Download Hypoglossal Nerve Response to 5-HT3 Drugs Injected into the XII

BASIC RESEARCH
Hypoglossal Nerve Response to 5-HT3 Drugs Injected into the XII Nucleus and
Vena Cava in the Rat
Polina Fenik,1,3 Hiromasa Ogawa,2,3 and Sigrid C. Veasey 1,3
1Department of Medicine, School of Medicine; 2Department of Animal Biology, School of Veterinary Medicine; 3Center for Sleep & Respiratory
Neurobiology, University of Pennsylvania, Philadelphia, PA
Summary: Systemically administered ondansetron, a 5-HT3 receptor
antagonist, reduces obstructive sleep-disordered breathing (OSDB)
events in the English bulldog. The neural mechanisms through which
ondansetron acts are unknown. 5-HT3 receptor immunoreactivity and
mRNA have been detected in the vicinity of upper airway dilator motoneurons (UAWDMn's), suggesting that this receptor may contribute to the
state-dependent modulation of UADMn's. To characterize 5-HT3 receptor
activity within a representative UAWD nucleus, we performed acute
microinjections of selective 5-HT3 drugs, 1-(m-Chlorophenyl)-biguanide
HCl, an agonist, and ondansetron, an antagonist, into a major population
of UADMn's, the hypoglossal nucleus (XII), in anesthetized, paralyzed and
mechanically-ventilated rats. The 5-HT3-selective drugs neither altered
the baseline XII nerve activity nor the excitatory effect of 5-HT microinjected into the XII. In contrast, systemic-administration of ondansetron (3
mg/kg) produced a significant increase in the inspiratory modulation of XII
nerve activity (to 195.8%±19.3 of control, p<0.001). Together, these data
suggest that 5-HT3 receptors within the XII nucleus do not mediate 5-HT
effects on XII motoneurons, rather antagonism of 5-HT3 receptors outside
the XII nucleus can increase respiratory drive to XII motoneurons. These
results highlight the importance of understanding serotonergic effects on
respiratory drive outside the UAWD motor nuclei as we search for 5-HT
drug therapies for OSDB.
Key words: Sleep apnea; serotonin; XII; motoneurons; 5-HT3; CPBG;
ondansetron; nodose ganglion
INTRODUCTION
tions in serotonin delivery to upper airway motoneurons may
contribute to the pathogenesis of this disorder.5,6,13,14 We have
shown that broad spectrum 5-HT antagonists systemically
administered to an animal model of OSDB, the English bulldog,
reduces upper-airway muscle activity and results in collapse of
the upper-airway.13 Further support that sleep-related withdrawal of 5-HT contributes to upper-airway muscle suppression
comes from a recent study finding that 5-HT dialyzed into the
hypoglossal airway largely prevents the reduction in XII
motoneuronal activity in NREM sleep.14
The mechanisms through which 5-HT may impact upon
breathing in sleep, however, likely extend beyond direct effects
at UADMn's (for review, see ref. 15). Specifically, 5-HT may
directly act on central respiratory neurons,9,16 reflex responses,10
and behavioral state control.17 In addition, peripheral effects of
5-HT may change respiratory output.18 In contrast to the excitatory effects of 5-HT within the hypoglossal nucleus, peripherally, 5-HT reduces respiratory drive. Understanding the 5-HT
receptor subtypes mediating each of these effects has obvious
pharmacological implications.
One of the 5-HT receptor subtypes that may impact upon respiration either at UADMn's or peripherally is the 5-HT3 receptor.
There are several recent reports suggesting that 5-HT, through a
peripherally mediated mechanism, has powerful effects on respiration and that ondansetron, administered systemically, reduces
the frequency of central sleep apnea events in rats.19 We have
recently shown that ondansetron, a 5-HT3 antagonist reduces
OSDB in the English bulldog, a natural animal model of this disease.20 5-HT3 mRNA21 and 5-HT3 immunoreactivity22 have been
identified in the XII nucleus. The XII nucleus is the source of
innervation of the genioglossal muscle of the tongue, a muscle
believed to play a critical role in upper-airway patency in persons
with OSDB.3 Within the large family of 5-HT receptor subtypes,
OBSTRUCTIVE SLEEP APNEA-HYPOPNEA SYNDROME,
ALSO REFERRED TO AS OBSTRUCTIVE SLEEP-DISORDERED BREATHING (OSDB), IS A PREVALENT DISORDER1 for which there is no widely effective pharmacotherapy.2
The pathophysiology of OSDB suggests that, in anatomically
predisposed individuals, sleep-related reductions in upper-airway
dilator activity play a permissive role in airway obstruction.3,4
There are many ways in which sleep states may influence upperairway dilator activity, including state-dependent changes in neurochemical delivery to the dilator motoneurons, changes in excitation of central respiratory neurons, and alterations in reflexes.
One of the neurochemicals implicated in sleep state-dependent changes in respiratory activity is serotonin (5-HT).5,6 5-HT
can significantly increase motoneuronal activity,7-10 including the
activity of upper-airway dilator motoneurons (UADMn's).5,9,10
Importantly, the activity of 5-HT neurons is reduced in non-rapid
eye movement (NREM) sleep and further reduced in rapid eye
movement sleep (REM).11 Further, the responsiveness of these
neurons to various motor challenges may be blunted in sleep.12
Therefore, 5-HT at upper-airway motoneurons may play an
important role in the maintenance of patent upper airways during
wakefulness in subjects with OSDB, and the sleep-related reduc-
Disclosure Statement
This work was supported by NIH, HL-60287.
Accepted for publication September 2001
Address correspondence to: Sigrid Carlen Veasey, MD, Center for Sleep and
Respiratory Neurobiology, University of Pennsylvania School of Medicine,
972 Maloney Bldg., 3600 Spruce St. Philadelphia, PA 19104-4283;
Tel: 215-349-8015; E-mail: [email protected]; Fax: 215-662-7749
SLEEP, Vol. 24, No. 8, 2001
871
5-HT3 Receptors and XII Activity—Fenik et al
XIIright
XIIleft
ETCO2
Ptrach
BP
60 s
Figure 1A—The effects of 1mM (10nl) 5-HT microinjected into the XII nucleus of a urethane-anesthetized, paralyzed rat. The top two traces show the moving averages of rectified nerve activity in the left and right XII nerves. The short dark line under the left hypoglossal nerve tracing (XIIleft) marks the onset through offset of
the injection of 5-HT. Immediately following injection there is a sustained (5—10 minute) increase in XII nerve activity on the right side, with only a minimal and delayed
effect of the left XII nerve. Baseline inspiratory XII nerve activity was measured as the difference between peak and expiratory values (arbitrary units). Tonic increase
in activity was quantitated as the percentage (%) increase in expiratory activity relative to the baseline inspiratory activity, while the increase in inspiratory activity was
calculated as the % increase relative to baseline inspiratory activity. In all rats, tonic XII activity was minimal or absent.
nists modify the activity of a representative population of upperairway motoneurons, XII motoneurons, when injected directly
into the XII nucleus. The second objective was to determine
whether systemically administered 5-HT3-active drugs can augment XII nerve activity. To address the first issue, we performed
microinjections of 5-HT3 receptor-specific drugs into the XII
nucleus in the rat while measuring XII motoneuronal activity,
with blood pressure, tracheal pressure, oxygenation, and endtidal CO2 held constant. To determine the effects of systemic 5HT3-active drugs on XII nerve activity, we injected selective agonist and antagonist drugs into the inferior vena cava, in anesthetized, vagotomized, mechanically ventilated, and paralyzed
rats. Preliminary results from these studies were published.25
Figure 1B—Histological confirmation of an injection site into the left XII nucleus.
Pontamine Sky Blue dye was injected with one of the drugs and then counterstain with Neutral Red allowed histological localization of injection sites.
Typically, penetrations of the nucleus with the pipettes containing different drugs
or concentrations of drug left a noticeable tract within the medulla. The arrow
points to the center of the Pontamine Sky Blue injection.
METHODS
Animals
Adult male Sprague-Dawley rats (Charles River) weighing
320-390g were used. All procedures were approved by the
Institutional Animal Care and Use Committee of the University
of Pennsylvania.
the 5-HT3 receptor is a ligand-gated ion channel, that when
opened results in depolarization.23 5-HT3 receptors have been
identified on GABAergic neurons within the forebrain.24 Thus,
one mechanism through which a 5-HT3 antagonist (i.e.,
ondansetron) might increase hypoglossal activity within the
hypoglossal nucleus may be through antagonizing a 5-HT excitatory influence on GABAergic or glycinergic interneurons.24
The intents of the present study were two-fold. The first goal
was to determine whether selective 5-HT3 agonists and antagoSLEEP, Vol. 24, No. 8, 2001
Surgery
Seventeen rats were pre-anesthetized with halothane (2.02.5%) and then anesthetized with either urethane 1g/kg intraperitoneally (n=12 rats) or ketamine (80 mg/kg) and xylazine (10
mg/kg) intramuscularly (n=5). A tracheostomy was performed to
place a tracheal cannula, and one femoral artery and vein were
872
5-HT3 Receptors and XII Activity—Fenik et al
cannulated for arterial pressure monitoring (Validyne pressure
transducer model) and drug and fluid administration, respectively. The vagi were exposed and transected at the level of the third
tracheal ring. Both XII nerves were dissected and transected at
the level of the ramification into medial and lateral branches.
The proximal ends of both XII nerves were placed in custommade cuff recording electrode. The animal was then placed
prone with its head positioned securely into a stereotaxic frame
and its neck flexed at 30.° Mechanical ventilation was started
using a custom-made constant flow small animal ventilator
attached to the tracheal tube. End-tidal CO2 was measured with
the small animal capnograph (Columbus Instruments) and maintained constant at 5.0—6.0%. The fraction of inspired oxygen
was held constant at 30%. With initiation of mechanical ventilation, the animal was paralyzed with pancuronium bromide (1
mg/kg, i.v.). A rectal thermistor probe (Yellow Springs) was
placed, and body temperature was then maintained at 36.5—
37.5°C with a heating pad and lamp.
The surgical approach used in this study has been described.26
Briefly, skin and muscles overlying the atlanto-occipital area
were cut along the midline and retracted. After removal of the
atlanto-occipital membrane, the caudal edge of the occipital bone
was removed; the dura was opened and the pia was gently
removed, exposing the calamus scriptorius (CS) and caudal tip of
the cerebellar vermis. Anesthesia and paralysis in all rats were
maintained by constant infusion of urethane (125 mg/kg/h) and
pancuronium (0.25 mg/kg/h) in sufficient saline to deliver 0.5 ml
volume/h.
testing 5-HT3 drugs, a pipette containing 5-HT (1 mM 5-HT creatinine sulphate in saline, Sigma) was inserted into the XII nucleus and injected to confirm proper selection of the microinjection
site. Sites were regarded appropriate when 5-HT injection produced at least a 25% increase in activity (Fig. 1A). In two animals, glutamate (1 mM L-glutamic acid monosodium salt,
Sigma), rather than 5-HT, was used to verify proper placement of
the pipette.
Occasionally, when the first injection of 5-HT or glutamate
did not produce an appropriate excitatory response, the placement of the pipette was changed by increments of 200 m rostrocaudally or 50 mm dorso-ventrally until the expected excitatory response to either 5-HT or glutamate was observed. After having established the desired coordinates for the injections, a
pipette containing one of the 5-HT3 drugs was inserted, and the
drug was microinjected; three to ten minutes later, the original
injection of either 5-HT or glutamate was repeated to determine
effect of the drug on 5-HT or glutamate effect and to confirm the
stability of the response to 5-HT or glutamate. Two to five trials
were performed in each animal gradually increasing concentrations of the 5-HT3 drugs.
Drugs
XII nerve activity was pre-amplified and filtered (50-2000 Hz;
BMA-830 amplifier, CWE, Inc.). A moving time average was
obtained from the rectified nerve signals (time constant 100 ms;
MA-821 RSP moving averager, CWE, Inc.). The raw and processed XII nerve signals, tracheal pressure, end-tidal CO2, and
blood pressure were monitored continuously on a chart recorder
(TA11, Gould) and simultaneously recorded on a digital tape
recorder (Cygnus Technology, CDAT-16).
The selective 5-HT3 agonist tested was 1-(m-Chlorophenyl)biguanide HCl (CPBG)25 (RBI, Natick, MA) and the selective
antagonist studied was ondansetron (Glaxo Wellcome
Pharmaceuticals).27 Both drugs are highly potent at 5-HT3 receptors within the nucleus tractus solitarius (NTS), where picomolar
amounts are effective in modifying blood pressure in urethaneanesthetized rats.28-30 Concentrations and volumes of the agonist
(0.1 µM to 50 mM) and antagonist (0.1 mM to 5 mM) were
selected based on the relative affinities (both in the nM range) of
each drug,25,265,30-32 and the effective doses used in earlier in vivo
microinjection studies in the NTS.27-298,33 2% Pontamine Sky
Blue dye was added to one of the 5-HT3 drug dilutions injected
later in each experiment, to mark the injection site. All injections
from which data are reported were then histologically confirmed
to have been placed within the XII nucleus (Figure 1B).
Microinjection Protocol
Histology
Microinjections of 5-HT and 5-HT3 receptor agonists and
antagonists were placed within one XII nucleus (typically the
right nucleus) using glass pipettes, as described previously.26
Single-barrel glass pipettes with tip diameters of 20—30 mm
were filled with one of the drugs and placed in a three-dimensional manipulator with hydraulically controlled vertical movement (David Kopf manipulator). While pressure was manually
applied to the fluid in the pipette, the microinjected volume was
monitored by direct observation of movement of the fluid meniscus within the pipette with aid of a calibrated microscope.
Microinjections were targeted towards one XII nucleus using
the following coordinates relative to the CS: ±1.5 mm rostral; 0.2
mm lateral and 1.05—1.25 mm below the dorsal medullary surface. The injected volumes were 10 nl for 5-HT and 5-HT3 agonists and 20 nl for antagonists, and were delivered over 5-15 seconds. After positioning the pipette at the appropriate coordinates,
at least two minutes of stable XII nerve activity was recorded
before drugs were administered. In most experiments, prior to
At the end of the experiments, under deep urethane anesthesia
(2-5 g/kg, i.v.), each animal was decapitated and the brainstem
was removed and placed for one to three days in 30% sucrose and
4% paraformaldehyde. Brainstems were sectioned coronally
using a vibratome. Sections containing the XII nucleus were
serially mounted and counter-stained with Neutral Red.
Placement of the Pontamine Sky Blue dye relative to the XII
nucleus was determined with microscopic review.
Recording Procedures
SLEEP, Vol. 24, No. 8, 2001
Data Analysis
As baseline recordings of XII nerve activity were stable breath
to breath and there was minimal or no end-expiratory activity in
the XII nerve recordings, the amplitude of phasic inspiratory
activity was measured from the moving average of the signal. A
control level of this activity was used to normalize responses to
drugs in each animal. The effect of microinjections on XII nerve
activity was measured 20—50 seconds after a drug injection was
873
5-HT3 Receptors and XII Activity—Fenik et al
Figure 2—An example of the effects of systemically administered ondansetron (2 mg/kg) on the moving average of the rectified XII nerve activity. In this sample,
ondansetron produced a two-fold increase in the amplitude of inspiratory XII nerve activity. Whereas the tonic activity is minimally affected. The timing of the
ondansetron injection (a) and a saline flush (b) are indicated by the two event markers.
prior to CPBG. Rats included in the microinjection studies and 3
additional rats with stable XII nerve recordings, blood pressure
and end-tidal CO2 received systemic injections of ondansetron.
completed. This delay of 20 seconds served to allow for recovery from mechanical effect of the injection, if any, and spread
within the nucleus.25 For quantifying systemic drug effect, XII
nerve activity and blood pressure effects were examined for two
to ten minutes after systemic injection over succesive one-minute
sampling periods.
Data were expressed relative to the pre-injection level of XII
nerve activity±standard error. Sample size for all statistical analyses was the number of rats tested under each specific condition,
rather than the number of trials performed. Drug responses were
examined using the Student's t-test for paired data. Two-way
analysis of variance with the Bonferroni/Dunn post-hoc test was
used to determine if 5-HT3 drugs altered the 5-HT excitation of
XII motoneurons. In all circumstances, values were interpreted
as statistically significant if probability of the null hypothesis was
<0.05.
Effects of Systemic Injections of the Selective 5-HT3
Antagonist, Ondansetron
Seven rats received a series of systemic injections of
ondansetron at cumulative doses 0.25-0.5 mg/kg, 1 mg/kg, 2
mg/kg and three mg/kg. The lowest dose did not alter XII nerve
activity, 99%±1 of control. 1 mg/kg increased the nerve activity
in 7 rats, to 134%±13 of control (t=-2.8, p=0.03). 2 mg/kg
appeared to increase XII activity (191% ± 30) but with n=4 this
did not reach significance, (t=-2.9, p=0.06). At the highest dose,
3 mg/kg, XII activity was increased to 195%±19 of control (n=7,
t = -7.2, p=0.0004). An example of the response to 3 mg/kg of
ondansetron is provided in Figure 2. At all doses, the phasic
inspiratory activity was significantly augmented with minimal or
no effect on tonic (expiratory) activity. The increase in phasic
XII nerve activity was dose dependent (r2=.81, F=48.8, p<0.001).
In addition, respiratory rate increased by 25%±4 (p<0.05) at the
highest dose without significant changes at smaller doses (Figure
3). There were no effects on blood pressure with any dose of
ondansetron.
RESULTS
Overview
Microinjection studies were successful in 14 rats. Specifically,
recordings of XII activity, blood pressure and end-tidal CO2 were
stable, excitatory responses in XII activity were consistent upon
local injections of either 5-HT or glutamate, and upon completion
of the experiment, Pontamine Sky Blue dye was localized to the
XII nucleus. In 12 of the 14 rats with 5-HT microinjections, 1
mM 5-HT (10nl) into the XII nucleus produced reproducible
excitatory responses to 5-HT throughout the duration of 5-HT3
drug trials. In the two rats in which CPBG was injected prior to
any 5-HT injections, glutamate produced excitatory responses
SLEEP, Vol. 24, No. 8, 2001
Effect of Systemic Injections of a 5-HT3 Agonist, CPBG
In contrast to systemic ondansetron, systemic CPBG had no
measurable effect on XII nerve activity or respiratory rate.
Systemically administered CPBG did, however, have a pronounced effect on blood pressure. At 2—3 mg/kg, CPBG pro874
5-HT3 Receptors and XII Activity—Fenik et al
rate or blood pressure.
Percentage of control XII phasic activity
250
Effect of a 5-HT3 Agonist (CPBG) Microinjection into the XIl
Nucleus
200
CPBG at concentrations of 0. 1 µM (n=2), 0.1 mM (n=6), 1
mM (n=4), 10mM (n=8) and 50 mM (n=2) had no effect on XII
nerve activity. Of the 22 injections, XII nerve activity appeared
to increase after injection in just two trials. In neither of these
animals were these results reproducible. In the two animals
receiving 50 mM CPBG prior to 5-HT injections, no visible
increase was detected. Data for all doses are presented in Table
1. 1 mM 5-HT was injected prior to all but the highest concentration of CPBG. The effect of CPBG on 5-HT excitation was
assessed by comparing 5-HT excitation at baseline to excitation
after CPBG injection. There were no changes in the magnitude
of excitation of XII motoneurons with 1mM 5-HT following any
of the doses of CPBG studied (0.1 µM to 10 mM) 72% ± 14, prior
and 74% ±17, after CPBG. In the two rats in which CPBG
microinjection (50 mM) was performed prior to any other
microinjections, 5-HT excitation of the XII nerve after CPBG
was within the range of values obtained in the other animals,
receiving 5-HT before CPBG (55% and 73% increases in XII
nerve activity in these two animals).
150
100
50
0
<0.6mg/kg
1mg/kg
2mg/kg
3mg/kg
Cumulative ondansetron dose (intravenous
Figure 3—Dose-dependence of the effects of systemically adminstered
ondansetron on the inspiratory activity in the hypoglossal (XII) nerve activity.
Data are average values with standard error bars. Significant increases from
baseline are marked with asterisks (*=p< 0.05).
Table 1—The average percentage increases from baseline in XII
nerve inspiratory activity in response to CPBG, a 5-HT3 agonist,
injected into the XII nucleus (10 nl). Paired-Student's t-values
and corresponding probabilities of the null hypothesis are also
provided.
Concentration
(Sample Size)
Mean %
change ± S.E.
t-value
P value
µM
0.1µ
(n=2)
0.1 mM 1 mM
(n=6)
(n=4)
10 mM 50 mM
(n=8)
(n=2)
1%±
1
NA
NA
5%±
16
0.72
0.50
-1%±
10
-0.02
0.99
2%±
7
0.23
0.83
Microinjections of a 5-HT3 Antagonist
Ondansetron was injected into the XII nucleus in 6 rats at concentrations of 0.1 mM and 5 mM (20 nl) with three trials at each
dose studied per animal. The effects on baseline XII activity
were not statistically significant, a 10%±5 increase from baseline
for 0.1 mM (d.f.=5), and 8%±14 increase from baseline for 5 mM
ondansetron (d.f.=4). There was no effect of ondansetron on
either the phasic or tonic 5-HT excitation of XII motoneurons;
these data are presented along with the statistics in Table 2.
5%±
15
NA
NA
DISCUSSION
We have evaluated the effects of selective 5-HT3 drugs on the
activity of inspiratory-modulated XII motoneurons. In contrast
to the robust excitatory effects of 1 mM 5-HT microinjected into
the XII nucleus, local application of 5-HT3-selective drugs neither altered the baseline activity of the XII nerve nor had any
effect on subsequent 5-HT injections. In contrast, systemic
administration of the 5-HT3 antagonist, ondansetron, significantly increased XII nerve activity. Together, these data suggest that
5-HT3 antagonists can enhance the phasic respiratory-related
activity of XII motoneurons, but the effect is not mediated to a
significant degree within the XII motor nucleus. This, in turn,
suggests that exploration of upper-airway control mechanisms
lying outside the motor nuclei may identify important therapeutic options for OSDB. Therefore, determining the site of action
of the 5-HT3 effect on respiratory drive will be important in elucidating pharmacotherapies for OSDB.
duced an initial decrease (45—150 s) followed by a long-lasting
>15-minute increase in both systolic and diastolic blood pressure
in all five rats studied (22%±3, p<0.002; increase in systolic from
120 ± 15mmHg, and an 18%±3.9, p<0.05 increase in diastolic
pressure from 72± 12 mmHg).
Effect of 5-HT Microinjected into the XII Nucleus
Serotonin (1 mM, 10 nl) produced a 75%±17 increase in XII
nerve activity (d.f. = 13, t= 14.07, p<0.0001). A typical example
of this effect is shown in Fig. 1A, with its corresponding injection site in Fig. 1B. The magnitude of the response across a typical four- to six-hour duration of experiments was constant;
74%±18 increase for the initial responses, vs. a 77%±19 increase
for the responses four to six hours later (n=12). There were significant differences in the excitatory responses for XII nucleus
injection sites placed up to 2 mm rostral to the CS versus up to 2
mm caudal to the CS. Rostral injections were of greater magnitude (110%±30 increase, n=7) compared to more caudally-placed
injection sites (32%±16 increase, n=5, p<0.05). 5-HT injected
into the XII nucleus had no apparent effect on either respiratory
SLEEP, Vol. 24, No. 8, 2001
The Absence of a Local Effect of 5-HT3 Drugs on XII Nerve
Activity
Microinjection studies of selective drugs can provide important information regarding regional receptor activity.
875
5-HT3 Receptors and XII Activity—Fenik et al
Table 2—The effects of microinjected ondansetron, a 5-HT3 antagonist, on 5-HT excitation of XII motoneurons. Shown are the effects
of locally injected 5-HT (1mM, 10 nl) before and after a 20 nl ondansetron injections of 0.1 mM (n=4 rats) and 5mM (n=4 rats).
Pre-0.1mM
Ondansetron
Post-0.1mM
Ondansetron
P value
Pre-5mM
Ondansetron
Post-5mM
Ondansetron
P value
Relative 5-HT increase
in tonic activity
59%±
34
44%±
19
P=0.41
43%±
16
49%±
19
P=0.23
Relative 5-HT increase
in inspiratory activity
78%±
20
65%±
12
P=.58
67%
±19
66%
±16
P=0.61
Interpretation of negative results, however, must be done with
caution. We believe that in the present experiments, we were
able to establish a lack of effect of 5-HT3-active drugs on XII
nerve activity, when they are injected into the XII nucleus, by
rendering the following potential causes of false negative results
unlikely.
First, we have shown activity for both of the 5-HT3-selective
drugs by documenting effects with systemic administration. The
doses effective in our study were similar to effective systemic
doses in previous reports.22,35 Systemically administered
ondansetron in the present study had a robust dose-dependent
excitatory effect on XII activity. Although the effect of this drug
on upper airway motor activity has not been previously studied,
it has been shown in rats that systemic administration of
ondansetron, in similar doses, increases minute ventilation.22 5HT3 agonists have not been reported to impact upon respiratory
activity, but have been shown to have significant effects on blood
pressure when applied to the NTS.28-30,36 We have found that, in
anesthetized rats with bilateral vagotomies, CPBG causes an initial drop in blood pressure followed by a long-lasting increase in
blood pressure. A similar effect has been shown for another 5HT3 agonist, phenylbiguanide, when injected intravenously.36
A second area of concern with negative results from microinjection experiments is the site and dosage of drugs used. To this
end, we confirmed that the injections were centered within the
XII nucleus after microinjecting Pontamine Sky Blue dye with
one of the 5-HT3-active drugs. We have also documented, before
and/or after 5-HT3-selective drug injections, an appropriate excitatory response to 5-HT microinjections. Concentrations of
drugs selected for this experiment spanned a range to cover the
expected in vitro affinities, as well as the reported effective doses
of these drugs when injected into the NTS.28-30 Therefore, neither
the site nor dose may explain the lack of effect of local 5-HT3
drugs on XII activity.
Rapid desensitization of the 5-HT3 receptor has been reported
in several populations of neurons,33 and this must be considered
as a potential cause of our negative results. The lack of any
immediate response to agonist, however, even when injected at
the very beginning of the experiment, without any pretreatment
with 5-HT, and the lack of effect of a 5-HT3 antagonist on both
baseline XII activity and on the excitatory effects of 5-HT
microinjected into the XII nucleus, suggest that 5-HT3 receptor
desensitization was not the cause of the lack of effect with 5-HT3
drugs. Thus, it appears, if there is any 5-HT3 effect, it is minimal,
and it does not contribute significantly to the excitation of XII
motoneurons by local injection of 5-HT, in this preparation.
SLEEP, Vol. 24, No. 8, 2001
Likewise CPBG did not alter the XII response to locally-injected
5-HT. Thus, even if densitization occurred, there was no impact
of the presumed desensitization to 5-HT3 receptors on the XII
response to local 5-HT, thereby confirming that the 5-HT3
response does not contribute significantly to the serotoninergic
excitation of XII motoneurons.
In performing experiments with anesthetized and paralyzed
animals, one must also consider the effects of anesthesia and paralyzing drugs on receptor activity. Microinjections of CPBG into
the NTS of urethane-anesthetized rats produce significant effects
on blood pressure,28-30 and this effect can be blocked with local
ondansetron.30 Further, when the NTS microinjections of 5-HT3selective drugs were tried in unanesthetized rats, similar doses of
the drugs were required for the effect on blood pressure as necessary in urethane anesthetized rats.29 Therefore, it is unlikely
that urethane anesthesia significantly alters 5-HT3 receptor activity in the brainstem. There is a recent report suggesting that nondepolarizing neuromuscluar blocking agents (such as pancuronium bromide used in our study) can antagonize 5-HT3 receptor
sites in vitro.37 However, in the present study, pancuronium did
not prevent a robust effect from systemic ondansetron on XII
activity. Further, there are several studies showing effects of similarly-dosed 5-HT3-active drugs having comparable effects in
pancuronium-paralyzed30 and non-paralyzed rats.28,29 Therefore,
neither the anesthesia nor paralyzing drug is likely to cause the
lack of effect of 5-HT on XII motoneurons when the drugs are
microinjected into the XII nucleus in the rat.
Are Functional 5-HT3 Receptors Present in the XII Nucleus?
The observed lack of 5-HT3 response for XII motoneurons is
rather unexpected, given the reports on the presence of 5-HT3
mRNA within the XII nucleus19 and the immunolabeling of a
fragment of the 5-HT3 receptor protein in close proximity to, if
not on, XII motoneurons.20 In fact, the brainstem motor nuclei
are some of the brain regions with the most intense immunoreactivity for the 5-HT3 receptor.20 It is important to recognize, however, that both studies detected one component of the 5-HT3
receptor, the A subunit, and therefore, cannot be regarded as sufficient evidence for the presence of an intact and functional 5HT3 receptor in XII motoneurons. Our lack of physiological
effect is consistent with ligand-binding assays, which have not
reported specific binding sites within upper-aiway dilator motor
nuclei.27,38-40 There are several possible explanations to explain
the presence of a component of the 5-HT3 receptor within the XII
nucleus and the lack of effect with locally injected 5-HT3 drugs.
876
5-HT3 Receptors and XII Activity—Fenik et al
First, autoradiography may be less sensitive than immunocytochemistry in detecting low levels of binding sites.40,41 With the
newer highly sensitive 5-HT3 ligands showing no sites in motor
nuclei,38 this potential reason for the discrepancy is unlikely. It is
also possible that the 5-HT3 receptor within hypoglossal
motoneurons is unassembled in the normal adult rat, and
becomes assembled and is rendered functional only under specific motor challenges, or conditions other than those operational
while our studies were conducted. In support of this are recent
reports showing that nerve growth factor can induce binding sites
(active receptor sites) for the 5-HT3 receptor42 and this receptor
subtype plays a role in at least one form of neuronal plasticity.43
Thus, a challenge to upper-airway dilator motoneurons may be
required to induce assembly of a functional receptor.
It is also possible that this receptor is indeed produced within
motoneurons, but rather than attaching to the somatic or proximal
dendritic membranes, the receptor may be transported from the
soma to distal dendrites or axons. There are XII dendrites within the NTS, as well as a high density of 5-HT3 receptor binding.40,44 5-HT3A immunoreactivity is associated, however, predominantly with terminals and axonal profiles. Thus, it is likely
that the 5-HT3A receptor subunit is transported distally.
Another way to reconcile the lack of physiological effect and
binding with the presence of a 5-HT3 subunit is that there may
be a unique 5-HT3 receptor on XII neurons that is structurally and
pharmacologically distinct from the 5-HT3 receptors identified to
date. There are electrophysiological data that demonstrate differences in both ion channel conductances and rectification within the rat 5-HT3 channels.22 Differences in receptor subunits
could alter the binding affinities for the established 5-HT3 drugs,
rendering these 5-HT3 drugs inactive at motoneurons. Further
information about variations in 5-HT3 receptor composition will
be necessary to determine if 5-HT3 receptors on motoneurons differ from those previously described.
receptor subtypes that active not only at the UAWDMn's but at
other regions within the brain and peripherally that may impact
on respiratory drive.
Conclusion
In summary, we have shown that 5-HT3 drugs do not significantly modulate XII nerve activity within the XII nucleus in the
rat. Findings in the present study along with observations made
with our bulldog model of sleep apnea23 and the work of Drs.
Radulovacki and Carley,22 strongly support a peripheral 5-HT3
mediated effect on upper-airway control. In contrast to the lack
of response locally, ondansetron, a 5-HT3 receptor antagonist
injected systemically produced a robust, dose-dependent excitatory effect on XII activity. An ondansetron-mediated increase in
respiratory drive to upper-airway dilators may provide a safe and
effective pharmacotherapeutic for OSDB in humans.
ACKNOWLEDGMENTS
We would like to thank Dr. Leszek Kubin for his advice, use
of his facilities and review of this manuscript.
REFERENCES
1. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The
occurrence of sleep-disordered breathing among middle-aged adults. N
Engl J Med 1993;328:1230-1235.
2. Hudgel DW, Thanakitcharu S. Pharmacologic treatment of sleepdisordered breathing. Am J Resp Crit Care Med 1998;158:691-699.
3. Remmers JE, Degroot WJ, Sauerland EK, Anch AM. Pathogenesis
of upper airway occlusion during sleep. J Appl Physiol 1978;44:931938.
4. Hendricks JC, Kline LR, Kovalski RJ, O'Brien JA, Morrison AR,
Pack AI. The English bulldog: a natural animal model of sleep-disordered breathing. J Appl Physiol 1987;63:1344-1350.
5. Kubin L, Tojima H, Davies RO, Pack AI. Serotoninergic excitatory
drive to XII motoneurons in the decerebrate cat. Neurosci Lett
1992;243-248.
6. Kubin L, Davies RO, Pack AI. Control of upper airway motoneurons during REM sleep. News Physiol Sci 1998;13:91-97.
7. Jackson DA, White SR. Receptor subtypes mediating facilitation
by
serotonin
of
excitability
of
spinal
motoneurons.
Neuropharm1990;29:787-797.
8. Al-Zubaidy ZA, Erickson RL, Greer JJ. Serotonergic and noradrenergic effects on respiratory neural discharge in the medullary slice
preparation of neonatal rats. Pflügers Arch 1996;431: 942-949.
9. Arita H, Ochiishi M. Opposing effects of 5-hydroxytryptamine on
two types of medullary inspiratory neurons with distinct firing patterns.
J Neurophysiol 1991;66:285-292.
10. Ribeiro-do-Valle LE, Metzler CE, Jacobs BL. Facilitation of masseter EMG and masseteric (jaw closure) reflex by serotonin in behaving
cats. Brain Res 1991;550:197-204.
11. Jacobs BL, Fornal CA. Serotonin and motor activity. Curr Op
Neurobio 1997;7:820-825.
12. Veasey SC, Fornal CA, Metzler CW, Jacobs BL. Response of serotonergic caudal raphe neurons in relation to specific motor activities in
freely moving cats. J Neurosci 1995;15:5346-5359.
13. Veasey SC, Panckeri KA, Hoffman EA, Pack AI, Hendricks JC.
The effects of serotonin antagonists in an animal model of sleep-disordered breathing. Am J Respir Crit Care Med 1996;153:776-786.
14. McCrimmon DR, Dekin MS, Mitchell GS. Glutamate, GABA, and
serotonin in ventilatory control. In: Dempsey JA, Pack AI, eds.
Regulation of breathing. New York: Marcel Dekker, 1995:175-218.
A Systemically Administered 5-HT3-Selective Antagonist
Enhances XII Motoneuonal Activity
High-dose systemically-administered ondansetron produced a
doubling in the amplitude of the phasic inspiratory XII nerve
activity. The site of this effect is unknown, except that it is
unlikely to occur in the XII nucleus, but there are recent data to
suggest that the mechanism of action may be through disinhibition of vagal influences on respiratory drive.45 This suggestion is
based on the finding that, when administered peripherally, 5-HT
causes a profound respiratory suppression and may increase
sleep-related apneas in rats.46 Further, 5-HT applied to the
nodose ganglion can mimic this effect,17 and finally, ondansetron
given peripherally antagonizes the effect of systemic 5-HT.45 In
our preparation, the animals were vagotomized at a level leaving
the nodose ganglion intact. Thus, the nodose ganglion may have
been a site of action for ondansetron in our model.
The observations that a 5-HT antagonist can augment upperairway activity, together with the possibility that the mechanism
is peripheral, have important clinical implications. The first is
that there are multiple serotoninergic mechanisms, beyond the
motoneurons, to explore for potential pharmacotherapeutics for
sleep apnea. Second, optimizing 5-HT pharmacotherapeutics for
OSDB will require a more extensive comprehension of 5-HT
SLEEP, Vol. 24, No. 8, 2001
877
5-HT3 Receptors and XII Activity—Fenik et al
15. Lalley PM. Bischoff AM, Schwarzacher W, Richter DW. 5-HT2
receptor-controlled modulation of medullary respiratory neurones in the
cat. J Physiol (Lond.) 1995;487:653-661.
16. Horner RL, Sanford LD, Annis D, Pack AI, Morrison AR.
Serotonin at the laterodorsal tegmental nucleus suppresses rapid-eyemovement sleep in freely behaving rats. J Neurosci 1997;17:7541-7552.
17. Yoshioka M, Goda Y, Togashi H, Matsumoto M, Saito H.
Pharmacological characterization of 5-hydroxytryptamine-induced
apnea in the rat. J Pharm Exp Therap 1992;260:917-924.
18. Fletcher S, Barnes NM. Desperately seeking subunits: are native 5HT3 receptors really homomeric complexes? Trends in Pharm Sci
1998;19: 212-215.
19. Okabe S, Mackiewicz M, Kubin L. Serotonin receptor mRNA
expression in the XII motor nucleus. Respir Physiol 1997;110:151-160.
20. Morales M, Battenberg E, Bloom EF. Distribution of neurons
expressing immunoreactivity for the 5HT3 receptor subtype in the rat
brain and spinal cord. J Comp Neuro 1998;402:385-401.
21. Radulovacki M, Trbovic SM, Carley DW. Serotonin 5-HT3-receptor antagonist GR 38032F suppresses sleep apnea in rats. Sleep
1998;21:131-136.
22. Veasey SC, Chachkes J, Fenik P, Hendricks JC. The effects of
ondansetron on sleep-disordered breathing in the English bulldog. Sleep
2001;24:155-60.
23. Veasey SC, Ogawa H, Morales M, Fenik P, Chachkes J, Kubin L.
Anatomical and physiological correlates of the 5-HT3 receptor within
the hypoglossal nucleus. Sleep 2000;23:A110.
24. Okabe S, Woch G, Kubin L. Role of GABAB receptors in the control of hypoglossal motoneurons in vivo. NeuroRep 1994;5: 25732576.
25. Sepulveda MI, Lummis SC, Martin IL. The agonist properties of mchlorophenylbiguanide and 2-methyl-5-hydroxytryptamine on 5-HT3
receptors in N1E-115 neuroblastoma cells. Br
J Pharmacol
1991;104:536-540.
26. Kilpatrick G, Jones BJ, Tyers MB. Identification and distribution of
5-HT3 receptors in rat brain using radioligand binding. Nature
1987;330:746-748.
27. Sevoz C, Nosjean A, Callera J-C, Machado B, Hamon M, Laguzzi
R. Stimulation of 5-HT3 receptors in the NTS inhibits the cardiac
Bezold-Jarisch reflex response. Am J Physiol 1996; 271:H80-81.
28. Callera J-C, Sevoz C, Laguzzi R, Machado BH. Microinjection of
a serotonin3 receptor agonist into the NTS of unanesthetized rats inhibits
the bradycardia evoked by activation of the baro- and chemoreflexes. J
Auton Nerv Sys 1997;63:127-136.
29. Sevoz C, Callera J-C, Machado BH, Hamon M, Laguzzi R. Role of
serotonin3 receptors in the nucleus tractus solitarii on the carotid
chemoreflex. Am J Physiol 1997;272:H1250-259.
30. Sevoz-Couche C, Nosjean A, Franc B, Hamon M, Laguzzi R.
Dorsal medullary 5-HT3 receptors and sympathetic premotor neurones
in the rat. J Physiol 1998;508:747-762.
31. Lummis SC, Sepulveda MI, Kilpatrick GJ, Baker J.
Characterization of [3H]meta-chlorophenylbiguanide binding to 5-HT3
receptors in N1E-115 neuroblastoma cells. Europ J Pharmacol.
1993;243:7-11.
32. Jones S, Yakel JL. Calcium influx through voltage-gated calcium
channels regulates 5-HT3 receptor channel desensitization in NG108-15
cells. Ann NY Acad Sci 1998;861:253-254.
33. Ali Z, Wu G, Kozlov A, Barasi S. The role of 5HT3 in nociceptive
processing in the rat spinal cord: results from behavioural and electrophysiological studies. Neurosci Lett 1996;208:203-207.
34. Nosjean A, Hamon M, Laguzzi R. c-Fos induction in the rostroventrolateral medulla by 5-HT3 receptor activation in the nucleus tractus solitarius. NeuroRep 1998;9:373-378.
35. Fu L-W, Longhurst JC. Reflex pressor response to arterial phenylbiguanide: role of abdominal sympathetic visceral afferents. Am J
Physiol 1998; 275:H2025-2035.
36. Min T, Wu CL, Yang J. Nondepolarizing neuromuscluar blockers
SLEEP, Vol. 24, No. 8, 2001
inhibit the serotonin-type 3A receptor expressed in Xenopus Oocytes.
Anesth Analg 2000;90:476-481.
37. Hewlett WA, Trivedi BL, Zhang ZJ, de Paulis T, Schmidt DE,
Lovinger DM, Ansari MS, Ebert MH. Characterization of (S)-des-4amino-3-[125I]iodozacopride ([125I]DAIZAC), a selective high-affinity
radioligand for 5-hydroxytryptamine3 receptors. J Pharmacol Exp
Therap 1999;288:221-231.
38. Steward LJ, West KE, Kilpatrick GJ, Barnes NM. Labelling of 5HT3 receptor recognition sites in the rat brain using the agonist radioligand [3H] meta-chlorophenylbiguanide. Europ J Pharmacol
1993;243:13-18.
39. Doucet E, Miquel MC, Nosjean A, Verge D, Hamon M, Emerit MB.
Immunolabeling of the rat central nervous system with antibodies partially selective of the short form of the 5-HT3 receptor. Neurosci
2000;95:881-892.
40. Gordon JC, Rowland HC. Nerve growth factor induces 5-HT3
recognition sites in rat pheochromocytoma (PC12) cells. Life Sci
1990;46:1435-1442.
41. Humblot N, Thiriet N, Gobaille S, Aunis D, Zwiller J. The serotonergic system modulates the cocaine-induced expression of the immediate early genes egr-1 and c-fos in rat brain. Ann NY Acad Sci
1998;844:7-20.
42. Ohuoha DC, Knable MB, Wolf SS, Kleinman JE, Hyde TM. The
subnucleuar distribution of 5-HT3 receptors in the human nucleus of the
solitary tract and other structures of the caudal medulla. Brain Res
1994;637:222-226.
43. Rende M, Morales M, Brizi E, Bruno R, Bloom F, Sanna PP.
Modulation of serotonin 5-HT3 receptor expression in injured adult rat
spinal cord motoneurons. Brain Res 1999;823:234-240.
44. Carley DW, Radulovacki M. Role of peripheral serotonin in the
regulation of central sleep apneas in rats. Chest 1999;115:1397-1401.
45. Szereda-Przestaszewska M, Wypych B. Effects of vagal and laryngeal afferents on apnoeic response to serotonin in cats. Resp Physiol
1995;101:231-237.
878
5-HT3 Receptors and XII Activity—Fenik et al