarinic agonist such as ACh. A greater effect
on the response to EF S than ACh suggests prejunctional modulation. TO use this
approach, it is necessary to assume that exogenous and neurally released ACh behave
similarly. However, this is not completely true. Endogenous ACh is released in small
amounts. Its effect is quickly terminated due to its breakdown to choline and acetate by
cholinesterase, which is located ahnost exclusively in or around the nerve terminals
(Baker et al. 1986). The density Of innervation Of airway smooth muscle is generally
sparse. Although smooth muscle cells near the varicosities of cholinergic nerve terminals
are believed to be directly activated by endogenous ACh, it is questionable if ACh
molecules can diffuse to the cells further away before its degradation. These latter cells

are probably activated by electrical coupling through gap junctions or other structures

112
(Gabella 1987; Kannan and Daniel 1980). By contrast, exogenous ACh is usually

administered in larger amounts so that the ACh molecules distribute throughout the tissue
preparation and induce smooth muscle contractions by acting on numerous muscarinic
receptors Of each individual cell. Because the amount Of exogenous ACh is usually in
great excess to the cholinesterase activity Of the tissue, a long-lasting plateau in tension
is Often Observed following addition of each concentration of ACh. Because of these
differences in the actions of endogenous and exogenous ACh, factors that directly affect
airway smooth muscle cells inﬂuence the response Of the tissue to endogenous ACh more
easily than to exogenous ACh. Among all the factors that have been raised as
prejunctional modulators Of airway cholinergic nerves, few lack a direct postjunctional
effect. In addition, all kinds Of nerves in the tissue preparation are activated during EF S,
release Of other neurotransmitters may inﬂuence the response to EFS but not to
exogenous ACh. Furthermore, if a factor has a modulatory effect on gap junctions of
airway smooth muscle cells, it may inﬂuence the contractile response to EFS but not to
exogenous ACh. A remarkable modulatory effect Of PGFa on gap junctions in canine
airway smooth muscle has already been reported (Agrawal and Daniel 1986). For the
above reasons, it may be not always reliable tO infer conclusions about prejunctional
modulation simply by comparing the contractile responses of airway smooth muscle to
EFS and exogenous ACh.

I was not the ﬁrst to Observe a species difference in the response of airway
cholinergic nerves to PGE,. Based on results of an indirect study, Black et al. (1989)
concluded that PGE, inhibits cholinergic neurotransrrrission in rabbit but not human

bronchus. They explained that the difference in the response to PGE2 between species

113

results from a difference in the dependency Of ACh release on voltage dependent Ca2+
channels, based on the Observation that the Ca2+ channel agonist BAY K8644 augmented
the contractile response to EF S in the rabbit but not human bronchus. However, it may
be questionable tO assume a prejunctional effect Of BAY K8644 in the rabbit bronchus
when a postjunctional effect was Observed in the same preparation.

ACh release from airway cholinergic nerves is subject to feedback inhibition
through muscarinic autoreceptors. The nonspeciﬁc muscarinic antagonist atropine can
remove this autoinhibitory effect (Chapters 2, 3). It has been reported that the
augmenting effect of indomethacin on ACh release from canine bronchial tissue can be
demonstrated in the presence but not in the absence of atropine (Deckers et al. 1989a).
Therefore, all tissue preparations were treated with atropine in the present study. In
order to demonstrate the neuromodulatory effect Of a factor, blockade Of muscarinic
autoreceptors may be necessary only in such direct studies in which ACh accumulates
due to prolonged EFS and inhibition Of ACh breakdown by a cholinesterase inhibitor.
The fact that the inhibitory effect of PGE2 on ACh release can be Observed in the
presence Of atropine indicates that PGE, does not exert its inhibitory effect through
muscarinic receptors.

In conclusion, the present study demonstrated that exogenous PGE2 and
endogenous prostanoids inhibit ACh release from canine but not equine airway
cholinergic nerves. The lack of an important role for PGEz in the regulation Of ACh
release from airway cholinergic nerves Of the control horses makes it unlikely that a
decrease Of PGE, production is an important factor in the cholinergically mediated airway

Obstruction characteristic of heaves.

mozmm: Moe ace £828: toeﬁoQ .Tm. oswk

114

Bea unease... n on aoﬁaﬁa 22c aoEooa n mum

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

06% an 26% mm 26% an 26% mm 06% mm 06% ma v
32% mm 06% we 06% mm 96% ma 96% mm 96% mm m a
one? :2 a: x 0 m
:2 B: was :2 B: Nmom :2 E: omen sea? was 5852:er 5852:er N
6.5.60 _
m5 N .NT— m .> w~ wagon—33a mn—m
58 5 so as on as A: 1: a n: 8:285
A: a: x o 5852:er as .ez ,2: x n: oeﬁeaooe A: L: x 0 2322 osewzmooe as A: we a c 2382 .
conﬁne—Sum
_ 0:0.—
m e m e m e m e m a. m e ohmwmesﬁeme
2 : 2 o m e o m e m N _ a sum

 

 

 

 

 

 

 

 

 

 

 

 

£0.”— mac—Swen.—

 

EEEEE 0383x6220

 

 

 

8:8: woe 6e £8666 “co—Eon— .—-m 935.

J armch— UA-Cr—‘UCD
"H
‘ _
[I]
a L rN WNOUOHOHQ UMVN If]!
My mQHUQ i I U
'h

 

115

eotun noes—08:0. H mm ”coca—=83 20c HOT—820 H mum

 

 

 

 

22%.» £2 A2 no: Nun—Om 06% ms 06% mm 06% mm 06% as v

£020.» £2 :2 90: £0; 06% mm 06% 0.0 06% we 06% mu m m
2oz? :2 vs a e A: no: m

0_0E0> «mOm :2 so: 60; 58508er 50050885 85:8 038206.008 N

 

 

 

 

 

 

6550

~

 

 

mEN .Nzn .> 2

ad. no .NI _ .> w—

maosoﬁaaa mum

 

SE o2 6e 32
v2 x 0 50285er 2% A2 92 e
mi 0583.80: A2 e9 x 0 0:623.

as on he :2 a: e n: erwEooe e5 A: L: a c oeaei

6:335

 

 

 

6:82:53“
as £0.85
m h m ._. m H m H m m k 6 PC «0:00;.
2 2 E o m N. e n v N s at Sam

 

 

 

 

 

 

 

 

 

 

 

0mg msoe0woxm

 

nerEE 3.0536220

 

 

 

8:3: 026: 6% £8865 8:800 .N-m 050,—.

 

 

Ll‘wr‘1

 

 

 

 

116

 

 

 

 

 

 

II I
, I],
\ (“'1 Iii
\x If ‘5 l 1"
$ ‘\ l ( ll \._._ c
e. ,I \ "M \
D ~\\"'\ ,/ 1 ' -'*\~\v.
] 5.02 pmol 1:724 meI
. . control ' 10 M PGE”
1
r
'\, r1 r»
,r. r 1
I I k‘.‘ {I ‘5 l l I I III
i \\~ r' \ l'6') 1' '1
A - Me.) A [ ‘l l \r r“, \
x,» ./ ‘ ""
3.47 pmol 5.53 pmol
control 3 x IO‘M indomethacin

 

 

Fig. 5-1. Typical chromatograms showing the effects of PGE2 and indomethacin on
acetylcholine (ACh) release from cholinergic nerves innervating dog trachea. Each pair
of chromatograms represents the ACh amount of two samples from the same tissue bath.
Because the two samples of each pair were analyzed a few hours apart and the detector

sensitivity tended to decrease over time, the proportionality between ACh amount and
the peak height was not constant.

 

at.E\m\_OEn—v 0000.0; Cnu<

Fig
(102
in,

d0g

117

 

 

150 -
- [:Jdog Ihorse
125 - —l—
. _L.
100 '-
75 ’-

0'1
0
l

 

 

ACh release (pmol/g/min)

 

 

 

 

 

 

 

 

 

.Hi
ea

 

 

 

Trachealis strips Bronchial segments

Fig. 5-2. Acetylcholine (ACh) release (pmol/g/min) from cholinergic nerves innervating
dog and horse airways. The values represent the ACh release rate during period 1 in
tissues used to study the effect Of indomethacin and its vehicle. n equals 12 and 10 for
dog and horse tissues, respectively. *P < 0.05 compared with the dog value.

Fig. i
inhen
WEre

Peleas

118

 

 

 

 

 

 

(A) .
320 - _ . *
+ Indomethacrn
*
_ Ovehicle
9 w
'8 240 _
‘C
8
“6
£3
3 160 —
3 + ‘3
g r
O
<
80 e
l l J 1
0 1 2 3 4
(B)
320 — . .
1" indomethacin
r Ovehicle
:: 24o —
B
5
o.
'5 ,, *
§ 160 _ *
o
i
1'» + 4
5 80 —
<
I I l l
0 1 2 3 4

Stimulation period

Fig. 5-3. Effect of indomethacin on acetylcholine (ACh) release from cholinergic nerves
innervating dog trachea (A) and bronchi (B). Four periods Of electrical ﬁeld stimulation
were applied to two groups (n = 6 in each group) of tissue. NO indomethacin was
administered during period 1. During periods 2 to 4, one group was treated with
indomethacin (3 x 1045 M), the other with its vehicle. Indomethacin augmented ACh
release signiﬁcantly. Although ACh release tended to increase progressively following
indomethacin in the trachea, there was no signiﬁcant difference among periods 2 to 4.
* P < 0.05 compared with the vehicle group.

ACh release (% oi period 1)

L

(3)
15c

120

ACh release (96 of period 1)
8

119

 

 

 

 

(A)
160—
. I 1
'8 120* /
’C
8
6 l- ‘ A
95 .
g 80— , * *
E
.C
0 * a *
< u-
40... "‘§" E
’ *vehicle +1o-9M *10'8M 910'7M
41 1 J J
o 1 2 3 4
(B)
160—

 

120*-

 

 

 

 

ACh release (96 of period 1)
8
1

~ +vehicle +1o-9M *10‘8M 010'7M

 

 

0 l l l l
1 2 3 4

Stimulation period

Fig. 5-4. Effect of PGE2 on acetylcholine (ACh) release from cholinergic nerves
innervating dog trachea (A) and bronchi (B). Four periods of electrical ﬁeld stimulation
were applied to four groups (n = 6 in each group) of tissue which had been pretreated
with indomethacin. NO PGE2 was administered during period 1. During periods 2 to
4, one group was treated with PGE2 vehicle, whereas the other three were treated with
different concentrations of PGE,. PGE, inhibited ACh release concentration-dependently
and the magnitude Of inhibition did not change over time. * P < 0.05 compared with
the control group.

(A)

: 8:8 .o as 38.2 eo<

: 8:3 do so noose 5e.

120

(A)

'0' indomethacin 0 control

33‘
l

s
T

 

ACh release (96 of period 1)
8
T I

8
I

 

 

(B)
160 r‘
1" indomethacin 0 control

140—

120—

 

too~

80—

ACh release (% of period 1)

 

60 L l l
1 2 3

Stimulation period

&

Fig. 5-5. Effect of indomethacin on acetylcholine (ACh) release from cholinergic nerves
innervating horse trachea (A) and bronchi (B). Four periods of electrical ﬁeld
stimulation were applied to two groups (n = 5 in each group) of tissue. NO indometha-
cin was administered during period 1. During periods 2 to 4, one group was treated with
indomethacin (3 x 1045 M) whereas the other with its vehicle. Indomethacin had no effect
on ACh release.

121

(A)

"' meclofenamate 9 control

'3‘

.a

N

o
1

100 -

ACh release (96 of period 1)

a)
O
l

 

 

(B)

.a.

a

O
1

» +mecloienamate ‘ control

a

d

.h

0
l

ACh release (96 of period 1)
8 r23
0 O
l I

O
O
l

L 1 r
1 2 3 4
Stimulation period

 

8

Fig. 5-6. Effect of meclofenamate on acetylcholine (ACh) release from cholinergic
nerves innervating horse trachea (A) and bronchi (B). Four periods of electrical ﬁeld
stimulation were applied to two groups (n = 5 in each group) of tissue. NO meclofena-
mate was administered during period 1. During periods 2 to 4, one group was treated
with meclofenamate (10‘ M) whereas the other served as control. Meclofenamate had
no effect on ACh release.

 

 

 

Fig,

ifiner
Slimu
Pierre
in the
withor

A

1)
tro
n
I co
e (% o
s
lea
e
h r
AC

 

 

122

(A)
1 60

l

T

+ PGE2. control

.A
hr
0
T

T

*
120
100»— N

80—

 

ACh release (96 of control)

 

 

60 I l ' 1
control 9 8
(3)

N1-

+ PGE2. control

‘ 91

120—

100—

ACh release (96 of control)

 

60 1 J 1 1
control 9 8 7

P652 concentration (409 M)

Fig. 5-7. Effect of PGE, on acetylcholine (ACh) release from cholinergic nerves
innervating horse trachea (A) and bronchi (B). Four periods of electrical ﬁeld
stimulation were applied to two groups (n = 5 in each group) of tissue that had been
pretreated with indomethacin. N o PGE, was given during period 1. Periods 2 to 4 were
in the presence of increasing concentrations of either PGE, or its vehicle. PGE, was
without effect on ACh release except that 10?7 M PGE, augmented ACh release
signiﬁcantly in the trachea. * P < 0.05 compared with the vehicle group.

Introd

in mar
factors
Sllggeg
1981a
Uncle;
airwa:
that t'

airwa

the e:
“1118c

Ont},

exper

CHAPTER 6
PREJUNCTIONAL az-ADRENOCEPTORS INHIBIT ACETYLCHOLINE

RELEASE FROM CHOLINERGIC NERVES IN EQUINE AIRWAYS‘

Introduction

The parasympathetic nervous system is the predominant excitatory neural pathway
in mammalian airways (Barnes 1986). Its activity can be modulated by a number of
factors through actions on prejunctional receptors. Inhibitory tip-adrenoceptors have been
suggested to be present on airway cholinergic nerves of guinea pigs (Grundstrtim et al.
1981a; Thompson et al. 1990) and humans (GrundstrOm and Andersson 1985). It is yet
unclear if prejunctional az-receptors are present on cholinergic nerves supplying equine
airway smooth muscle. The interest in equine airways was stimulated by the observation
that there is a predominant muscarinic component to airway obstruction in the equine
airway disease known as heaves (Broadstone et al. 1988).

The most common approach for studying prejunctional modulation is to compare
the effect of a potential neuromodulator on the contractile response of airway smooth
muscle to EFS and to an exogenous muscarinic agonist such as ACh. A1 greater effect

on the response to EFS than to ACh suggests that this factor has a prejunctional

 

‘ This study was conducted in conjunction with Mingfu Yu, who performed the
experiments involving muscle tension measurements.

123

modula
contrac
not be

factors
moduli
AChn
studies
indirec
1990) .

direct]

adrenr
comp ;
Conch

aCIiVa

124
modulatory effect. In this indirect approach, ACh release is inferred solely from the

contractile response of smooth muscle. Conclusions about prejunctional modulation may
not be reliable, because inhibitory neurotransmitters released concurrently with ACh or
factors that inhibit the spread of excitation between gap junctions would also cause
modulation of the response to EFS but not to ACh. Although direct measurement of
ACh release can provide dependable evidence about prejunctional modulation, previous
studies on prejunctional az-receptors of airway cholinergic nerves have been mostly
indirect (Grundstrbm et al. 1981a; Grundstrt'im and Andersson 1985; Thompson et al.
1990). In a study in which ACh release from canine airway cholinergic nerves was
directly measured, NE did not inﬂuence ACh release (Martin and Collier 1986).

In the present study, I investigated the presence and function of prejunctional a2-
adrenoceptors on cholinergic nerves innervating horse airway smooth muscle by
comparing the responses to EFS and ACh and by measurement of ACh release. I
concluded that az-adrenoceptors are present on equine airway cholinergic nerves and that

activation of these receptors inhibits cholinergic neurotransmission.

Materials and methods

Sixteen horses (body weight: 381.0 i 25.8 kg, 2 i SE) were used in this study
which was approved by the All-University Committee on Animal Use and Care of
Michigan State University. Horses had no clinical signs of respiratory disease for several
weeks before they were euthanized by injection of an overdose of pentobarbital sodium
through the jugular vein. Post-mortem, their lungs were normal in gross appearance.

A variety of non-puhnonary tissues was utilized by several other investigators. A

segment (
part of ti
(composi
2.6, Mg'
C0,.

1
and was
muscle
dissecte
left inn
Studies

release

MusClr

rnl ﬁg:
1112lint
Bloch».
force

“11ch
gene,

We re

125
segment of trachea between the 6th and 15th cartilaginous rings above the carina and a

part of the caudal lung were removed quickly, immersed in Krebs-Henseleit solution
(composition in mM: NaCl 118.4, NaHCO3 25.0, dextrose 11.7, KC14.7, CaCl2 - 2HZO
2.6, MgSO4 - 7HZO 1.19, KHZPO, 1.16), and gassed continuously with 95% O2 - 5%
C02.

The trachea was opened longitudinally through the cartilages in its anterior aspect
and was pegged ﬂat. Trachealis strips (2 X 30 mm) were cut with a template along the
muscle ﬁber direction. Two bronchial rings (OD = 3-5 mm, length = 15 mm) were
dissected free from the surrounding tissues from each of 5 animals. The epithelium was
left intact in all the preparations. Trachealis strips were used for both muscle tension
studies and ACh release measurement, while bronchial rings were used only in ACh

release measurement.

Muscle tension study

Trachealis strips (2 x 10 mm) were tied with 3-0 silk suture and suspended in 15-
ml tissue baths containing Krebs-Henseleit solution bubbled with 95 % Oz - 5 % CO2 and
maintained at 38°C. One end of each strip was tied to a hook at the lower end of an
electrode holder that was placed in the muscle bath. The other end was attached to a
force transducer (Grass FT03, Grass Instrument Company, Quincy, MA) mounted on a
micromanipulator so that the tissue length could be adjusted. The isometric force
generated by the muscle was recorded on a Grass polygraph (model 7E or 7D). Tissues

were suspended between platinum electrodes for EFS. Electrical pulses were produced

bya

II. D

stirr

solt
mu:

mu:

ind.

COB

pla
cot
14,
One

[1‘ e;

gen

126
by a stimulator (Grass SS8) and passed through a stimulus power booster (Stimu-Splitter

II, Med Lab Instrument, Loveland, CO).

Tracheal strips were equilibrated for approximately 100 min with a predetermined
optimal passive tension of 2 grams applied and maintained. During the equilibration,
EFS (20 V, 16 Hz, 0.5 ms) was applied to tissues for durations of 2—3 min at 8- to 10-
min intervals until the baseline was stable and the magnitude of the response to this
stimulus was consistent. The bath solution was changed every 15 min. At the end of
the equilibration, the contractile response to 127 mM KCl-substituted Krebs-Henseleit
solution was determined. This response was used to normalize the strength of subsequent
muscle contractions. Tissues were then rinsed with Krebs-Henseleit solution until the
muscle tension returned to baseline.

Protocol 1: Eﬁ'ect of az-receptor agonists clonidine and UK 14, 304 on EFS-
induced contraction—Control frequency-response curves (0.1, 0.5, 2, 8, 32 Hz) were
constructed by applying increasing frequency of supramaximal EFS (20 V, 0.5 ms) to
the tissues. Each stimulus lasted approximately 2-3 min until the tension reached a
plateau. The interval between consecutive stimuli was 6 min. Afterwards, three
concentrations of clonidine (107, 10‘, and 105 M) and three concentrations of UK
14,304 (10‘, 107 and 10‘ M) were added to six muscle baths. Each bath received only
one drug at one concentration. One strip from each animal did not receive drug
treatment and served as time control. After a 30-min incubation period for clonidine or
a 10-min incubation period for UK 14,304, the second frequency-response curve was

generated.

respor
conce
Acety
10‘3 IN
a plat
tion-r

Hens

or U
had i
M or

freqt

az‘l‘l
Wen
PIUS
freq

effe

0rd
rele

and

127
Protocol 2: Eﬂect of clonidine (10’ M) and UK 14, 304 (10‘ M) on the contractile

response to exogenous ACh—This protocol was similar to protocol 1 except that ACh
concentration-response curves were constructed instead of frequency-response curves.
Acetylcholine was added to the muscle baths cumulatively at concentrations from 10‘ to
10'3 M in log increments. When the response to one concentration of ACh had reached
a plateau, the next concentration of ACh was added. After the control ACh concentra-
tion-response curves had been obtained, tissues were washed with warmed Krebs-
Henseleit solution until the tension retumed to baseline.

Protocol 3: Eﬁ’ects of a,— or az-receptor antagonists on the actions of clonidine
or UK 14, 304—When the frequency-response curves before and after clonidine (10‘ M)
had been obtained (as described in protocol 1), an biz-receptor antagonist yohimbine (10'7
M or 10‘ M) was added to the tissue baths. After a 30-min incubation period, the third
frequency-response curves were generated.

To conﬁrm that the effects of clonidine and UK 14,304 were mediated through
az-receptors, the highly selective az-antagonist idazoxan and the a,-antagonist prazosin
were used. Tissues were ﬁrst incubated with idazoxan for 20 min and then with idazoxan
plus either clonidine (10‘ M) or UK 14,304 (10‘7 M) for another 10 min. EFS
frequency-response curves were constructed before and after the drug treatment. The
effect of prazosin on the clonidine response was Studied with a similar protocol.

Protocol 4: Eﬁect of yohimbine on EFS-induced smooth muscle contraction—In
order to determine if prejunctional az-receptors are normally activated by endogenous NE
released from adrenergic nerves, EFS frequency-response curves were obtained before

and after incubation of the tissues with yohimbine (107 and 10‘ M). To determine if any

effect
nonsp

guane

CODII’:

Meas

3-0 s
2 brc
prep;
Morn
main
Wall
glass
steel

Unaf

bath
incu
of r
Chol
inCu'

atro;

128

effect of yohimbine was due to inhibition of activation of prejunctional org-receptors or
nonspeciﬁc effect, the effect of yohimbine was also studied in the presence of
guanethidine (105 M). This concentration of guanethidine abolishes the EFS-induced

contraction of equine digital artery.

Measurement of ACh release induced by EF S

Four trachealis strips (each measuring 2 x 15 mm) were tied into a bundle with
3-0 silk suture. Four such trachealis bundles (weight: 221.3 :1: 13.6 mg, X :l: SE) and
2 bronchial rings (231.1 :1: 19.9 mg, i :1; SE) were prepared from each animal. Each
preparation was suspended in a 2-ml tissue bath (Radnoti Glass Technology Inc. ,
Monrovia, CA) containing Krebs-Henseleit solution bubbled with 95 % 02 - 5 % C02 and
maintained at 38°C. A pair of parallel platinum wire electrodes was built against the
wall of the muscle bath in the vertical direction. One end of the tissues was tied to a
glass tissue holder that was placed in the muscle bath; the other end was attached to a
steel bar above the bath by an elastic thread. As indicated in Chapter 2, ACh release is
unaffected by muscle preload alterations.

Tissues were equilibrated for approximately 2 hours, during which time the
bathing solution was changed every 10-15 min. After equilibration, tissues were
incubated with 10" M neostigmine (a cholinesterase inhibitor to prevent the breakdown
of released ACh), 107 M atropine (to block the muscarinic autoreceptors on the
cholinergic nerves), and 105 M guanethidine for 30 min. Two bundles of trachealis were
incubated with 107 M and 10" M yohimbine, respectively, in addition to neostigmine,

atropine, and guanethidine. Electrical ﬁeld stimulation (20 V, 0.5 ms, 0.5 Hz for

tracht
perim
10‘5 l
yohir
applil
expel
and 5
each
blottl

relea

Was
Same

Cha]

Sulfa
Yohj
(Res
wen
of t}

Wale

129

trachea; 20 V, 1 ms, 2 Hz for bronchi) was applied to all the tissues for four 15-min
periods. After the ﬁrst EFS period, increasing concentrations of clonidine (107, 10‘, and
105 M) were administered to four tissue baths, including the two trachealis bundles with
yohimbine and one without, and one bronchial segment. Electrical stimulation was
applied after a 30—min incubation with each concentration of clonidine. In each
experiment, one bundle of trachealis and one bronchial segment did not receive clonidine
and served as time control. Tissue bath solution was collected upon the completion of
each EFS for the measurement of ACh. At the end of the experiment, tissues were
blotted dry and weighed. The effect of UK 14,304 alone or plus idazoxan on ACh
release was studied in a similar way in the trachealis.

High-performance liquid chromatography coupled with electrochemical detection
was used to measure the amount of ACh in the tissue bath liquid. The technique was the
same as described in Chapter 2 except for some modiﬁcations, which were addressed in

Chapter 5.

Drugs

The following drugs were used in the present study: ACh chloride, atropine
sulfate, clonidine hydrochloride, neostigmine methylsulfate, prazosin hydrochloride,
yohimbine hydrochloride (all from Sigma Chemical Co. , St Louis, MO) and UK 14,304
(Research Biochemical International, Natick, MA). All the drugs except UK 14,304
were dissolved in deionized water and diluted with Krebs-Henseleit solution on the day
of the experiment. UK 14,304 was dissolved in ethanol (10 mg/ml) and then diluted with

water to 102 M. This solution was stored at 4°C as aliquots and diluted with

10‘]
contr
baths

ﬁnal

Snui

as a
sanu
and
to e
adrn
depe
inhil
0fci
the!
POSI

On;

130

Krebs-Henseleit solution on the day of the experiment. The concentration of ethanol in
10" M UK 14,304 is 0.0029%, which had no inﬂuence on EFS-induced smooth muscle
contraction as determined in a previous study. Drug solution was pipetted into the tissue
baths at 1% of the bath volume. The concentrations of the drugs were expressed as their

ﬁnal bath molar concentrations.

Statistical analysis

The contractile responses of the muscle strips to EF S and ACh were expressed
as a percentage of the tension produced by 127 mM KCl. The ACh amount in the
samples was expressed as a percentage of that measured in the ﬁrst sample of each bath
and also as pmol/ g/ min. A paired t-test was used to compare the contractile responses
to each frequency of EF S or each concentration of ACh before and after the drug
administration. To determine if the effect of clonidine was frequency- and concentration-
dependent, the effect of clonidine on EFS-induced contraction was also calculated as %
inhibition and analyzed by two-way analysis of AN OVA. The disinhibition of the effect
of clonidine by yohimbine, and the ACh measurement data in trachea were analyzed by
the use of single factor repeated and randomized design AN OVA, respectively. Tukey’s
post-Hoe test was used to identify signiﬁcantly different means. The effect of clonidine
on ACh release in bronchial rings was tested by unpaired t-test.

All values were expressed as K :1; SE. P < 0.05 was considered statistically

signiﬁcant. 11 equals the number of animals.

Raul

Muse.

induc

(Fig
anta;
not I
EFS
prod
idaz
com
infll
incr
y 0b
irre

10‘:

AC

on

131
Results

Muscle tension study

Clonidine (107-10s M) had no effect on passive tension but inhibited the EFS-
induced smooth muscle contraction in a concentration- and frequency-dependent manner.
The inhibition was absent at 32 Hz but increased in magnitude as frequency decreased
(Fig. 6-1). Inhibition induced by clonidine (10‘ M) was attenuated by the az-receptor
antagonists yohimbine (107 and 10‘ M, Fig. 6-2) and idazoxan (10‘ M, Fig. 6-3), but
not by the al-receptor antagonist prazosin (10‘ M, Fig. 6—3). UK 14,304 inhibited the
EFS-induced smooth muscle contraction at 10'7 and 10‘ M (Fig. 6-4). The inhibition
produced by 10'7 M UK 14,304 was prevented by pretreating the tissues with 107 M
idazoxan (Fig. 6-4). Neither clonidine (105 M) nor UK 14,304 (10‘ M) affected the
contractile response to exogenous ACh (data not shown). Yohimbine at 10” M had no
inﬂuence on the response to EFS (data not shown), but yohimbine at 10‘ M signiﬁcantly
increased the response to 0.1 Hz (Fig. 6-5). Although a statistically signiﬁcant effect of
yohimbine was also observed at 32 Hz, the effect was trivial and probably biologically
irrelevant. The effect of 10‘ M yohimbine was unaffected by pretreating the tissues with

10‘ M guanethidine.

ACh measurement

During the ﬁrst stimulation period, the ACh release rate of trachealis strips was
7.23 :1; 2.10 pmol/g/min. Incubation with yohimbine or idazoxan alone had no inﬂuence
on ACh release. In trachealis strips, UK 14,304 (10‘-10‘ M) inhibited ACh release

concentration-dependently and this inhibition was attenuated by pretreating the tissues

(I!

132
with idazoxan (Fig. 6-6A). Similarly, clonidine (107 to 10‘ M) decreased ACh release

concentration-dependently and this inhibition was antagonized by pretreating the tissues
with yohimbine (10‘7 and 10‘ M, Fig. 6B and C). Clonidine (10'7—10‘ M) also
signiﬁcantly inhibited ACh release in bronchi (Fig. 6—7). All of the above-mentioned
results were obtained at a single stimulation frequency (0.5 Hz for trachea, 2 Hz for
bronchi) and voltage (20 V). The clonidine—induced inhibition on ACh release was also
studied at several frequencies (0.5, 2, 8, 32 Hz) and voltages (10, 15, 20, 24 V) in
trachealis. Clonidine signiﬁcantly inhibited ACh release at 0.5 and 2 Hz but not at 8 and
32 Hz (20 V, 0.5 ms, Fig. 8A). The inhibition observed at 0.5 Hz was present at all

voltages (Fig. 6-8B).

Discussion

The role of az-adrenoceptors on equine airway cholinergic neurotransmission was
examined in the present study. In the indirect experiment, az-receptor agonists clonidine
and UK 14,304 inhibited the contractile response of trachealis to EFS but did not affect
the response to ACh. The lack of an effect of az-agonists on ACh-induced contractions
suggests either that there are no az-receptors on smooth muscle cells or that their
activation does not interfere with the muscarinic response of smooth muscle. Therefore,
the inhibition of EFS-induced contraction by clonidine or UK 14,304 could be explained
by 1) a decreased ACh release from cholinergic nerves, 2) an increased release of
neurotransmitters from inhibitory nerves; or 3) other possible effects of clonidine or UK
14, 304, such as modulation of the spread of excitation through gap junctions. It is

difﬁcult to determine which possibility is true simply through the measurement of muscle

 

tensic
ACh

induc

an a1
(Ken;
the c]
to E3
antag
at th.
Yohn
relea
Effec
neun
ICCCI
az‘re
FESpc
a hig
We c

aCtiv

theF

With

133

tension. However, direct measurements indicated that clonidine and UK 14,304 inhibited
ACh release signiﬁcantly. Thus the inhibition by clonidine or UK 14,304 of EFS-
induced contractions must be a result of decreased ACh release from cholinergic nerves.

Clonidine is a selective az-receptor agonist (Doxey et al. 1985) and yohimbine is
an antagonist speciﬁc for az-receptors at concentrations equal to or below 107 M
(Kenakin 1987). In the present study, yohimbine concentration-dependently attenuated
the clonidine-induced inhibition of both the contractions and the ACh release in response
to EF S, and this effect of yohimbine was apparent at 107 M concentration. The
antagonism by yohimbine of the action of clonidine is most likely a result of competition
at the receptor level (genuine antagonism) rather than functional antagonism, because
yohimbine per se had little or no effect on either EFS-induced contractions or ACh
release. Furthermore, prazosin, a selective ail-antagonist, did not prevent clonidine’s
effect. These observations suggest that the inhibitory effect of clonidine on cholinergic
neurotransmission and its reversal by yohimbine were through their actions on 0:;-
receptors. This conclusion was conﬁrmed by the observation that the highly selective
az-receptor agonist, UK 14,304, concentration-dependently inhibited both the contractile
response and ACh release induced by EFS. This inhibition was antagonized by idazoxan,
a highly speciﬁc az-antagonist (Doxey et a1. 1985). Based on the above observations,
we conclude that az-receptors are present on equine airway cholinergic nerves and the
activation of these receptors inhibits cholinergic neurotransmission.

Isolated trachealis strips may contain both the preganglionic nerve terminals and
the postganglionic neurons of the parasympathetic nerves. Because ganglionic blockade

with hexamethonium has no effect on either muscle contractions (Chapter 4) or ACh

II

134
release in response to EFS (Chapter 2), the prejunctional az-receptors that inhibited ACh

release must be located on postganglionic neurons. However, the possibility cannot be
excluded that az-adrenoceptors may also exist on parasympathetic ganglia (Baker et al.
1983).

Based on the dominance of the parasympathetic nerves in control of airway tone
and the huge depressant effect of clonidine and UK 14,304 on cholinergic neurotransmis-
sion, it seems likely that prejunctional az-adrenoceptors constitute a potent regulatory
mechanism of smooth muscle tone in equine airways. The physiological stimulants of
these receptors could be N E released by sympathetic nerves and/ or circulating
catecholamines. The observation that sympathetic and parasympathetic nerves are in
close association in guinea pig trachea (Jones et al. 1980) furnishes the anatomical
possibility for interaction between these two nervous systems. Electrical ﬁeld stimulation
of isolated trachealis strips activates all the nerves in the preparation. If NE released
from the sympathetic nerves in response to EFS can activate az-receptors on cholinergic
nerves, blockade of the az-receptors with yohimbine should enhance ACh release and the
contractile response to EFS. In the present study, EFS-induced contraction was
augmented by yohimbine (10‘ M) but only at a frequency of 0.1 Hz. It is unlikely that
this augmentation was a result of blockade of az-receptors, because depletion of the
sympathetic neurotransmitter with guanethidine did not abolish the augmentation. These
results suggest that, under the present experimental conditions, endogenous NE does not
inhibit ACh release from equine airway cholinergic nerves via prejunctional az-receptors.
This conclusion is not unexpected considering the sparse sympathetic innervation in most

species and the rapid removal by neuronal and extraneuronal uptake or enzymatic

135
breakdown of NE (Barnes 1986). Thompson et al. (1990) reached a similar conclusion

in experiments with guinea pig tracheal tissues.

Circulating catecholamines may also stimulate az-receptors on cholinergic nerves.
In the guinea pig airways, epinephrine and NE, in the presence of B-blockers, inhibit
airway smooth muscle contraction induced by nerve stimulation but not by exogenous
ACh (Grundstrom et al. 19813) suggesting prejunctional modulation by catecholamines.
However, it is not known if circulating catecholamines have such an effect at their
physiological concentrations. The concentration of epinephrine and NE in equine
plasma are 9 x 101° M and 7 x 101° M respectively at rest and increase to 1.53 x 107 M
and 1.48 x 107 M respectively after exercise (Snow et al. 1992). Whether these
concentrations of catecholamines, especially after exercise, can modify equine airway
cholinergic activity has not been determined in vitro. Failure of intravenous
yohimbine to alter pulmonary function in ponies implies that the az-inhibitory receptors
are not activated under physiological conditions in the resting pony (Broadstone et al.
1992).

The existence of az-receptors on cholinergic nerves may have some clinical
importance in the treatment of airway obstruction. Xylazine, an az-receptor agonist and
potent tranquilizer that inhibits the EFS-induced contractions in equine peripheral airways
(LeBlanc et al. 1993), causes bronchodilation in heavey ponies and this effect is reduced
by yohimbine (Broadstone 1992). Although az-receptors also occur on the airway
smooth muscle cells of some species (Barnes et al. 1982; Takaganagi et al. 1990), their

activation causes smooth muscle contraction. Therefore, if xylazine-induced

136
bronchodilation is mediated at the level of the airway rather than centrally, the most

likely mechanism is its action on prejunctional az-receptors.

137

 

 

 

 

 

 

°/o inhibition
120 r
_ 0 0.1Hz +0.5Hz '0' 2H2 “A“ 8H2 *32Hz
90 —
60 -
b 0
3O — o
b o
O 5;, -
I l 1 J
0 0.1 1 10

Clonidine concentration (uM)

Fig. 6-1. Inhibitory effect of clonidine on the contractile response of equine trachealis
to electrical ﬁeld stimulation (EFS, 20 V, 0.5 ms). Percent inhibition was calculated
from the amplitude of muscle contraction as [(pre-drug - post-drug)/pre-drug] x 100.

The inhibitory effect of clonidine was concentration— (P = 0.0012) and frequency—
dependent (P = 0.0000).

138

Force (% of response to 127 mM KCI)

 

 

160 -
/ *
‘ O
120 —
+/
_ q *
*
80 _ +/+
*
+ +*
40 —
A
O
O * T I I 1
0.1 0.5 2 8 32

Frequency (Hz)

Fig. 6—2. Frequency-response curves of equine trachealis strips to electrical ﬁeld
stimulation (20 V, 0.5 ms). 0 Control; [I] 10‘ M clonidine, n = 6; O 10‘ M clonidine
plus 107 M yohimbine, n = 5; A 10‘ M clonidine plus 10‘ M yohimbine, n = 4. The
inhibition of EFS-induced smooth muscle contraction by clonidine was concentration-

dependently attenuated by yohimbine. * Signiﬁcantly different from control, + signiﬁ-
cantly different from the 10‘ M clonidine.

139

Force (% of response to 127 mM KCI)

 

 

160-
2
120~ o/.*
_ .
80- o *
+ /u
i *
o
40— + I
_ * *
O . * I l I J
01 05 2 8 32

Frequency (Hz)

Fig. 6-3. Frequency-response curves of equine trachealis strips to electrical ﬁeld
stimulation (20 V, 0.5 ms). 0 control, 0 10‘ M clonidine alone, n = 6; El 10‘ M
prazosin plus 10‘ M clonidine, n = 4; A 10‘ M idazoxan plus 10‘ M clonidine, n = 3.
The clonidine-induced inhibition was prevented by pretreating the tissues with idazoxan

but not with prazosin. * Signiﬁcantly different from control, + signiﬁcantly different
from the 10‘ M clonidine.

'1

140

Force (°/o of response to 127 mM KCI)

 

 

 

 

 

160 —
O
120 ~—
80 L °
,— + *
0
40 —
b + I
* * *
0 a 3 l l I
0.1 0.5 2 8 32

Frequency (Hz)

Fig. 6-4. Frequency-response curves of equine trachealis strips to electrical ﬁeld
stimulation (20 V, 0.5 ms). 0 control; B 10‘ M UK 14,304 , n = 4; O 107 M UK
14,304, 11 = 3; A 10‘ M UK 14,304, 11 = 4; v 107 M idazoxan added 20 min before
the addition of 10'7 M UK 14,304, n = 3. * Signiﬁcantly different from control,
+ signiﬁcantly different from the 10’7 M UK 14,304.

 

 

141

Force (% of response to 127 mM KCI)

 

160 —
0control
_ *1 uM yohimbine *
n=5
120 r
80 —
40 —
O l l 1 I J

 

 

0.1 0.5 2 8 32
Frequency (Hz)

Fig. 6-5. Effect of yohimbine (10‘ M) on the contractile response of equine trachealis
strips to electrical ﬁeld stimulation (EFS, 20 V, 0.5 ms) in the absence of guanethidine.
Yohimbine increased the response to 0.1 Hz and this increase was not affected by

pretreating the tissues with guanethidine (105 M; data not shown). * Signiﬁcant effect
of yohimbine.

 

142

A ACh relese (% ol pre-UK 14.304)

   

 

 

 

 

 

 

 

 

 

ACh release (96 of pre-clonidine)

   

 

 

 

 

 

 

 

 

 

 

140[
120 - 120 _
100 ~ 100 F
so » 80 ~ * *
50 I. +* 60 _
Ocontrol Oeontrol **
40 _ oux 14.304 40 _*clonidino
OUK 14,304 +100 nM idazoxan 9clonidino +100 nM yohimbine
20 _ tux 14,304 +1 uM idazoxan * 20 -*clonidine +1 or yohimbim *
”=4 [1:4
0 ‘ l l l 0 e-ckl) 'd' 1001 M 1 lM iOIIJM
pre-UK14.304 10nM 100 nM IuM ” "' '"° " “
UK 14.304 concentration Clonidine concentration
C ACh release rate(pmol/g/min)
1 2 F
10 - T
8 I— J-
2 — _L
control clonidine 10'7 M 10" M
(104 M) g l
yohimbine
+
clonidine(10‘ M)

Fig. 6-6. Effects of az-receptor agonists and antagonists on acetylcholine (ACh) release
from equine trachealis strips in response to electrical ﬁeld stimulation (EFS, 20 V, 0.5
ms, 0.5 Hz). * Signiﬁcantly different from time control, + signiﬁcantly different from
UK 14,304 (A) or clonidine (B). A shows the inhibition of ACh release by UK 14,304
and the antagonizing effect of idazoxan. B shows the inhibition of ACh release by
clonidine and the antagonizing effect of yohimbine. C gives the ACh release rate
expressed as pmol/g/min at a clonidine concentration of 10‘ M.

 

 

143

ACh release (% of pre-clonidine)

140*

120—

100—

80—

60*—

40*—

20—

l—

 

 

0 control + clonidine

I

(n=5)
I I I

pre-clonidine 7 6 5

Clonidine concentration (-log M)

Fig. 6-7. Effect of clonidine on acetylcholine release from equine bronchial rings in
response to electrical ﬁeld stimulation (EFS, 20 V, 1 ms, 2 Hz). * Signiﬁcantly
different from time control.

 

 

ﬂow
in re
mem
Sign}

 

144

A ACh release (96 of control)

 

 

 

 

 

 

 

 

 

 

 

 

 

200} Dtime com: Ecronrdsm (10;— M) I /
‘60— fl /
I 2 2
.01: * /

 

 

 

 

 

0.5 2
Frequency (Hz)

ACh release (96 of control)

 

 

 

 

1501
Deontrol chonidine (10‘M)
n = 5 I
T
.1...
100 "
.L

 

 

50-

 

 

 

 

 

 

 

 

 

 

 

 

 

B\\\\\\\s*
‘\\\\\\\\-I»

F—l
.. \\\\\ﬂ »

1O 1 2

Voltage (V)

Fig. 6-8. Effect of frequency (A; 20 V, 0.5 ms) and voltage (B; 0.5 Hz, 0.5 ms) on
clonidine-induced inhibition of acetylcholine (ACh) release from equine trachealis strips
in response to electrical ﬁeld stimulation. Data were normalized to an initial measure-
ment of ACh release (2 Hz, 20 V, 0.5 ms) obtained prior to addition of drug. *
Signiﬁcant difference between time control and clonidine.

 

Intror

hay a1
"heav
resist:
muse:
trachr
hype:
Sllgg.
heavI
Obstr
rCSpc

Incas

 

CHAPTER 7
ACH RELEASE FROM AIRWAY CHOLINERGIC NERVES IN

HORSES WITH AIRWAY OBSTRUCTION (HEAVEs)

Introduction

Some horses and ponies deve10p airway obstruction when stabled and fed dusty
hay and enter remission at pasture. This type of airway obstruction is colloquially called
"heaves. " Muscarinic antagonists such as atropine markedly reduce the increased airway
resistance in affected animals (Broadstone et al. 1988), indicating the involvement of a
muscarinic mechanism. In in vitro studies, airway smooth muscle preparations from
trachea and bronchi of heavey horses are hyporesponsive to exogenous ACh but
hyperresponsive to EFS compared with those of normal horses (Broadstone et al. 1991),
suggesting that ACh release from airway cholinergic nerves is probably increased in
heavey horses. An increase in ACh release may contribute to the development of airway
obstruction. Because this conclusion has been inferred by comparing the contractile
response of airway smooth muscle to EFS and ACh, it needs to be veriﬁed by direct
measurement of ACh release.

In early chapters of my dissertation, I reported that ACh release from horse
airway cholinergic nerves is inhibited by activation of prejunctional muscarinic

autoreceptors (Chapters 2 and 3) and az-adrenoceptors (Chapter 6). Alterations in the

145

146

function of these receptors might result in increased ACh release. There is evidence to
suggest that muscarinic autoreceptors on airway cholinergic nerves are probably
dysftmctional in the lungs of asthmatics (Minette et al. 1989) and of guinea pigs infected
with parainﬂuenza vinrs (Fryer et al. 1990) or challenged by ovalbumin (Fryer and
Wills-Karp 1991). It is possible that other prejunctional receptors such as the a2-
adrenoceptors may be also affected in diseases characterized by airway inﬂammation and
obstruction.

Prostaglandin E, inhibits ACh release from canine airway cholinergic nerves
(Deckers et al. 1989a). In the horse, however, PGE2 lacks such an effect in both the
trachea and bronchi (Chapter 5). It was of interest to know if horses with heaves
respond differently to PGE2.

The present study was conducted to answer the following questions: 1) Is ACh
release from airway cholinergic nerves of the heavey horse increased under the in vitro
conditions? 2) Are muscarinic autoreceptors or prejunctional rah-adrenoceptors on airway
cholinergic nerves dysfunctional in the heavey horse? 3) Does PGFq modulate ACh

release from airway cholinergic nerves in heavey horse?

Materials and methods
Animals

Airway tissues were obtained from nine normal (weighing 387.4 i 24.0 kg) and
ﬁve heavey (weighing 406.2 :1; 29.8 kg) adult horses of either sex. The normal horses
had no clinical signs of pulmonary disease and, post mortem, their lungs were normal

in gross appearance. The heavey horses had a history of heaves. Several days before

the exp,
were fe ..
such as
presen
Results
RL of
:1: 0.1
on ave
Cay, o
compo
mg/kg
Cd,“ cl
and 0
admll

trial .

Prep

S0diI
Wide
Segm

lechn

147

the experiment, they were transferred from pasture to straw-bedded stalls where they
were fed dusty hay. They remained in the stall until clinical signs of airway obstruction,
such as ﬂared nostrils and forced abdominal effort during expiration, appeared. The
presence of airway obstruction was conﬁrmed by pulmonary function measurements.
Results of previous studies indicate that normal horses in the barn environment have a
RL of 0.87 :1: 0.10 cm HZO/L/sec, a APpl,,m of 6.7 :1; 0.4 cm H20, and a C.lyn of 1.26
:1: 0.15 L/cm H20 (Gray et al. 1992a). The heavey horses used in the present study had
on average a RL of 2.6 :1; 0.3 cm H,O/L/sec, a APplm of 36.6 :1; 6.2 cm H20, and a
Cdyn of 0.35 i 0.08 L/cm H20 during airway obstruction. The strong muscarinic
component to airway obstruction was conﬁrmed by administration of atropine (0.2
mg/kg), which dilated the airways in all the animals. Pulmonary resistance, APplm, and
CM changed to 1.2 3; 0.3 cm HZO/L/sec (P < 0.05), 15.6 :1; 2.9 cm H20 (P < 0.05),
and 0.74 :1; 0.18 L/cm H20 (P > 0.05), respectively, within 5 to 15 min of atropine
administration. In vitro experiments were conducted at least 3 days after the atropine

trial.

Preparation of tissues

Animals were euthanized by intravenous injection of an overdose of pentobarbital
sodium. Trachealis strip bundles with epithelium (each consisting of four strips of 2 mm
wide and 17— 19 mm long) were prepared from the dorsal membrane of the tracheal
segment between the 16th and 25th cartilaginous rings above the carina using the
technique described in Chapter 2. Bronchial segments (3-5 mm outside diameter, ~15

mm long) with epithelium were prepared from the left or right lung. The tissue

 

Prel

M01
the‘
was

to a

the 1
relea
guid.
bronI
hook
(Com
2.6 I

95%

Protc

SOlut:
HECeS
bath,

148
preparations were then suspended in 2-ml tissue baths (Radnoti Glass Technology, Inc. ,

Monrovia, CA). Each bath had a pair of parallel platinum wire electrodes built against
the wall in the vertical direction. For trachealis strip bundles, one end of the preparation
was ﬁxed to the tip of a glass tissue holder. The suture on the other end was attached
to a steel bar through a piece of elastic. The tension applied to the tissue was well within
the range of 2-20 grams, which has been determined to have no inﬂuence on ACh
release from airway cholinergic nerves (Chapter 2). The bronchial tubes were gently
guided onto the shorter arm of an "L”-shaped glass tissue holder. To prevent the
bronchial tube from ﬂoating above the surface of the bath solution, it was tied to the
hook at the end of the tissue holder. Tissue baths contained Krebs-Henseleit solution
(composition in mM: NaCl 118.4, NaHCO, 25.0, dextrose 11.7, KCl 4.7, CaC12-2H20
2.6, MgSO4-7H20 1.19, KH2P04 1.16), which was maintained at 38°C and bubbled with

95% 045% co,

Protocols

Tissues were equilibrated in the baths for 2-3 hours, during which time the bath
solution was replaced at 10. to 20-min intervals. This extended equilibration period was
necessary in order to reduce the amount of foam formed during bubbling. In a 2-ml
bath, foam formation results in overﬂow of bath solution so that ACh release cannot be
accurately determined.

In order to determine if muscarinic autoreceptors are defective in the heavey
horse, one trachealis strip bundle and one bronchial segment were prepared from each

control or heavey horse. After incubation with 10 uM neostigmine for 60 min, 5 periods

of

ﬁr:
tis:
atrI

inc

pre.
pre;
pre;
atrc

COD

Upr

Inez
incl
inﬁ
rest
SUIT
C01]
Epir

me.

149
of 20-min EFS (20 V, 4 Hz, 2 ms) were applied to the tissues at 30-min intervals. The

ﬁrst period was applied in the absence of atropine. During the remaining 4 periods,
tissues were stimulated in the presence of logarithmically increasing concentrations of
atropine (10‘ ~ 105 M) in addition to 10 uM neostigmine. EFS was applied after
incubation with each concentration of atropine for approximately 25 min. To make sure
that the effect of atropine was not a time-related response, one bronchial segment was
prepared from each normal and heavey horse and one trachealis strip bundle was also
prepared from each heavey horse. Except for the omission of atropine, these tissue
preparations otherwise received the same treatment as those used to study the effect of
atropine. A time control experiment was not conducted for trachealis tissues of the
control horse because I have previously demonstrated that ACh release is constant over
up to 6 EFS periods in the absence of atropine (Chapter 3).

In trachealis tissues used to examine the effect of atropine, the ACh release data
measured during the ﬁrst EFS period were also used to determine if ACh release is
increased in the heavey horse. In an effort to eliminate the possible inﬂuences of edema,
inﬁltration of inﬂammatory cells, or muscular hypertrophy on the interpretation of our
results, ACh release was expressed in units of pmol/g tissue dry weight/min, pmol/mm2
surface area of the trachealis/min, as well as pmol/ g tissue wet weight/min prior to
comparisons between the control and heavey horses. For the purpose of calculating the
epithelial surface area of the trachealis, the length of each trachealis strip bundle was
measured carefully. The surface area of each bundle was calculated as the product of

the length of the bundle, the width of each strip, and the number of strips in each bundle.

150

To determine if prejunctional cab-adrenoceptors are dysfunctional in the airways
of the heavey horse, one pair of trachealis strip bundles and one pair of bronchial
segments were prepared from each heavey horse. After incubation with neostigmine ( 1
nM), atropine (0.1 nM), and guanethidine (10 nM) for 60 min, 4 periods of 15-min EFS
(20 V, 0.5 Hz, 0.5 ms for trachea; 20 V, 2 Hz, 1 ms for bronchi) were applied to the
tissues at 30-min intervals. The ﬁrst period was applied in the absence of clonidine.
During the remaining 3 periods, one preparation of each pair was stimulated in the
presence of logarithmically increasing concentrations of clonidine (107-10‘ M) in
addition to neostigmine, atropine, and guanethidine, whereas the other preparation did
not receive clonidine and served as time control. The tissues were incubated with each
concentration of clonidine for approximately 25 min before stimulation. Historical data
from normal horses (Chapter 6) were cited for comparing the effect of clonidine between
heavey and normal animals.

Four bronchial segments were prepared from each heavey horse to determine the
effect of exogenous PGE2 (109-10‘7 M) and the effect of indomethacin (3 x 10‘ M), a
cyclooxygenase inhibitor, on ACh release. Experiments were not performed with
trachealis tissues because there were only four tissue baths available for this part of the
study , and histological lesions are more prominent in the bronchi than trachea of the
heavey horse (Breeze 197 9). The protocols were the same as those used for the
bronchial segments of the normal horse (Chapter 5). Brieﬂy, four 12-min periods of
EFS were applied to the tissues. Period 1 was applied in the absence of PGE2 and
indomethacin. During the remaining three periods, PGE2, PGE2 vehicle, indomethacin,

and indomethacin vehicle were added separately to the four baths. The concentrations

“

151
of indomethacin (3 x 10‘ M) and its vehicle were constant during the three periods,

whereas the concentrations of PGE, (109-10'7 M) and its vehicle were increased
logarithmically over successive periods. For details of the protocols, please refer to
Chapter 5.

Tissue bath liquid was collected after EFS for ACh measurement. The bath
solution was replaced before the start and at the end of each EFS period. Each time the
bath solution was replaced, the tissue bath was ﬂushed 2-3 times with Krebs-Henseleit
solution.

The wet weight of each tissue preparation was determined at the end of the
experiment after blotting with ﬁlter paper. For tissues used to compare ACh release rate
between normal and heavey horses, the dry weight was also determined after keeping the
tissues in a desiccator containing anhydrous CaCl2 for at least one week, which was

adequate for the tissue to dry thoroughly.

ACh analysis

Collected samples were ﬁltered through 0.2 1‘ nylon membrane ﬁlters (Acrodiscs
13, Gelman Sciences, Ann Arbor, MI). The ACh content was determined by HPLC
coupled with electrochemical detection. The technique was described in Chapter 2. The
following modiﬁcations were made in this study. 1) Some pieces of equipment were
replaced: electrochemical detector (Coulochem II, ESA Inc. , Bedford, MA); ACh and
choline analytical column (ESA Inc.); autosampler (ISS 200, Perkin-Elmer, Norwalk,
CT). 2) The mobile phase (water solution) contained 100 mM NaQI-IPO4, 0.5 mM

tetramethylammonium chloride, 2 mM l-octanesulfonic acid, and 0.005 % reagent MB

152
(BSA) with pH adjusted to 8.0 with H3PO4. At a ﬂow rate of 0.35 ml/min, the retention

times for choline and ACh were 3.9 and 6.2 min, respectively. 3) 25 pl of calibration
standard or bath solution was injected. The standard was injected every 4 samples and
the ACh amount in the samples was calculated based on the bracketed calibrations. The
lower limit of determination was 50 femtomoles. 4) To prevent the breakdown of ACh
molecules at room temperature as observed previously (Chapter 2), samples were kept
refrigerated (1-3 °C) in the autosampler by a refrigerated bath circulator (RTE—110D,
NESLAB Instruments, Inc. , Portsmouth, NH) during the analysis, which lasted for less

than 12 hours.

Drugs

N eostigmine methylsulfate, atropine sulfate, clonidine hydrochloride, and
guanethidine monosulfate (Sigma Chemical Co. , St. Louis, M0.) were dissolved in and
diluted with water. PGEz (Cayman Chemical Company, Ann Arbor, MI) was ﬁrst
dissolved with methanol (1 mg/ml) and subsequently diluted to 105 M with distilled
water. The 105 M PGE, was stored as a stock solution at -15 °C and diluted with water
before use. Indomethacin (Sigma) was dissolved in solutions of equimolar sodium
carbonate. The concentrations of all drugs are expressed as their ﬁnal bath concentra-

tions.

Statistical analysis
To examine the effect of clonidine, PGE2, and indomethacin, ACh release in each

bath during the ﬁrst EFS period was regarded as 100% and that during subsequent

3 $.47.

153

periods was expressed as a percentage of this value. The ACh release during each period
was compared between the control and drug-treated tissues by an unpaired t-test. To
determine if ACh release rate differed in trachealis tissues of the control and heavey
horse, the release data were compared between the two groups by an unpaired t-test. To
investigate if muscarinic autoreceptors are dysfunctional in the heavey horse, ACh release
during the ﬁrst EFS period was regarded as l, and that during subsequent periods was
calculated as multiples of 1. The calculated values were compared between the heavey
and control horses by an unpaired t-test. All values were expressed as mean :1; SE.
P < 0.05 was considered statistically signiﬁcant. Except when otherwise speciﬁed, 11

equals the number of animals.

Results

All samples contained a measurable amount of ACh. ACh content remained
constant during the period when samples were kept in the autosampler. The content
measured after 12 hours varied from the original value by no more than -5 % or +1 %
(average -1.23%, n = 13).

In the presence of neostigmine alone, the rate of EFS-induced ACh release from
the tracheal preparations was the same in the control and heavey horses. The values
were 11.4 i 2.0 versus 14.1 j: 3.0 pmol/g wet weight/min, 53.0 i 8.7 versus 65.4 :1;
13.9 pmol/g dry weight/min, and 1.38 i 0.22 versus 1.68 :1: 0.32 pmol/cm2 epithelial

surface area of trachealis/min for tissues of normal (n = 5) and heavey (n = 5) horses,

respectively.

154
Atropine augmented ACh release in both the trachea and bronchi. Chromato-

grams of a few samples collected from a single tissue bath showing the effect of atropine
on ACh release are displayed in Fig. 7-1. The magnitude of augmentation was similar
between heavey and control horses (Fig. 7-2). ACh release rate was constant over the
ﬁve stimulation periods in the absence of atrOpine (data not shown).

In the heavey horse trachea, clonidine did not inhibit ACh release signiﬁcantly
at 107 M (Fig. 7-3A) as it did in the trachea of normal horses (Fig. 6-6B). At 10‘ and
10‘ M concentrations, clonidine inhibited ACh signiﬁcantly (Fig. 7-3A). However, the
magnitude of inhibition was less than that observed in the normal horses (about 30% less
at 10‘ M clonidine) (Figure 6-6B), although the difference was not statistically signiﬁcant
(P = 0.06 at 10‘ M clonidine).

In the heavey horse bronchi, clonidine did not inhibit ACh release signiﬁcantly
(Fig. 7-3B). This result is in contrast to its signiﬁcant inhibitory effect in the normal
horse bronchi (Fig. 6-7). However, when the magnitude of inhibition was compared,
there was no signiﬁcant difference between the normal and heavey horses.

Neither PGFQ (Fig. 7-4A) nor indomethacin (Fig. 7-4B) had any effect on ACh
release in the bronchial tissues. The results were similar to those obtained in the normal
horses (Fig. 5-7B and Fig. 5-5B)

The wet weight of each tissue preparation averaged 192.6 :1; 7.1, 180.6 :1: 27.4,
178.7 :1; 7.3, and 201.2 :1: 14.0 mg for control trachea (n = 5 bundles), control bronchi

(n = 10 segments), heavey trachea (n = 9 bundles), and heavey bronchi (n - 40

segments), respectively. In trachealis tissues whose dry weight was determined, the

155
dry/wet weight ratio was 0.214 :1: 0.003 (n = 5 bundles) and 0.214 :1: 0.002 (n = 5

bundles) in the normal and heavey horses, respectively.

Discussion

In the present study, I compared ACh release from airway cholinergic nerves of
trachealis preparations of normal and heavey horses. To do this, I ﬁrst needed to decide
how to normalize my data. In studies described in the earlier chapters, I expressed the
ACh release data as pmol/ g wet weight of tissue/min. Because the airways of heavey
horses may have edema, inﬁltration of inﬂammatory cells, or muscular hypertrophy,
comparison of data expressed in this unit may be unable to reﬂect the real situation. I
therefore also normalized the ACh release data by the dry weight and surface area of the
trachealis. Normalization by dry weight would remove the possible inﬂuence of edema.
Because the surface area of the trachea is unlikely to change signiﬁcantly in the heavey
horse, normalization by the surface area would remove the possible inﬂuence of muscular
hypertrophy and inﬁltration of inﬂammatory cells, which may affect both the wet and dry
weight.

No matter how I normalized my ACh release data, all the evidence indicates that
the ACh release rate was similar in the trachea between the two groups of horses. This
result is in contradiction to the conclusion of a previous study that suggests an increased
ACh release from airway cholinergic nerves in the heavey horse (Broadstone et al.
1991). The conclusion of the previous study was inferred by comparing the contractile
responses of airway smooth muscle to exogenous ACh and to EFS that releases

endogenous ACh. To use this latter approach, it is necessary to assume that exogenous

156
and endogenous ACh behave similarly. However, because the amount of the neurally

released ACh is much less than the exogenously administered ACh, contractions induced
by EF S are more likely to be inﬂuenced by modulators of muscle contractions than
contractions induced by exogenous ACh. Besides, various neurotransmitters may be
released from all kinds of nerves during EFS and may therefore affect the response to
EF S but not to exogenous ACh. Furthermore, if a factor has an effect on gap junctions,
it may inﬂuence the response to EFS but not to exogenous ACh. For these reasons, the
indirect approach may not always be able to provide a correct conclusion. I therefore
believe the evidence obtained in the present study.

In the heavey horse, blockade of muscarinic receptors by intravenous administra-
tion of atropine markedly reduces airway resistance during acute exacerbations of airway
obstruction. However, atropine does not have such an effect in the control horse or in
the heavey horse during remission (Broadstone et al. 1988). The inhibitory effect of
atropine on airway resistance in the heavey horse may be due to either an increased
sensitivity of airway smooth muscle to ACh or an increased local ACh concentration or
both. In vivo, the airway of heavey horses is hyperresponsive to aerosol of methacholine
(Armstrong et al. 1986), suggesting that increased sensitivity of airway smooth muscle
to neurally released ACh is a possible cause for the development of airway obstruction
and, therefore, for the response to atropine. However, isolated airway smooth muscle
preparations from heavey horses are not hyperresponsive to ACh in the tissue bath
(Broadstone et al. 1991), suggesting that airway smooth muscle may behave differently
in its neurohumoral environment in vivo. In the present study, ACh release reﬂects only

changes in efferent nerve function. I observed that ACh release from cholinergic nerves

157

innervating the trachea was not increased in the heavey horse under the in vitro
conditions. However, I could not exclude the possibility that ACh release is increased
in vivo, particularly if the increase is due to reﬂex activation or increased central ﬁring.
In addition, an increase in ACh release may happen only in peripheral airways where
airway inﬂammation is most severe.

Muscarinic autoreceptors are potent inhibitory receptors on airway cholinergic
nerves. The neurotransmitter ACh can act on prejunctional muscarinic autoreceptors to
inhibit further ACh release (Chapters 2, 3). When the receptors are blocked, ACh
release increases. In the present study, the non-selective muscarinic antagonist atropine
augmented ACh release to the same extent in the normal and heavey horses, suggesting
that the autoreceptors are not defective in heavey horses. This observation is in contrast
to those made in humans with asthma (Minette et al. 1989) and in guinea pigs infected
with parainﬂuenza virus (Fryer et al. 1990) or challenged by ovalbumin (Fryer and
Wills-Karp, 1991). In guinea pigs, antigen challenge is associated with a large inﬂux of
inﬂammatory cells, especially eosinophils, into the airway. Eosinophil proteins, such as
major basic protein, are antagonists for M2 receptors (Jacoby et al. 1992). Although
heaves is also a type of allergic pulmonary disease, the inﬂammatory cells in the lungs
are predominantly neutrophils and few eosinophils are present, as revealed by broncho-
alveolar lavage (Derksen et al. 1985c). If the presence of a large number of eosinophils
is necessary for the dysfunction of muscarinic autoreceptors, lack of numerous
eosinophils in the lungs may explain why the function of the autoreceptors is not altered
in the heavey horse. Furthermore, the subtype of the muscarinic autoreceptors in the

horse airways is not clear. Although the Mz-selective muscarinic antagonist gallamine

158

potentiates the contractile response of airway smooth muscle to EFS in guinea pigs (Fryer
and Maclagan 1984), cats (Blaber et al. 1985), and dogs (Ito and Yoshitomi 1988), which
has been attributed to the blockade of muscarinic autoreceptors, it does not have such an
effect in the horse (Y u et al. 1992a), suggesting that there are no gallamine-sensitive
functional inhibitory muscarinic receptors on the cholinergic nerve terminals. If only the
gallamine-sensitive muscarinic receptors are affected by the airway inﬂammation, the
autoreceptors should then have a normal function in the heavey horse because the
autoreceptors on horse airway cholinergic nerves probably belong to another subtype.

The inhibitory effect of the az-adrenoceptor agonist clonidine on ACh release was
remarkably yet insigniﬁcantly less in the heavey horse trachea than in the normal horse
trachea (Chapter 6). In the bronchi, clonidine had no signiﬁcant effect on ACh release
in the heavey horse as observed in the present study, but it inhibited ACh release
signiﬁcantly in the normal horse (Chapter 6). However, the magnitude of inhibition in
the bronchi was not signiﬁcantly different between the normal and heavey horses.
Therefore, the difference in prejunctional az-adrenoceptor function between normal and
heavey horses is unlikely to be substantial even if it truly exists. In heavey horses,
xylazine causes bronchodilation during acute exacerbations of heaves (Broadstone et al.
1992). Since xylazine is an az-agonist, the bronchodilatory response to xylazine may
suggest the presence of functional viz-receptors on airway cholinergic nerves.

My results obtained with PGE2 and indomethacin in the heavey horse were similar
to those obtained in the normal horse (Chapter 5). Neither PGEz nor indomethacin had

any effect on ACh release. Therefore, although the airway mucosal PGEz production is

 

159
decreased in heavey horses (Gray et al. 1992a), such a change in PGE2 production would

not augment ACh release from airway cholinergic nerves.

160

.833.— nU< Bonus—mam mucoucoooeéozgeoocoo BEBE .2953 some “8 380? we? Eng 53 3mm: 3 at.“
-5539 outshone <= N a can 5: some he 58 m “we: 05 waist <: R: “a .8 33 .8693 _8_anuobou_o 05 mo ESEEP. 05.
.359: 8.8: mﬁuatoaﬁ mote: £305.23 89¢ ammo—8 :U< :o oﬁqoba .«o Soto 05 wEBoam mEEonEoEu 18E» h AA. .wE

 

 

 

3352 3.352 335.2 . 33:22
. O . m:
M. m. a. N... o.
.92. 2.... u 05.2.0 .2.a «2: u 8:20 .9E 8.2 u 252.0 . .85 2.3 u 8:29
.2a «2 u 5< .28 «3 u .6... .28 m: u =o< .25 and u 5e.
./ 053...: 2 ob— ucﬁoha 2 «-3 2.383 2 $3 “EBB; 5933
I./ ./
7 # x . x H / ,
a/ I. / ﬁ/f _ \ ’1/
/ ~~ . r ax: \ P...................... /\ /.... a h ...........
r ...... a _ l .................. n....c./ .. .....o.....o.....u.... ...... ... 7’ 7 cl _
l/ a ’ V
. _ ,.. J ,. m

 

 

 

 

 

 

auuoua ._
__.4-
aurloqa
143‘.“ J”:—
3'.”qu H-ﬂ

 

 

 

HOV

 

 

 

 

 

 

auuoua

HOV

 

 

 

 

 

 

161

*control '9' heavey
n = 5

 

ACh release (multiples of pre-atropine)
b

0 1 I l 1
pre-atropine 8 7 6

 

m.—

Atropine concentration (-log M)

     

ﬁcontrol 9 heavey
n = 5

 

ACh release (multiples of pre-atropine)
b
l

1 l L 1

pre-atropine 8 7 6 5
Atropine concentration (—|og M)

L-

Fig. 7-2. Effect of atropine (10‘8—10‘ M) on ACh release from cholinergic nerves
innervating the trachea (A) and bronchi (B) of normal and heavey horses. Acetylcholine
release rate during the ﬁrst EFS period was regarded as l, and the rate during subsequent
periods was calculated as multiples of the rate of period 1. There was no difference in
the augmenting effect of atropine between normal and heavey horses.

162

A 120*

,.._ ﬂ

 

I——-)(——I

ACh release (96 of pre-clonidine)
8
l

 

 

*
*
40 —
l.
20 _ *control +clonidine
n = 5
0 l l L l
pre-clonidine 7 6 5

Clonidine concentration (-Iog M)

B 140—-

 

 

" A
100— ‘— ‘i.<—'i

I
I——<
I—«L

ACh release (% of pre-clonidine)

 

 

80
60 .-
40 .—
_ * control +clonidine
2° - (n=5)
0 l l l J
pre-clonidine 7 6 5

Clonidine concentration (-Iog M)

Fig. 7-3. Effect of clonidine (107—105 M) on ACh release from cholinergic nerves
innervating the trachea (A) and bronchi (B) of heavey horses. Clonidine inhibited ACh
release concentration-dependently in the trachea but was without an effect in the bronchi.

 

163

 

 

 

 

 

 

 

A 160 —
E
E
8
.. 120 -
o
12, .
8
80 —
1'3
9
5 'D'vehicle
< 40 -
O PGE2
n = 5
o I l l 9
control 9 8 7
PGE2 concentration (-log M)
r: 160 —
'o
.9
‘2. H
22’ .
o . . .
g 80 _ : Indomethacin or vehicle 1
.‘2
92
5 *vehlcle
< 40 ’ .lndomethacin (3 nM)
n = 5
1 l l !_J
0 1 2 3 4

Stimulation period

Fig. 7-4. Effects of PGE2 (109—10?7 M) (A) and indomethacin (3 x 10‘ M) (B) on ACh
release from cholinergic nerves innervating the heavey horse bronchi. Neither PGE2 nor
indomethacin showed any effect on ACh release.

CHAPTER 8

SUMMARY AND CONCLUSIONS

During my dissertation research, I studied prejunctional modulation of equine
airway cholinergic neurotransmission. My interest in equine airways was stimulated by
the Observation that there is a predominant muscarinic component to airway Obstruction
in the equine airway disease known as heaves.

In order to obtain direct evidence of airway cholinergic neurotransmission, I
adapted an ACh analysis technique to measure EFS-induced ACh release from isolated
equine airway smooth muscle preparations suspended in 2-ml tissue baths. With this
technique, I ﬁrst studied the effects of stimulation intensity and muscle preload on ACh
release (Chapter 2) because such information was important for the proper design of later
experiments. In the absence of EFS, ACh content in the incubation solution was hardly
detectable. Acetylcholine release increased sharply in response to EFS. Tetrodotoxin
was able to abolish the EFS-induced ACh release, indicating the cholinergic neural origin
of ACh. Acetylcholine release was dependent on stimulation intensity. Increasing
frequency (0.5-16 Hz) and voltage (5-20 V) increased ACh release; increasing pulse
duration (0.5-3 ms) had only a minor effect. Alterations in muscle preload (2-20 g) had
no effect on ACh release. ACh release rate was constant within each and over different

stimulation periods.

164

165

My observation that atropine augmented ACh release (Chapter 2) indicates the
presence of muscarinic autoreceptors on horse airway cholinergic nerves. In order to
determine which subtype of muscarinic receptors the autoreceptors belong to, the effects
of several selective muscarinic antagonists as well as the nonselective muscarinic
antagonist atropine were compared (Chapter 3). Although all muscarinic antagonists
augmented ACh release, neither hexahydrosiladifenidol (HHSiD, M3-selective),
pirenzepine (Ml-selective), nor AF-DX 116 (Mz-selective) was as potent or augmented
ACh release as much as atropine. Hexahydrosiladifenidol inhibited muscle contraction
more potently than ACh release, indicating that autoreceptors are not M3. Autoreceptors
are unlikely to be M1 because the augmenting effect of pirenzepine emerged at a
concentration 240 times its Kd for M1 receptors. Evidence was inadequate either to
conﬁrm or deny M2 receptors as muscarinic autoreceptors. Although the concentration
at which AF-DX began to augment ACh release was only 21 times its Kd for M2
receptors, the maximal effect Of AF-DX 116 was no greater than that Of HHSiD or
pirenzepine.

TO determine if PGE2 is important in regulating ACh release from equine airway
cholinergic nerves, I conducted two studies. In ponies, I studied the effects of exogenous
PCB, and endogenous prostanoids on ACh release by comparing the contractile response
of trachealis Strips to EFS and ACh (Chapter 4). Although exogenous PGE2 inhibited
the contractile response to both BF S and ACh in a concentration-dependent manner, the
concentration required to inhibit the response to EFS (10 nM) was less than that required
to inhibit the response to ACh (0.1 uM). Cyclooxygenase inhibition with aspirin

(10‘5-103 M) or meclofenamate (10‘3—10“ M) had no effect on either the response to EFS

 

 

 

166
or to ACh even though PGE,2 production was inhibited. My results suggest that

exogenous PGE,) inhibit airway cholinergic neurotransmission but endogenous prostanoids
do not have such an effect.

In horses, the effects of exogenous PGEI and endogenous prostanoids on ACh
release were determined by direct measurement of ACh release (Chapter 5). Neither
exogenous PGE2 (10'9-107 M) nor endogenous prostanoids were demonstrated to have
an inhibitory effect on ACh release in both the tracheal and bronchial preparations. In
contrast, I demonstrated that both exogenous PGE (10"-—10I7 M) and endogenous
prostanoids inhibit ACh release from canine airway cholinergic nerves.

In order to determine the presence and function of (viz-adrenoceptors on horse
airway cholinergic nerves, I studied the effects of several all-receptors agonists and
antagonists on ACh release (Chapter 6). The az-receptor agonists clonidine (10'7—105
M) and UK 14,304 (10'8—10‘ M) concentration-dependently inhibited ACh release and
this inhibition was attenuated by the az-receptor antagonists yohimbine and idazoxan.
These results indicate that az-adrenoceptors exist on horse airway cholinergic nerves and
activation of these receptors inhibits cholinergic neurotransmission.

I then turned my attention to horses with heaves in order to determine if ACh
release from airway cholinergic nerves is increased and if the prejunctional muscarinic
autoreceptors and org-adrenoceptors are dysfunctional under in vitro conditions. I
Observed that ACh release rate from cholinergic nerves innervating the trachealis was not
higher in heavey than normal horses. Blockade of muscarinic autoreceptors by atropine
(10'3-10‘ M) augmented ACh release to the same extent in normal and heavey horses,

suggesting that there is no dysfunction of muscarinic autoreceptors in the heavey horse.

 

167
Although clonidine inhibited ACh release in both the trachea and bronchi of normal

horses, it inhibited ACh release only in the trachea in horses with heaves. Furthermore,
the effect Of clonidine in the trachea was less in the heavey horse than in the normal
horse. The results suggest that the prejunctional az-inhibitory mechanism is probably
dysfunctional. The effect of exogenous PGE2 and endogenous prostanoids on ACh
release was only examined in the bronchi. As was true in the normal horses, neither of
them had any effect on ACh release. Lack of an effect of endogenous prostanoids on
airway cholinergic neurotransmission suggests that the reduced airway mucosal PGE,
production Of heavey horses (Gray et al. 1992) has no inﬂuence on ACh release from
airway cholinergic nerves.

For technical reasons, direct measurements of ACh release from airway
cholinergic nerves became possible only a few years ago. Traditionally, information
about ACh release from airway cholinergic nerves was inferred from comparisons of the
contractile response of airway smooth muscle to exogenous ACh and to neural activation
that releases endogenous ACh. Although conclusions Obtained by the two different
approaches are Often compatible, such as the case of prejunctional inhibitory a2-
adrenoceptors on horse tracheal cholinergic nerves (Chapter 6) and the effect of PGE2
on ACh release in canine airways (Deckers et al. 1989; Waters et al. 1984), I met with
several occasions when the two approaches led to different conclusions. First, cyclooxy-
genase inhibition markedly augments the contractile response of horse bronchial smooth
muscle to EFS with only a slight effect on the response to exogenous ACh, suggesting
that endogenous prostanoids inhibit airway cholinergic neurotransmission (Yu et al. in

press a), but direct ACh measurements indicate that ACh release did not increase

168
following administration of cyclooxygenase inhibitors (Chapter 5). Second, although

activation of prejunctional az-adrenoceptors by clonidine (10‘ and 10‘ M) inhibits ACh
release from cholinergic nerves in horse bronchi (Chapter 6), comparison of the response
of bronchial smooth muscle to EFS and ACh was unable to demonstrate such an effect
of clonidine over the same concentration range (LeBlanc et a1. 1993). Third, although
comparison Of the response of airway smooth muscle to EFS and exogenous ACh
suggests that ACh release from airway cholinergic nerves innervating trachealis of heavey
horses is increased under the in vitro conditions (Broadstone et al. 1991), direct
measurement Of ACh release could not conﬁrm this ﬁnding (Chapter 7). The speciﬁc
causes for these discrepancies are unclear. AS I indicated in the introduction Of my
dissertation, there are some potential problems in evaluating ACh release or prejunctional
modulation by comparing the response of airway smooth muscle to EFS and ACh. It
remains to be determined when it is appropriate to use the indirect approach to study
prejunctional modulation. The above-mentioned discrepancies highlight the importance

of direct measurement of ACh release.

REFERENCES

REFERENCES

Aas, P. Prejunctional control of cholinergic nerves in airway smooth muscle exerted by
muscarinic, purinergic and glutamergic receptors. In: Progress in Brain Research, Vol.
84, S-M Aquilonius, PG Gillberg eds., pp. 219-225, Elsevier Science Publishers B. V.
(Biomedical Division), 1990.

Aas P, Fonnum F . Presynaptic inhibition of acetylcholine release. Acta Physiol Scand
127: 335-342, 1986.

Aas P, Maclagan J . Evidence for prejunctional M2 muscarinic receptors in pulmonary
cholinergic nerves in the rat. Br J Pharmacol 101: 73-76, 1990.

Agrawal R, Daniel EE. Control of gap junction formation in canine trachea by
arachidonic acid metabolites. Am J Physiol 250 (Cell Physiol): C495-C505, 1986.

Andersson RGG, Grundstriim N. The excitatory non-cholinergic, non-adrenergic nervous
system of the guinea-pig airways. Eur J Respir Dis 64: 141—157, 1983.

Armstrong PJ , Derksen FJ , Slocombe RJ, Robinson NE. Airway responses to aerosolized
methacholine and citric acid in ponies with recurrent airway obstruction (heaves). Am Rev
Respir Dis 133: 357-361, 1986.

Ayala LE, Ahmed T. Is I there loss of a protective muscarinic receptor mechanism in
asthma? Chest 96: 1285—1291, 1989.

Baker DG, Brown JK. Potent muscarinic cholinergic regulation of acetylcholine release
from nerve endings in airway smooth muscle. Am Rev Respir Dis 143: A358, 1991.

Baker DG, Basbaum CB, Herbert DA, Mitchell RA. Transmission in airway ganglia of
ferrets: inhibition by norepinephrine. Neurosci Lett 41: 139—143, 1983.

Baker DG, McDonald DM, Basbaum CB, Mitchell RA. The architecture Of nerves and
ganglia Of the ferret trachea as revealed by acetylcholinesterase histochemistry. J Comp
Neurol 246: 513-526, 1986.

Baker DG, Don HF , Brown JK. Direct measurement of acetylcholine release in guinea
pig trachea. Am J Physiol 263 (Lung Cell Mol Physiol 7): L142-L147, 1992.

169

 

170

Barnes PJ. Neural control of human airways in health and disease. Am Rev Respir Dis
134: 1289-1314, 1986.

Barnes PJ . Cell surface receptors in airway smooth muscle. In: Airway Smooth Muscle
in Health and Disease, RF Coburn, ed. New York: Plenum Press, pp. 77-97, 1989a.

Barnes PJ. Muscarinic autoreceptors in airways: their possible role in airway disease.
Chest 96: 1220—1221, 1989b.

Barnes PJ . Muscarinic receptor subtypes: implications for lung disease. Thorax 44:
161-167, 1989c.

Barnes PJ . Neural control of airway smooth muscle. In: The Lung: Scientiﬁc
Foundations, RG Crystal, JB West, eds. New York: Raven Press, pp. 903-916, 1991.

Barnes P], Brown M], Silverman M, Dollery CT. Circulating catecholamines in exercise
and hyperventilation-induced asthma. Thorax 36: 435-440, 1981.

Barnes PJ , Skoogh B-E, Nadel JA, Roberts JM. Postsynaptic alpha2-adrenoceptors
predominate over alphal-adrenoceptors in canine tracheal smooth muscle and mediate
neuronal and hormonal alpha-adrenergic contraction. Mol Pharmacol 23: 570-57 5, 1982.

Barnes PJ , Basbaum CB, Nadel J A. Autoradiographic localization Of autonomic receptors
in airway smooth muscle: marked differences between large and small airways. Am Rev
Respir Dis 127: 758-762, 1983.

Baumann PA, Koella WP. Feedback control of noradrenaline release as a function of
noradrenaline concentration in the synaptic cleft in cortical slices of the rat. Brain Res
189: 437—448, 1980.

Beani L, Bianchi C, Siniscalchi A, Sivilotti L, Tanganelli S, Veratti E. Different
approaches to study acetylcholine release: endogenous ACh versus tritium efﬂux.
Naunyn-Schmiedeberg ’s Arch Pharmacol 328: 119-126, 1984.

Belvisi M, Stretton D, Barnes PJ. Nitric oxide as an endogenous modulator of

cholinergic neurotransmission in guinea-pig airways. Eur J Pharmacol 198: 219—221,
1991.

Belvisi MG, Stretton CD, Verleden GM, Ledingham SJL, Yacoub MH, Barnes PJ.
Inhibition of cholinergic neurotransmission in human airways by Opioids. J Appl Physiol
72: 1096—1100, 1992.

Belvisi MG, Miura M, Stretton D, Barnes PJ . Endogenous vasoactive intestinal peptide

and nitric oxide modulate cholinergic neurotransmission in guinea-pig trachea. Eur J
Pharmacol 231: 97-102, 1993.

171

Berne RM, Levy MN. Physiology, 2nd edition, Chapter 20: The autonomic nervous
system and its central control. The CV. Mosby CO., St. Louis, pp. 280-296, 1988a.

Berne RM, Levy MN. Physiology, 2nd edition, Chapter 54: The adrenal glands. The
CV. Mosby CO., St. Louis, pp. 280-296, 1988a.

Bethe] RA, McClure CL. Cyclooxygenase inhibitors increase canine tracheal muscle
response to parasympathetic stimuli in situ. J Appl Physiol 1990; 68: 2597-2603.

Blaber LC, Fryer AD, Maclagan J. Neuronal muscarinic receptors attenuate vagally-
induced contraction of feline bronchial smooth muscle. Br J Pharmacol 86: 723-728,
1985.

Black JL, Johnson PRA, Alfredson M, Armour CL. Inhibition by prostaglandin E2 of
neurotransmission in rabbit but not human bronchus: a calcium-related mechanism?
Prostaglandins 37: 317-330, 1989.

Bloom JW, Baumgartener-Folkerts C, Palmer JD, Yamamura HI, Halonen M. A
muscarinic receptor subtype modulates vagally stimulated bronchial contraction. J Appl
Physiol 65: 2144-2150, 1988.

Blusztajn JK, Wurtrnan RJ . Choline biosysthesis by a preparation enriched in
synaptosomes from rat brain. Nature 290: 417-418, 1981.

Borgent P, Samuelsson B. Arachidonic acid metabolism in polymorphonuclear
leukocytes: effects of ionophore A 23187. Proc Natl Acad Sci USA 76: 2148-2152,
1979.

Breeze RG. Heaves: The problem Of disease deﬁnition. Vet Clin North Am [Large An
Pract] 1: 219-230, 1979.

Breeze RG, Lee H, Grant BD. Toxic lung disease. Mod Vet Pract 59: 301, 1978.

Brichant J-F, Warner DO, Gunst SJ, Rehder K. Muscarinic receptor subtypes in canine
trachea. Am J Physiol 258 (Lung Cell Mol Physiol 2): L349-L354, 1990.

Broadstone RV, J. S. Scott, F. J. Derksen, and N. E. Robinson. Effects of atropine in
ponies with recurrent airway Obstruction. J Appl Physiol 65: 2720—2725, 1988.

Broadstone RV, Leblanc PH, Derksen FJ , Robinson NE. In vitro responses of airway

smooth muscle from horses with recurrent airway Obstruction. Pulmon Pharmacol 4:
191-202, 1991.

Broadstone RV, Gray PR, Robinson NE, Derksen FJ . Effects of xylazine on airway
function in ponies with recurrent airway obstruction. Am J Vet Res 53: 1813-1817, 1992.

172

Burnstock G. Autonomic neural control mechanisms: with Special reference to the
airways. In: The Airways: Neural Control in Health and Disease, MA Kaliner, PJ
Barnes, eds. New York: Marcel Dekker, Inc, pp. 1-22, 1988.

Cameron AR, Johnston CD, Kirkpatrick CT, Kirkpatrick CA. The quest for the
inhibitory neurotransmitter in bovine tracheal smooth muscle. Q J Exp Psychol 68:
413-426, 1983.

Chand N, Diamantis W, Soﬁa RD. Airway histamine receptors and their signiﬁcance in
allergic lung diseases. In: Airway Smooth Muscle: Modulation of Receptors and
Response, DK Agrawal, Townley RG. Boca Raton: CRC Press, pp. 259-269, 1990.

Cherubini E, Morita K, North RA. Opioid inhibition of synaptic transmission in the
guinea-pig myenteric plexus. Br J Pharmacol 85: 805-817, 1985.

Christoﬁdes ND, Yianggou Y, Piper PJ, Ghatei MA, Sheppard MN, Tatemoto K, Polak
JM, Bloom SR. Distribution of peptide histidine isoleucine in the mammalian respiratory
tract and some aspects of its pharmacology. Endocrinology 115: 1958-1963, 1984.

Chung KF, Evans TW, Graf PD, Nadel JA. Modulation Of cholinergic neurotransmission
in canine airways by thromboxane mimetic U46619. Eur J Pharmacol 117: 373-375,
1985.

Chung KF. Inﬂammatory mediators in asthma. In: Asthma as an Inﬂammatory Disease,
PM O’Byrne, ed. New York: Marcel Dekker, Inc, pp. 159-183, 1990.

Coburn RF. Peripheral airway ganglia. Ann Rev Physiol 49: 573—582, 1987.

Cushley MJ, Tattersﬁeld AE, Holgate ST. Inhaled adenosine and guanosine on airway
resistance in normal and asthmatic subjects. Br J Clin Pharmacol 15: 161—165, 1983.

Cushley MJ, Tattersﬁeld AE, Holgate ST. Adenosine induced bronchoconstriction in
asthma: antagonism by inhaled theophylline. Am Rev Respir Dis 129: 380-384, 1984.

D’Agostino G, Chiari MC, Grana E, Subissi A, Kilbinger H. Muscarinic inhibition of
acetylcholine release from a novel in vitro preparation of the guinea-pig trachea. Naunyn-
Schmiedeberg’s Arch Pharmacol 342: 141-145, 1990.

Daffonchio L, Abbracchio MP, DiLuca M, Hernandez A, Amadeo L, Cattabeni F,
Omini C. B-Adrenoceptor desensitization induced by antigen challenge in guinea-pig
trachea. Eur] Pharmacol 178: 21-27, 1990.

Dahlstrom A, Fuxe K, Hokfelt T, Norberg K-A. Adrenergic innervation Of the bronchial
muscle of the cat. Acta Physiol Scand 66: 507-508, 1966.

173

Daniel EE, Davis C, Sharma V. Effects of endogenous and exogenous prostaglandin in
neurotransmission in canine trachea. Can J Physiol Pharmacol 65: 1433-1441, 1987.

De J ongste J C, J ongejan RC, Kerrebijn KF. Control Of airway caliber by autonomic
nerves in asthma and in chronic Obstructive pulmonary disease. Am Rev Respir Dis 143:
1421-1426, 1991.

de Troyer A, Yernault J -C, Rodenstein D. Effect of vagal blockade on lung mechanics
in normal man. J Appl Physiol 46: 217—226, 1979.

Deckers IA, Rampart M, Bult H, Herman AG. Evidence for the involvement Of
prostaglandins in modulation of acetylcholine release from canine bronchial tissue. Eur
J Pharmacol 167: 415-418, 1989a.

Deckers IA, Rampart MR, Herman AG. Evidence for presynaptic modulation of
acetylcholine release in canine airways using a sensitive HPLC method (abstract). Arch
Int Pharmacodyn 298: 297, 1989b.

Derksen FJ , Robinson N E, Slocombe RF. Ovalbumin-induced lung disease in the pony:
role Of vagal mechanisms. J Appl Physiol 53: 719—725, 1982.

Derksen FJ, Robinson NE, Armstrong PJ, Stick JA, Slocombe RF. Airway reactivity in
ponies with recurrent airway Obstruction (Heaves). J Appl Physiol 58: 598—604, 1985a.

Derksen FJ , Scott D, Robinson NE, Slocombe RF, Armstrong PJ . Intravenous histamine
administration in ponies with recurrent airway obstruction (heaves). Am J Vet Res 46:
774-777, 1985b.

Derksen FJ , Scott J S, Miller DC, Slocombe RF, Robinson NE. Bronchoalveolar lavage
in ponies with recurrent airway obstruction (heaves). Am Rev Respir Dis 132:
1066-1070, 1985c.

Derksen FJ , Scott J S, Slocombe RF, Robinson N E. Micropolyspora faeni causes airway
inﬂammation but not hyperresponsiveness in sensitized ponies. J Appl Physiol 62:
1398—1404, 1987.

Dey RD, Shannon WA, Said SI. Localization of VIP-immunoreactive nerves in airways
and pulmonary vessels of dogs, cats and human subjects. Cell Tissue Res 220: 231-238,
1981.

Diamond L, Lantta J, Thompson DC, Altiere RJ. Nitric oxide synthase inhibitors fail to
affect cat airway nonadrenergic noncholinergic inhibitory (NAN CI) responses. (abstract)
Am Rev Respir Dis 145: A382, 1992.

174

Ding DJ, Martin JG, Macklem PT. Effects of lung volume on maximal methacholine-
induced bronchoconstriction in normal humans. J Appl Physiol 62: 1324-1330, 1987 .

Doldstein IM, Malmsten CL, Samuelsson B, Weissman G. Prostaglandins,
thromboxanes, and polymorphonuclear leukocytes. Inﬂammation 2: 309-317, 1977.

Doxey JC, Lane AC, Roach AG, Smith CFC, Watter DS. Selective az-adrenoceptor
agonists and antagonists. In: Pharmacology of Adrenoceptors. E Szabadi, CM Bradshaw,
SR Nahorski, eds. London: Macmillan Press Ltd, PP. 13-22, 1985.

Ducis I. The high-afﬁnity choline uptake system. In: The Cholinergic Synapse, VP
Whittaker, ed. Berlin: Springer-Verlag, pp. 409-445, 1988.

Duckles SP, Budai D. Stimulation intensity as critical determinant Of presynaptic receptor
effectiveness. T iPS 11: 440—443, 1990.

Dunnill MS. The pathology of asthma, with special reference to changes in the bronchial
mucosa. J Clin Pathol 13: 27-33, 1960.

Faulkner D, Fryer AD, Maclagan J. Postganglionic muscarinic inhibitory receptors in
pulmonary parasympathetic nerves in the guinea-pig. BrJ Pharmacol 88: 181-187, 1986.

Fink JN . Hypersensitivity pneumonitis. In: Immunologic and Infectious Reactions in the
Lung, CH Kirkpatrick, HY Reynolds eds. New York: Marcel Decker Inc, pp. 229-241,
1976.

Fisher JT, Anderson JW , Waldron MA. Nonadrenergic noncholinergic neurotransmitter
Of feline trachealis: VIP or nitric oxide?. J Appl Physiol 74: 31-39, 1993.

Fredholm BB, Hedqvist P. Modulation of neurotransmission by purine nucleotides and
nucleosides. Biochem Pharmacol 29: 1635-1643, 1980.

Fryer AD, El-Fakahany EB, Jacoby DB. Parainﬂuenza virus type 1 reduces the afﬁnity
Of agonists for muscarinic receptors in guinea-pig lung and heart. Eur J Pharmacol 181:
51-58, 1990.

Fryer AD, Maclagan J. Muscarinic inhibitory receptors in pulmonary parasympathetic
nerves in the guinea-pig. Br J Pharmacol 83: 973-978, 1984.

Fryer AD, Wills-Karp M. Dysﬁmction of Mz-muscarinic receptors in pulmonary
parasympathetic nerves after antigen challenge. J Appl Physiol 71: 2255-2261, 1991.

Gabella G. Innervation of airway smooth muscle: ﬁne structure. Ann Rev Physiol 49:
583-594, 1987.

175
Gerber H. Chronic pulmonary disease in the horse. Equine Vet J 5: 26-32, 1973.

Gill KK, Kroeger EA. Effects of indomethacin on neural and myogenic components in
equine airway smooth muscle. J Pharmacol Exp Ther 252: 358-364, 1990.

Gray PR, Derksen FJ, Robinson NE, Carpenter-Deyo Ll, Johnson HG, Roth RA. The
role of cyclooxygenase products in the acute airway obstruction and airway
hyperreactivity Of ponies with heaves. Am Rev Respir Dis 140: 154-160, 1989.

Gray PR, Derksen FJ, Broadstone RV, Robinson NE, Peters-Golden M. Decreased
airway mucosal prostaglandin E) production during airway Obstruction in an animal
model Of asthma. Am Rev Respir Dis 146: 586-591, 1992a.

Gray PR, Derksen FJ , Robinson N E, Peters-Golden ML. Epithelial strips: An alternative
method for examining arachidonate metabolism in equine tracheal epithelium. Am J
Respir Cell Mol Biol 6: 29—36, 1992b.

Grodzinska L, Panczenko B, Gryglewski RJ. Generation of prostaglandin E-like material
by the guinea-pig trachea contracted by histamine. J Pharm Pharmacol 27: 88-91, 1975.

Grundemar L, Widmark E, Waldeck B, Hakanson R. Neuropeptide Y: prejunctional
inhibition of vagally induced contractions in the guinea pig trachea. Regulatory Peptides
23: 309-313, 1988.

Grundstréim N, Andersson RGG, Wikberg JES. Prejunctional a, adrenoceptors inhibit
contraction Of tracheal smooth muscle by inhibiting cholinergic neurotransmission. Life
Sci 28: 2981-2986, 1981a.

Grundstréim N, Andersson RGG, Wikberg JES. Pharmacological characterizaton of the
autonomic innervation of the guinea pig tracheobronchial smooth muscle. Acta Pharmacol
et Toxicol 49: 150-157, 1981b.

Grundstréim N, Andersson RGG. Inhibition Of the cholinergic neurotransmission in
human airways via prejunctional alpha-Z-adrenoceptors. Acta Physiol Scand 125:
513-517, 1985.

Hahn HL, Wilson AG, Graf PD, Fischer SP, Nadel J A. Interaction between serotonin
and efferent vagus nerves in dog lungs. J Appl Physiol 44: 144-149, 1978.

Halliwell REW, Fleischman JB, Mackay-Smith M, Beech J, Gunson DE. The role Of
allergy in chronic pulmonary disease of horses. J Am Vet Med Assoc 174: 277-281,
1979.

176

Holgate ST, Mann J S, Cushley MJ. Adenosine as a bronchoconstrictor mediator in
asthma and its antagonism by methylxanthines. J Allergy Clin Immunol 74: 302—306,
1984.

Hughes AR, Horstrnan DA, Takemura H, Putney JW, Jr. InositOl phosphate metabolism
and signal transduction. Am Rev Respir Dis 141: S115—S118, 1990.

Ichinose M, Barnes PJ . Inhibitory histamine H3-reeeptors on cholinergic nerves in human
airways. Eur J Pharmacol 163: 383—386, 1989.

Ichinose M, Stretton CD, Schwartz JC, Barnes PJ. Histamine H3-receptors inhibit
cholinergic neurotransmission in guinea-pig airways. Br J Pharmacol 97: 13-15, 1989.

Inoue T, Ito Y, Takeda K. Prostaglandin-induced inhibition of acetylcholine release from
neuronal elements of dog tracheal tissue. J Physiol 349: 553-570, 1984.

Inoue T, Ito Y. Pre- and post-junctional actions of prostaglandin I2, carbocyclic
thromboxane A2 and leukotriene C4 in dog tracheal tissue. Br J Pharmacol 84: 289-298,
1985.

Ito I Suzuki H, Aizawa H, Hirose T, Hakoda H. Pre-junctional inhibitory action of
prostaglandin E2 on excitatory neuro-effector transmission in the human bronchus.
Prostaglandins 39: 639—655, 1990.

Ito Y. Pre- and post-junctional actions of procaterol, a Bz-adrenoceptor stimulant, on dog
tracheal tissue. Br J Phamtacol 95: 268-274, 1988.

Ito Y. Prejunctional control of excitatory neuroeffector transmission by prostaglandins
in the airway smooth muscle tissue. Am Rev Respir Dis 143: S6-SlO, 1991.

Ito Y, Tajima K. Actions Of indomethacin and prostaglandins on neurO—effector
transmission in the dog trachea. J Physiol 319: 379-392, 1981a.

Ito Y, Taj ima K. Spontaneous activity in the trachea of dogs treated with indomethacin:
an experimental model for aspirin-related asthma. Br J Pharmacol 73: 563—571 , 1981b.

Ito Y, Yoshitomi T. Autoregulation of acetylcholine release from vagus nerve terminals
through activation of muscarinic receptors in the dog trachea. Br J Pharmacol 93:
636-646, 1988.

J acobowitz D, Kent KM, Fleisch JH, Cooper T. Histoﬂuorescent study Of catecholamine-

containing elements in cholinergic ganglia from the calf and dog lung. Proc Soc Exp
Biol. Med 144: 464-466, 1973.

177

Jacobsson L, Grundstrom N, Andersson RGG. Inﬂuence Of some org-receptor agonists
and antagonists on the excitatory non-adrenergic, non-cholinergic neurotransmission in
the airways of guinea-pigs in vivo. Acta Physiol Scand 142: 91-96, 1991.

Jacoby DB, Gleich GJ, Fryer AD. Human eosinophil major basic protein is an
endogenous, allosteric antagonist at the inhibitory muscarinic M2 receptor. Am Rev Respir
Dis 145: A436, 1992.

J anssen LJ, Daniel EE. Characterization of the prejunctional beta adrenoceptors in canine
bronchial smooth muscle. J Pharmacol Exp Ther 254: 741-749, 1990a.

Janssen LI, Daniel EE. Pre- and postjunctional muscarinic receptors in canine bronchi.
Am J Physiol 259: L304-L314, 1990b.

Jones TR, Kannan MS, Daniel EE. Ultrastructural study of guinea pig tracheal smooth
muscle and its irmervation. Can J Physiol Pharmacol 58: 974-983, 1980.

Kagaya M, Shioya T, Onodera A, Sano M, Miura S, Ito Y, Kashima M, Kadowaki Y,
Kurokawa H, Itaba M, Miura K, Miura M. Muscarinic receptor subtypes in airways. Am
Rev Respir Dis 145: A56, 1992.

Kalsner S. Heteroreceptors, autoreceptors, and other termial sites. In: Presynaptic
receptors and the question of autoregulation of neurotransmitter release, S Kalsner, TC
Westfall, eds. New York: The New York Academy of Sciences, pp. 1-6, 1990a.

Kalsner S. The problem with autoreceptors. In: Presynaptic Receptors and the Question
of Autoregulation of Neurotransmitter Release, S Kalsner, TC Westfall. New York: The
New York Academy Of Sciences, pp. 414-438, 1990b.

Kamikawa Y, Shimo Y. Inhibitory effects of catecholamines on cholinergically and non-
cholinergically mediated contractions of guinea-pig isolated bronchial muscle. J Pharm
Pharmacol 42: 131—134, 1990.

Kanazawa H, Kurihara N, Hirata K, Takeda T. The role of free radicals in airway
Obstruction in asthmatic patients. Chest 100: 1319-1322, 1991.

Kandel ER. Factors controlling transmitter release. In: Principles of Neural Sciences,
second edition, ER Kandel and JH Schwartz, eds, New York: Elsevier Science
Publishing Co, Inc, pp. 120-131, 1985.

Kannan MS, Daniel EE. Structural and functional study of control of canine tracheal
smooth muscle. Am J Physiol 238 (Cell Physiol): C27-C33, 1980.

178

Kannan MS, Johnson DE. Nitric oxide mediates the neural nonadrenergic, noncholinergic
relaxation of pig tracheal smooth muscle. Am J Physiol 262 (Lung Cell Mol Physiol 2):
L511-L514, 1992.

Kawashima K, Ishikawa H, Mochizuki M. Radioimmunoassay for acetylcholine in the
rat brain. J Pharmacol Methods 3: 115-123, 1980.

Kay AB. Leucocytes in asthma. Immunol Invest 17: 679-705, 1988.

Kenakin TP. Drug-receptor theory. In: Pharmacologic Analysis of Drug-Receptor
Interaction. New York: Raven Press, pp. 1-30, 1987.

Kilbinger H, Schneider R, Sieﬂten H, Wolf D, D’Agostino G. Characterization of
prejunctional muscarinic autoreceptors in the guinea-pig trachea. Br J Pharmacol 103:
1757-1763, 1991.

Knight DS, Hyman AL, Kadowitz PJ . Innervation of intrapulmonary airway smooth
muscle of the dog, monkey and baboon. J Autonomic Nerv Syst 3: 31—43, 1981.

Knoll J, Vizi ES. Effect of frequency of stimulation on the inhibition by noradrenaline
of the acetylcholine output from parasympathetic nerve terminals. Br J Pharmacol 41:
263—272, 1971.

Kubes P, Suzuki M, Granger DN. Nitric oxide: an endogenous modulator of leukocyte
adhesion. Proc Natl Acad Sci USA 88: 4651-4655, 1991.

Laitinen LA, Laitinen A. Neural system. In: The Lung: Scientiﬁc Foundations, RG
Crystal, JB West et al, eds. New York: Raven Press, Ltd, pp. 759-765, 1991.

Laitinen LA, Heino M, Laitinen A, Kava T, Haahtela T. Damage of the airway
epithelium and bronchial reactivity in patients with asthma. Am Rev Respir Dis 131:
599-606, 1985a.

Laitinen A, Partanen M, Hervonen A, Pelto—Huikko—M A, and Laitinen LA. VIP like
immunoreactive nerves in human respiratory tract. Light and electron miscropic study.
Histochemistry 82: 313—319, 1985b.

Lammers JW, Minette P, McCusker, Barnes PJ. The role of pirenzepine-sensitive (M,)
muscarinic receptors in vagally mediated bronchoconstriction in humans. Am Rev Respir
Dis 139: 446-449, 1989.

Lansing MW, Ahmed A, Cortes A, Sielczak MW, Wanner A, Abraham WM. Oxygen
radicals contribute to antigen-induced airway hyperresponsiveness in conscious sheep. Am
Rev Respir Dis 147: 321-326, 1993.

179

Lawson, GHK, McPherson EA, Murphy JR, Nicholson JM, Wooding P. The presence
of precipitating antibodies in the sera of horses with chronic Obstructive pulmonary
disease (COPD). Equine Vet J 11: 172-176, 1979.

LeBlanc PH, Broadstone RV, Derksen FJ, Robinson NE. In vitro responses of distal
airways in horses with recurrent airway Obstruction. Am J Vet Res 52: 999-1003, 1991.

LeBlanc PH, Eberhart SW, Robinson NE. In—vitro effects of alpha2 adrenergic receptor
stimulation on cholinergic contractions of horse distal airways. Am J Vet Res 54:
788—792, 1993.

Leff AR, Tallet J, Munoz NM, Shoulberg N. Physiological antagonism caused by
adrenergic stimulation of canine tracheal smooth muscle. J Appl Physiol 60: 216-224,
1986.

Li CG, Rand MJ. Evidence that part of the NAN C relaxant response of guinea-pig
trachea to electrical ﬁeld stimulation is mediated by nitric oxide. Br J Pharmacol 102:
91-94, 1991.

Loenders B, Rampart M, Herman AG. Effect of selective muscarinic antagonists on the
release of acetylcholine from rabbit trachea. FASEB J 6: A1865, 1992a.

Loenders B, Rampart M, Herman AG. Selective M3 muscarinic receptor antagonists
inhibit smooth muscle contraction in rabbit trachea without increasing the release Of
acetylcholine. J Pharmacol Exp Ther 263, 773, 1992b.

Lowell FC. Observations on heaves. An asthma-like syndrome in the horse.J Allergy 35:
322-330, 1964.

Lundberg JM, Brodin E, Saria A. Effects and distribution of vagal capsaicin-sensitive
substance P neurons with special reference to the trachea and lungs. Acta Physiol Scand
119: 243—252, 1983a.

Lundberg JM, Martling C-R, Saria A. Substance P and capsaicin-induced contraction of
human bronchi. Acta Physiol Scand 119: 49-53, 1983b.

Lundberg JM, Saria A, Brodin E, Rusells S, Folkers R. A substance P antagonist inhibits
vagally induced increase in vascular permeability and bronchial smooth muscle
contraction in the guinea-pig. Proc Nat Acad Sci USA 80: 1120-1124, 1983c.

Lundberg JM, Fahrenkrug J, Hokfelt T, Martling C-R, Larsson O, Tatemoto K, Anggard
A. Coexistence of peptide HI (PHI) and VIP in nerves regulating blood ﬂow and
bronchial smooth muscle tone in various mammals including man. Peptides 5: 593-606,
1984a.

180

Lundberg JM, HOkfelt T, Martling C-R, Saria A, Cuello C. Substance P-immunoreactive
sensory nerves in the lower respiratory tract Of various mammals including man. Cell
Tissue Res 235: 251—261, 1984b.

Luz S, Pinchasi I, Michaelson DM. Differential inhibition of the release of endogenous
and newly synthesized acetylcholine from Torpedo synaptosomes by presynaptic
muscarinic receptors. FEBS Letters 164: 9—12, 1983.

MacIntosh FC. Formation, storage, and release of acetylcholine at nerve endings. Can
J Biochem Physiol 37 : 343-356, 1959.

Macklem PT. Mechanical factors determining maximum bronchoconstriction. Eur Respir
J 6: 5165-5195, 1989.

Macklem PT, Mead J. Resistance of central and peripheral airways measured by a
retrograde catheter. J Appl Physiol 22: 395-401, 1967.

Maclagan J, Faulkner D. Effect of pirenzepine and gallamine on cardiac and pulmonary
muscarinic receptors in the rabbit. Br J Pharmacol 97: 506-512, 1989.

Malik KU, Sehic E. Prostaglandins and the release of the adrenergic transmitter. In:
Presynaptic Receptors and the Question of Autoregulation of Neurotransmitter Release.
S Kalsner, TC Westfall, eds. New York: The New York Academy of Sciences, pp.
222-236, 1990.

Mann SP. The innervation of mammalian bronchial smooth muscle: the localization Of
catecholamines and cholinesterases. Histochem J 3: 319-331, 1971.

Mann J S, Holgate ST. Speciﬁc antagonism of adenosine induced bronchoconstriction in
asthma by oral theophylline. Br J Clin Pharmacol 19: 685-692, 1985.

Marmo E, Cazzola M, Chieppa S, Lampa E, Lobefalo G, Filippelli A. Documentation
Of presence of a purinergic reactivity at tracheobronchial level in man. Curr Ther Res 38:
118-134, 1985.

Martin JG, Collier B. Acetylcholine release from canine isolated airway is not modulated
by norepinephrine. J Appl Physiol 61: 1025-1030, 1986.

Martin JG, Wang A, Zacour M, Biggs DF. The effects Of vasoactive intestinal
polypeptide on cholinergic neurotransmission in an isolated innervated guinea pig tracheal
preparation. Respir Physiol 79: 111-122, 1990.

Matsuzaki Y, Hamasaki Y, Said SI. Vasoactive intestinal peptide: a possible transmitter
of nonadrenergic relaxation of guinea pig airways. Science 210: 1252-1253, 1980.

181

McPherson EA, Lawson GHK, Murphy JR, Nicholson JM. Chronic obstructive
pulmonary disease (COPD) in horses: Aetiological studies: Responses to intradermal and
inhalation antigenic challenge. Equine Vet J 11: 159—166, 1979a.

McPherson EA, Lawson GHK, Murphy JR, Nicholson JM, Breeze RG, Pirie HM.
Chronic obstructive pulmonary disease (COPD): factors inﬂuencing the occurrence.
Equine Vet J 11: 167—171, 1979b.

Mei L, Roeske WR, Yamamura HI. 1989. Molecular pharmacology of muscarinic
receptor heterogeneity. Life Sci 45, 1831-1851, 1989.

Minette PA, Barnes PJ . Prejunctional inhibitory muscarinic receptors on cholinergic
nerves in human and guinea pig airways. J Appl Physiol 64: 2532-2537, 1988.

Minette PA, Barnes PJ. Muscarinic receptor subtypes in lung: clinical implications. Am
Rev Respir Dis 141: S162-Sl65, 1990.

Minette PAH, Lammers JWJ, Dixon CMS, McCusker MT, Barnes PJ. A muscarinic
agonist inhibits reﬂex bronchoconstriction in normal but not in asthmatic subjects. J Appl
Physiol 67: 2461-2465, 1989.

Mitchell HW, Sparrow MP, Tagliaferri RP. Inhibitory and excitatory responses to ﬁeld
stimulation in fetal and adult pig airways. Ped Res 28: 69—74, 1990.

Miura M, Ichinose M, Kirnura K, Katsumata U, Takahashi T, Inouse H, Takishima T.
Dysfunction of nonadrenergic noncholinergic inhibitory system after antigen inhalation
in actively sensitized cat airways. Am Rev Respir Dis 145: 70—74, 1992.

Moreno RH, Hogg JC, Paré PD. Mechanics of airway narrowing. Am Rev Respir Dis
133: 1171-1180, 1986.

Munakata M, Masaki Y, Sakuma I, Ukita H, Otsuka Y, Homma Y, Kawakami Y.
Pharmacological differentiation of epithelium-derived relaxing factor from nitric oxide.
J Appl Physiol 69: 665-670, 1990.

N adel J A, Cabezas GA, Austin JHM. In vivo roentgenographic examination of
parasympathetic innervation of small airways: use of powdered tantulum and a ﬁne focal
spot x-ray. Invest Radio 6: 9-17, 1971.

Nakanishi H, Yoshida H, Suzuki T. Inhibitory effects of prostaglandin E1 and E2 on
cholinergic transmission in isolated canine tracheal muscle. Japan J Pharmacol 26:
669-674, 1976.

Obel NJ, Schmiterlow CG. The action of histamine and other drugs on the bronchial tone
in horses suffering from alveolar emphysema (heaves). Acta Pharmacol 4: 71-80, 1948.

182

Ollerenshaw, S, Jarvis D, Woolcock A, Sullivan C, Scheibner T. Absence of
immunoreactive vasoactive intestinal polypeptide in tissue from the lungs of patients with
asthma. N Eng J Med 320: 1244-1248, 1989.

Olsen CR, Colesbarh HJH, Mebel PE, Nadel JA, Staub NC. Motor control Of pulmonary
airways studied by nerve stimulation. J Appl Physiol 20: 202—208, 1965.

Orehek J, Douglas J S, Lewis AJ, Bouhuys A. Prostaglandin regulation of airway smooth
muscle tone. Nature New Biology 245: 84—85, 1973.

Ozaki T, Rennard SI, Crystal RG. Cyclooxygenase metabolites are compartmentalized
in the human lower respiratory tract. J Appl Physiol 62: 219-222, 1987.

Pack RJ, Richardson PS. The aminergic innervation of the human bronchus: a light and
electron microscopic study. J Anat 138: 493-502, 1984.

Palmer JB, Cuss FM, Barnes PJ. VIP and PHM and their role in nonadrenergic
inhibitory responses in isolated human airways. J Appl Physiol 61: 1322-1328, 1986.

Partanen M, Laitinen A, Hervonen A, Toivanen M, Laitinen LA. Catecholamine and
acetylcholinesterase containing nerves in human lower respiratory tract. Histochemistry

76: 175-188, 1982.

Polak JM, Bloom SR. Regulatory peptides in the respiratory tract of man and other
animals. Exp Lung Res 3: 313-328, 1982.

Post MJ, te Biesebeek JD, Werner J, van Rooij I-IH, Porsius AJ. Adenosine enhances
antigen-induced bronchoconstriction and histamine release in rat isolated lungs. Agents
and Actions 30: 30—33, 1990.

Potter PE, Meek JL, Neff NH. Acetylcholine and choline in neuronal tissue measured
by HPLC with electrochemical detection. J Neurochem 41: 188-194, 1983.

Prast H, Philippu A. Nitric oxide releases acetylcholine in the basal forebrain. Eur J
Pharmac01216: 139-140, 1992.

Racké K, Biihring J, Langer C, Brautigam M, Wessler I. Prostanoids inhibit release of
endogenous norepinephrine from rat isolated trachea. Am Rev Respir Dis 146:
1182-1186, 1992.

Rhoden KJ, Meldrum LA, Barnes PJ . Inhibition of cholinergic neurotransmission in
human airways by Bz-adrenoceptors. J Appl Physiol 65: 700-705, 1988.

Richardson J, Beland J. Nonadrenergic inhibitory nervous system in human airways. J
Appl Physiol 41: 764-771, 1976.

 

183

Robinson C, Holgate ST. Regulation by prostanoids. In: The Lung: Scientiﬁc
Foundations, RG Crystal, JB West. New York: Raven Press, pp. 941-951, 1991.

Robinson NE. Functional abnormalities caused by upper airway obstruction and heaves:
their relationship to the etiology of epistaxis. Vet Clin North Am [Large An Pract] 1:
17—34, 1979.

Robinson NE. Airway Obstruction in the horse. J Equine Vet Sci 9: 155-160, 1989.

Russell JA. Nonadrenergic inhibitory innervation Of canine airways. J Appl Physiol 48:
16—22, 1980.

Said SI. Nitric oxide and vasoactive intestinal peptide: cotransmitters of smooth muscle
relaxation. NIPS 7: 181-183, 1992.

Sekizawa K, Tamaoki J, Graf PD, Nadel J A. Modulation Of cholinergic neurotrans-
mission by vasoactive intestinal peptide in ferret trachea. J Appl Physiol 64: 2433-2437 ,
1988.

Sekizawa K, Graf PD, Nadel J A. Somatostatin potentiates cholinergic neurotransmission
in ferret trachea. J Appl Physiol 67: 2397-2400, 1989.

Sheppard MN, Kurian SS, Henze-Logmans SC, Michetti F, Cocchia D, Cole P, Rush
RA, Marangos PJ, Bloom SR, Polak JM. Neuron-speciﬁc enolase and S-100. New
markers for delineating the innervation of the respiratory tract in man and other
mammals. Thorax 38: 333-340, 1983.

Shore S, Collier B, Martin JB. Effect of endogenous prostaglandins on acetylcholine
release from dog trachealis muscle. J Appl Physiol 62: 1837-1844, 1987.

Small RC, Boyle JP, Foster RW, Good DM. Airway smooth muscle: electrophysiological
properties and behavior. In: Airway Smooth Muscle: Modulation of Receptors and
Response, DK Agrawal, RG Townley, eds. Boca Raton: CRC Press, pp. 69-94, 1990.

Snow RW, Weinreich D. Presynaptic and postsynaptic effects of histamine and histamine
agonists in the superior cervical ganglion of the rat. Neuropharmacology 26: 743-752,
1987.

Snow DH, Harris RC, Macdonald IA, Forster CD, Marlin DJ. Effects of high-intensity
exercise on plasma catecholamines in the thoroughbred horse. Equine Vet J 24: 462-467,
1992.

Starke K. Presynaptic cit-autoreceptors. Rev Physiol Biochem Pharmacol 107: 73-146,
1987.

 

 

184

Steinberg TH. Adenosine and adenosine receptors. Am J Respir Cell Mol Biol 2:
125-126, 1990.

Takaganagi I, Kawano K-I, Koike K. az-adrenoceptor mechanisms in guinea-pig trachea.
Eur J Pharmacol 182: 577-580, 1990.

Tamaoki J, Sekizawa K, Graf PD, N adel JA. Cholinergic neuromodulation by
prostaglandin D2 in canine airway smooth muscle. J Appl Physiol 63: 1396-1400, 1987.

Tanaka DT, Grunstein MM. Maturation of neuromodulatory effect Of substance P in
rabbit airways. J Clin Invest 85: 345-350, 1990.

Taylor SM, Pare PD, Schellenberg R. Cholinergic and nonadrenergic mechanisms in
human and guinea pig airways. J Appl Physiol 56: 958-965, 1984.

Thompson DC, Diamond L, Altiere RJ. Presynaptic alpha adrenoceptor modulation of
neurally mediated cholinergic excitatory and nonadrenergic noncholinergic inhibitory
responses in guinea pig trachea. J Pharmacol Exper Ther 254: 306-311, 1990.

Thurlbeck WM, Lowell FC. Heaves in horses. Am Rev Respir Dis 89: 82-88, 1964.

Tsuchiya Y, Hosokawa T, Kaswa Y. Effects of histamine on neurally mediated
contraction of canine tracheal smooth muscle. Japan J Pharmacol 52: 647-651, 1990.

Tucek S. Choline acetyltransferase and the synthesis of acetylcholine. In: The Cholinergic
synapse, VP Whittaker, ed. Berlin: Springer-Verlag, pp. 125—165, 1988.

Tucker JF, Brave SR, Charalambous L, Hobbs AJ, Gibson A. L-NG-nitro argine inhibits
non-adrenergic, non-cholinergic relaxations of guinea-pig isolated tracheal smooth
muscle. Br J Pharmacol 100: 663-664, 1990.

Tiirker RK, Zengil H. Release of prostaglandin-like material from isolated cat tracheal
muscle by electrical and mechanical stimulation. Arch Int Physiol Biochim 84: 833-841,
1976.

van Koppen CJ, Blankesteijn WM, Klaassen ABM, Rodrigues de Miranda JF, Beld AJ,
van Ginneken CAM. Autoradiographic visualization of muscarinic receptors in human
bronchi. J Pharmacol Exp Ther 244: 760—764, 1988

Vincent NJ, Knudson R, Leith DE, Macklem PT, Mead J. Factors inﬂuencing
puhnonary resistance. J Appl Physiol 29: 236-243, 1970.

Vizi ES, Kiss J, Elenkov IJ. Presynaptic modulation of cholinergic and noradrenergic
neurotransmission: Interaction between them. NIPS 6: 119-123, 1991.

185

Wang SR, Yang CM, Wang SSM, Han SH, Chiang BN. Enhancement of A 23187-
induced production of the slow-reacting substance on peripheral leukocytes from subjects
with asthma. J Allergy Clin Immunol 77: 465-471, 1986.

Wang Z, Yu M, Robinson N E, Broadstone RV, LeBlanc PH, Derksen FJ . Exogenous
but not endogenous PGE, modulates pony tracheal smooth muscle contractions. Pulmon
Pharmacol 5: 225-231, 1992.

Warren JB, Keynes RJ, Brown MJ, Jenner DA, McNicol MW. Blunted sympathoadrenal
response to exercise in asthmatic subjects. Br J Dis Chest 76: 147-150, 1982.

Waters EH, O’byrne PM, Fabbri LM, Graf PD, Holtzman J, Nadel JA. Control of
neurotransmission by prostaglandins in canine trachealis smooth muscle. J Appl Physiol
57: 129-134, 1984.

Wharton J, Polak JM, Bloom SR, Will JA, Brown MR, Pearse AGE. Substance P-like
immunoreactive nerves in ammmalian lung. Invest Cell Pathol 2: 3—10, 1979.

Widdicombe JG. Action potentials in vagal efferent nerve ﬁbers to the lungs of cat.
Naunyn Schmiedebergs Arch Pharmacol 241: 415-432, 1961.

Wood JD. Enteric neurophysiology. Am J Physiol 247: GS85—G598, 1984.

Woods PSA, Robinson NE, Swanson MC, Reed CE, Broadstone RV, Derksen FJ.
Airborne dust and aeroallergen concentration in a horse Stable under two different
management systems. Equine Vet J 25: 208-213, 1993.

Xie Z, Hirose T, Hakoda H, Ito Y. Effects of vasoactive intestinal polypeptide

antagonists on cholinergic neurotransmission in dog and cat trachea. Br J Pharmacol 104:
938-944, 1991.

Yu M, Robinson NE, Wang Z, Derksen FJ . Muscarinic receptor subtypes in equine
tracheal smooth muscle. Vet Res Commun. 16: 301-310, 1992a.

Yu M, Wang Z, Robinson NE, Derksen FJ . Furosemide’s inhibitory effect on the
contractile response Of equine trachealis to cholinergic nerve stimulation. Pulm
Pharmacol 5: 233-238, 1992b.

Yu M, Robinson NE, Wang Z, Derksen FJ . Independent modulation of horse airway
smooth muscle contraction by epithelium and prostanoids. Respir. Physiol. (in press 3).

Yu M, Wang 2, Robinson NE, LeBlanc PH. Distribution of inhibitory nerves and
mediation of nonadrenergic-noncholinergic relaxation by nitric oxide in equine airway
smooth muscle (in press b).