Neurogastroenterol Motil (2004) 16, 81–98
Nitrergic–purinergic interactions in rat distal colon
motility
K. VAN CROMBRUGGEN & R. A. LEFEBVRE
Heymans Institute of Pharmacology, Ghent University, Ghent, Belgium
activation of SK-channels and induces neuronal release of NO. Both nitrergic and purinergic pathways
must be blocked to inhibit EFS-induced relaxations.
Abstract Responses of rat distal colon circular muscle
strips to exogenous nitric oxide (NO) and adenosine
5¢-triphosphate (ATP) and to electrical field stimulation (EFS) were assessed in the absence/presence of
various agents that interfere with nitrergic–purinergic
pathways.
Exogenous NO (10)6 to 10)4 mol L)1) elicited concentration-dependent, tetrodotoxin (TTX)-insensitive
relaxations. The soluble guanylyl-cyclase (sGC) inhibitor 1H[1,2,4,]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ)
reduced duration and amplitude; the small conductance
Ca2+-sensitive K+ (SK)-channel blocker apamin (APA)
only shortened the relaxations. ODQ + APA showed a
marked inhibitory effect on duration and amplitude.
TTX, APA, the NO-synthase inhibitor N(omega)nitro-L-arginine methyl ester (L-NAME) and the purinergic receptor P2Y antagonist Reactive Blue 2 (RB2)
shortened the relaxations by exogenous ATP
(10)3 mol L)1) but did not influence the amplitude.
ODQ had no effect. TTX + L-NAME did not yield a
more pronounced inhibitory effect than TTX alone.
The effect of ATP-c-S was similar to that of ATP.
Electrical field stimulation (EFS) (40 V, 0.05 ms,
0.5–4 Hz for 30 s) yielded TTX-sensitive relaxations
that were not altered by L-NAME, ODQ or RB2. APA
shortened the relaxations. L-NAME + APA nearly
abolished these relaxations. ODQ + APA and RB2 +
L-NAME reduced the duration.
These results suggest that distinct sets of small
conductance SK-channels are involved in the amplitude and the duration of the relaxations and that NO
increases their sensitivity to NO and ATP via guanosine 3¢,5¢-cyclic monophosphate (cGMP). ATP elicits
relaxations via P2Y receptors with subsequent
Keywords colonic motility (distal), non-adrenergic,
non-cholinergic, nitrergic–purinergic, rat (Wistar),
small conductance Ca2+-dependant K+-channels,
smooth muscle relaxation.
INTRODUCTION
Non-adrenergic, non-cholinergic (NANC) inhibitory
neurones play an important role in the physiology of
gastrointestinal (GI) motility. Nitric oxide (NO)1,2 is
the major NANC inhibitory neurotransmitter of the GI
tract but depending upon species and part of the tract
adenosine 5¢-triphosphate (ATP) also can play a role.3
In rat distal colon, the role of NO as a mediator in
NANC neurotransmission has been addressed but this
did provide equivocal results. Several studies reported
NO to be involved in relaxation4–6 or inhibition of
spontaneous contractile activity in the longitudinal
muscle of rat distal colon,7 but others found no
evidence for NO as a mediator of these effects.8,9
Okishio et al.10 reported that the mediators of NANC
relaxations in distal colon might differ between strains
of rats: NO participates in the NANC relaxation of
longitudinal and circular muscle of distal colon in
Sprague–Dawley but not in Wistar rats. This indeed
corresponds with the above-mentioned positive/negative studies on rat distal colon longitudinal muscle.
However, for rat distal colon circular muscle, the
sparsely available studies do not support the strain
dependency of NO participation in NANC responses;
in both strains NO was reported to be involved in
inhibition of spontaneous contractile activity.11,12 In
addition, Okishio et al.10 also reported that the
NO-synthase immunoreactivity in the myenteric
plexus of Wistar and Sprague–Dawley rats was similar.
Exogenously applied ATP was reported to relax
carbachol (CCh) precontracted longitudinal muscle
Address for correspondence
Dr R. A. Lefebvre, Heymans Institute of Pharmacology, De
Pintelaan 185, 9000 Ghent, Belgium.
Tel.: +32 (0)9 240 33 74; e-mail: romain.lefebvre@ugent.be
Received: 4 February 2003
Accepted for publication: 22 July 2003
2004 Blackwell Publishing Ltd
81
K. Van Crombruggen & R. A. Lefebvre
Neurogastroenterology and Motility
muscle strips (3 · 10 mm) were cut along the circular
axis. All experimental procedures were approved by
the Ethical Committee for laboratory animals from
the Faculty of Medicine and Health Sciences at Ghent
University.
strips of Wistar rat distal colon in a concentrationdependent way.13 Other authors reported that nerve
depolarizations induced by the nicotinic receptor
agonist 1,1-Dimethyl-4-phenylpiperazinium DMPP
and by hypertonic solutions of potassium caused nonadrenergic relaxations of CCh precontracted longitudinal muscle of Sprague–Dawley rat distal colon, in
which ATP was involved.5,6 As far as we know, no data
are available on the involvement of ATP in NANC
neurotransmission of rat colon circular muscle. It was,
however, reported that ATP and NO are co-localized in
the NANC neurones of Sprague–Dawley rat colon.14
In circular muscle of Wistar rat distal colon, muscle
relaxation to NO was associated with an increase in
guanosine 3¢,5¢-cyclic monophosphate (cGMP) concentration and was inhibited by the soluble guanylate
cyclase (sGC) inhibitor methylene blue.12 In the same
tissue, the inhibition of the phasic contractile activity
by the NO donor sodium nitroprusside (SNP) was
reduced by the sGC inhibitor ODQ.15 These observations suggest that cGMP is the second messenger of
NO in this preparation. In analogy with other tissues,
this elevated cGMP production could induce relaxations via phosphorylation of cellular proteins by
cGMP-dependent protein kinases,16,17 although direct
activation of apamin (APA)-sensitive small conductance Ca2+-dependant K+ (SK)-channels by cGMP has
been suggested.18 Apamin-sensitive SK-channels are
also thought to be activated by ATP. This possible
interplay between the purinergic and nitrergic components of the NANC inhibitory responses in the GI tract
is not fully elucidated.
The aim of our study was therefore to verify (i) the
involvement of NO and ATP, (ii) their possible reciprocal interactions and (iii) transduction pathways in
inhibitory neurotransmission in Wistar rat distal colon
circular muscle.
Isometric tension recording
Muscle strips were mounted in 10 mL organ baths
between two platinum electrodes (22 · 7 mm, 6 mm
apart) for the functional experiments and in 300 lL
vials for cGMP-measurements. The organ baths contained aerated (5% CO2 in O2) PSS, maintained at
37 C in the presence of 4 · 10)6 mol L)1 guanethidine
to block noradrenergic responses. Changes in isometric
tension were measured using a Grass force-displacement transducer on a Graphtec linearcorder F WR3701
(Graphtec, Yokohama, Japan). Tissues were mounted
with an initial load of 0.25 g.
After an equilibration period of 60 min with flushing
every 15 min, the optimal length–tension relationship
was determined. Muscle strips were stretched by load
increments of 0.25 g and repeatedly exposed to
10)6 mol L)1 CCh to determine the optimal load (Lo;
the load at which maximal active tension response to
the contractile agent occurred). Active tension was
defined as the difference in total recorded tension and
the passive tension due to the load increases.
Functional study
Strips were allowed to equilibrate for 30 min at Lo with
flushing every 5 min and then precontracted with
10)4 mol L)1 methacholine. In the first part of the
experiment, relaxations were induced under control
conditions by either application of exogenous NO
(10)6 mol L)1 to 10)4 mol L)1), ATP (10)5 mol L)1 to
10)3 mol L)1 or 10)4 mol L)1 to 10)3 mol L)1), the
metabolic
stable
ATP
analogue
Ôadenosine
5¢-[c-thio]triphosphate tetralithium saltÕ (ATP-c-S;
10)4 mol L)1) and adenosine hemisulphate salt (adenosine; 10)4 mol L)1 to 10)3 mol L)1), or by electrical
field stimulation (EFS; 40 V, 0.05 and 1 ms, 0.5–4 Hz
for 30 s or 4 Hz for 30 s) via the platinum plate
electrodes by means of a Grass S88 stimulator (Grass,
W. Warwick, RI, USA). An interval of 5 min was
respected between the administration of the different
concentrations of NO or ATP, or the EFS trains at
different frequency.
In the second part of the experiment, the responses
to NO, ATP, ATP-c-S, adenosine or EFS were studied
again in the presence of one of the following drugs
(1) for NO-induced relaxations: the nerve blocker
MATERIALS AND METHODS
Tissue preparation
Male Wistar–Han rats (310–530 g) were purchased
from Janvier, Le Genest St-Isle, France. After fasting
overnight (18 h) with ad libitum access to water, rats
were killed by a blow on the head. A ±4 cm long
fragment of the distal colon was removed from 2 cm
above the pelvic brim in oral direction, opened along
the mesenteric border and pinned mucosa side up in
physiological salt solution (PSS in mmol L)1; NaCl:
118.5, KCl: 4.8, CaCl2: 1.9, MgSO4: 1.2, NaHCO3: 25,
KH2PO4: 1.2 and glucose: 10.1). The mucosa was
carefully dissected away and eight full-thickness
82
2004 Blackwell Publishing Ltd
Volume 16, Number 1, February 2004
Nitrergic–purinergic interaction in rat colon
tetrodotoxin (TTX; 3 · 10)6 mol L)1), the small conductance Ca2+-dependent K+ (SK)-channel blocker APA
(5 · 10)7 mol L)1; or 3 · 10)8 mol L)1 when indicated)
and the sGC inhibitor 1H[1,2,4,]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ; 10)5 mol L)1); (2) for ATP and
EFS-induced relaxations: TTX, APA, ODQ, the purinergic P2Y receptor antagonist Reactive Blue 2 (RB2;
3 · 10)5 mol L)1) and the NO-synthase inhibitor
N(omega)-nitro-L-arginine methyl ester (L-NAME;
3 · 10)4 mol L)1); (3) for ATP-c-S: APA; (4) for adenosine: RB2 and L-NAME.
In the final part of the experiment, the effect of the
following drug combinations was examined (1) on the
relaxations induced by NO: APA + ODQ and
ODQ + APA; (2) on the relaxations induced by ATP:
TTX + L-NAME, TTX + RB2, TTX + APA, APA +
ODQ, APA + RB2, APA + TTX, ODQ + APA, RB2 +
L-NAME, L-NAME + APA and L-NAME + RB2; (3) on
the relaxations induced by ATP-c-S: APA + ODQ; (4)
on the relaxations induced by EFS: APA + L-NAME,
APA + ODQ, APA + RB2, APA + TTX, ODQ + APA,
RB2 + L-NAME and L-NAME + APA. The respective
order in which the drugs are mentioned, corresponds to
the order of administration during the experiment; e.g.
APA + ODQ means that APA was present during the
second part of the experiment, and APA + ODQ during
the third part of the experiment.
The strips were washed between each part of the
experiment for 20 min with flushing every 5 min.
Interfering drugs were then incubated for 30 min and
methacholine was applied 10 min before the first
relaxant stimulus was applied. The reproducibility of
the responses to NO, ATP or EFS was evaluated by
running time-control strips in parallel, that did not
receive the interfering drugs or that received the
solvent of these drugs when it was not water; in all
these tissues, the responses were reproducible. The
tissue wet weight of the strips was measured after the
experiment.
as a control reference strip and was not given ATP.
Two minutes after adding ATP (or the corresponding
moment in the control strip), the tissues were snapfrozen in liquid nitrogen and stored at )80 C until
further processing. The strips were pulverized by
means of a Mikro-dismembrator U (B-Braun Biotech
International, Melsungen, Germany) and subsequently
dissolved in cold 6% trichloroacetic acid to give a
10% (w/v) homogenate. The homogenate was centrifuged at 2000 · g for 15 min at 4 C; the supernatant
was recovered and washed four times with five
volumes of water saturated diethyl ether. The aqueous extract was then dried under a stream of nitrogen
at 60 C and dissolved in a 10 times volume of assay
buffer. cGMP concentrations were determined using
an enzyme immunoassay kit (EIA Biotrak System;
Amersham Biosciences, Bucks, UK) after acetylation
of the samples and according to the manufacturer’s
instructions. The optical density (OD) was measured
with a 96-well plate reader (Biotrak II; Amersham
Biosciences) at 450 nm.
Data analysis
The amplitude and the duration of the relaxant
responses to NO, ATP and EFS were assessed. The
amplitude of the muscle responses was measured as
described in the results section and expressed as gram
of tension per cross-section area. Cross-section area
(mm2) ¼ tissue wet weight (mg)/[tissue length at Lo
(mm) · density (mg mm)3)]. The density of smooth
muscle tissue was estimated to be 1.05 mg mm)3. As
an example, in a series of eight strips, mean weight was
6.67 ± 0.88 mg, mean length at Lo was 14 ± 1.5 mm
and the cross-section area was 0.44 ± 0.04 mm2. The
duration of the responses was expressed in seconds (s).
cGMP concentrations were expressed as pmol per gram
tissue. All results were expressed as mean ± SEM.
Number (n) refers to tissues obtained from different
animals. Statistical analysis was performed by repeated
measures ANOVA followed by a Bonferroni corrected
t-test or by one-way ANOVA followed by a Bonferroni
corrected t-test when suitable. A P-value less than or
equal to 0.05 was considered to be statistically significant (GRAPHPAD, San Diego, CA, USA).
cGMP analysis
Strips were allowed to equilibrate for 30 min at Lo
with flushing every 5 min; the cGMP-specific phosphodiesterase (PDE) inhibitor zaprinast (10)5 mol L)1)
was added to the bath solution 10 min after the strips
were precontracted with 10)4 mol L)1 methacholine
and was allowed to incubate for 15 min. Two strips
were then relaxed by 10)4 mol L)1 or 10)3 mol L)1
ATP in the absence of L-NAME. A third strip was
studied in the presence of L-NAME (3 · 10)4 mol L)1)
from the beginning of the experiment and also
exposed to 10)4 mol L)1 ATP; a fourth strip was used
2004 Blackwell Publishing Ltd
Drugs used
Adenosine hemisulphate salt (adenosine; Sigma, Bornem, Belgium), adenosine 5¢-[c-thio]triphosphate tetralithium salt (ATP-c-S; Sigma), ATP (Boehringer
Mannheim, Mannheim, Germany), APA (Alomone
labs, Jerusalem, Israel), atropine (ATR; Sigma), CCh
83
K. Van Crombruggen & R. A. Lefebvre
Neurogastroenterology and Motility
duration of the relaxations induced by NO, ATP and
EFS was determined as the period calculated from
application of the respective stimuli on, till the
occurrence of the first phasic contraction.
(Fluka AG, Buchs, Switzerland), guanethidine (Sigma),
ODQ (Tocris Cookson, Bristol, UK), methacholine
(Schuchardt, Munchen, Germany), N(omega)-nitro-Larginine methyl ester (L-NAME; Sigma), RB2 (Sigma),
TTX (Alomone labs, Israel), zaprinast (a gift of RhonePoulenc; Dagenham, UK). All drugs were dissolved in
water except ODQ and zaprinast, which were dissolved
in 100% ethanol and 2% (v/v) triethanolamine respectively up to a concentration of 10)2 mol L)1. Saturated
NO solution was prepared from gas (Air Liquide,
Aalter, Belgium) as described by Kelm and Schrader.19
Responses to exogenously applied NO
Addition of exogenous NO (10)6 to 10)4 mol L)1) to
precontracted circular muscle strips of rat distal colon
produced TTX (3 · 10)6 mol L)1)-insensitive relaxations (Fig. 1A). The amplitude only moderately
increased upon increasing concentrations of NO, but
the duration markedly increased. APA (5 · 10)7
mol L)1) shortened the relaxations at all concentrations of NO but did not alter their amplitude (Figs 1A
and 2A). ODQ (10)5 mol L)1) on the other hand
reduced the amplitude and duration of the relaxations
induced by 10)5 to 10)4 mol L)1 NO and abolished the
relaxations induced by 10)6 mol L)1 NO (Fig. 2B). The
combination APA + ODQ (Fig. 2A) and ODQ + APA
(Fig. 2B) showed a more pronounced inhibitory effect
on both duration and amplitude of the relaxations
induced by 10)5 to 10)4 mol L)1 NO, compared with
APA or ODQ alone. APA (3 · 10)8 mol L)1), alone and
added to ODQ, had the same influence on NO-induced
relaxations as 5 · 10)7 mol L)1 APA (data not shown).
RESULTS
General observations
At optimal load (Lo), the tissues displayed on top of the
basal tone a variable amount of phasic activity. The
amplitude maximally reached 8.15 g mm)2. When
drugs were administered to study their influence on
the responses to NO, ATP or EFS, their influence on
tone and phasic activity during the incubation period
was evaluated; a change in tone and/or amplitude of
phasic activity was defined as an additional effect of
the drugs under study on these respective parameters
during the incubation time, compared with the spontaneous change of these parameters in parallel control
strips in the same period. The degree in effect of the
interfering drugs was variable from strip to strip but
the general tendency can be summarized as follows:
ODQ increased the tone of the strips and APA
increased the amplitude of the phasic activity; TTX
and L-NAME increased both these parameters and RB2
had no effect per se.
Applying methacholine (10)4 mol L)1) to the circular muscle strips of rat distal colon yielded contractions
comprised of an initial fast component followed by a
lower maintained tonic component with a superimposed phasic activity (Figs 1A, B and 6). The response
to methacholine was not altered by any of the abovementioned drugs or drug combinations. The mean
amplitude of the initial component was 16.85 ±
1.10 g mm)2 (n ¼ 8); the mean amplitude of the phasic
activity superimposed on the tonic component was
11.25 ± 1.10 g mm)2 (n ¼ 8), and the mean increase in
tone, measured from the baseline tone just before
administration of methacholine to the mean lower
level of five consecutive phasic contractions when the
response to methacholine was stabilized, was
2.05 ± 0.35 g mm)2 (n ¼ 8). The amplitude of the
relaxations induced by NO, ATP or EFS was defined
as the amount of relaxation below the mean lower
level of the five preceding phasic contractions. The
Responses to exogenously applied ATP, ATP-c-S
and adenosine
Exogenously applied ATP elicited concentrationdependent relaxations at 10)4 to 10)3 mol L)1.
10)5 mol L)1 ATP had no influence (Fig. 1B). Because
of the lack in response to 10)5 mol L)1 ATP, this
concentration was omitted from further experiments.
APA (5 · 10)7 mol L)1) reduced the amplitude and the
duration of the relaxations induced by 10)4 mol L)1
ATP but only shortened the relaxations induced by
10)3 mol L)1 ATP (Fig. 3A). ODQ (10)5 mol L)1)
reduced the duration of the relaxations induced by
10)4 mol L)1 ATP but had no effect on the relaxations
induced by 10)3 mol L)1 ATP (Fig. 3B). The combinations APA + ODQ (Figs 1B and 3A) and ODQ + APA
(Fig. 3B) abolished the effect of 10)4 mol L)1 ATP, but
did not influence the effect of 10)3 mol L)1 ATP more
than APA alone (Fig. 2). APA (3 · 10)8 mol L)1), alone
or in combination with ODQ, had the same influence
on ATP-induced relaxations as 5 · 10)7 mol L)1 APA
(Fig. 3C, D). In further experiments with APA vs ATP,
only 5 · 10)7 mol L)1 APA was used.
)4
L-NAME (3 · 10
mol L)1) shortened the relaxa)4
tions induced by 10 mol L)1 and 10)3 mol L)1 ATP
but did not influence the amplitude (Fig. 4A). Adding
84
2004 Blackwell Publishing Ltd
Volume 16, Number 1, February 2004
Nitrergic–purinergic interaction in rat colon
Figure 1 Original traces showing the
responses of precontracted (methacholine;
10)4 mol L)1) circular muscle strips from
rat distal colon to (A) exogenously applied
nitric oxide (NO; 10)6 to 10)4 mol L)1) in
control conditions (top), in the presence of
apamin (APA, 5 · 10)7 mol L)1; middle)
and APA + 1H[1,2,4,]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ, 10)5 mol L)1; bottom); to (B) exogenously applied adenosine
5¢-triphosphate (ATP, 10)5–10)3 mol L)1)
in control conditions (top), in the presence
of APA (5 · 10)7 mol L)1; middle) and
APA + ODQ (10)5 mol L)1; bottom).
Arrows indicate the moment of administration. .. indicates an interruption of
the trace; after the first interruption, the
paper speed was increased.
2004 Blackwell Publishing Ltd
85
K. Van Crombruggen & R. A. Lefebvre
Neurogastroenterology and Motility
APA (5 · 10)7 mol L)1) to L-NAME resulted in a more
pronounced reduction in the duration of the ATPinduced relaxations than with L-NAME alone; the
combination decreased the amplitude of the relaxations induced by 10)4 mol L)1 ATP, but still had no
effect on the amplitude of those induced by
10)3 mol L)1 ATP (Fig. 4A). RB2 (3 · 10)5 mol L)1)
almost abolished the relaxations induced by
10)4 mol L)1 ATP (two of the eight strips still showed
a minor response to 10)4 mol L)1 ATP) but only
shortened the relaxations induced by 10)3 mol L)1
ATP (Fig. 4B). When adding L-NAME to RB2, none of
the eight strips tested showed a response to
10)4 mol L)1 ATP (Fig. 4B). Adding RB2 to L-NAME
also nearly abolished the relaxations induced by
10)4 mol L)1 ATP (two of the eight strips showed a
minor response to 10)4 mol L)1 ATP) (Fig. 4C). Both
combinations (RB2 + L-NAME/L-NAME + RB2) further
reduced the duration but did not influence the amplitude of the relaxation induced by 10)3 mol L)1 ATP
compared with RB2 and L-NAME alone.
In the strips where the influence of APA and
subsequently APA + RB2 was studied, APA per se
had the same influence on the responses to ATP as
described above i.e. reduction of the amplitude and
duration of the response to 10)4 mol L)1 ATP and of
the duration of the response to 10)3 mol L)1 ATP. The
reduction of the amplitude of the relaxations induced
by 10)4 mol L)1 ATP in the presence of APA did not
reach significance, although the amplitude was
decreased for six of the eight strips (of which four
showed no response anymore in the presence of APA)
and was not changed for two strips. Adding RB2 to APA
abolished the relaxations induced by 10)4 mol L)1 ATP
and further reduced the duration of the relaxations
induced by 10)3 mol L)1 ATP (Fig. 4D). For the
responses to 10)4 mol L)1 ATP, no statistical significance was reached for the duration between the APA
and the APA + RB2 conditions because four of eight
strips already showed no response to 10)4 mol L)1 ATP
in the presence of APA alone; the responses of the
other four strips were abolished by the addition of
RB2.
TTX reduced both the amplitude and duration of the
relaxations induced by 10)4 mol L)1 ATP but only
reduced the duration of those induced by 10)3 mol L)1
ATP (Fig. 5A–C). Adding APA to TTX nearly abolished
the relaxations induced by 10)4 mol L)1 ATP (one of
the seven strips still showed a response) and further
reduced the duration of those induced by 10)3 mol L)1
ATP (Fig. 5A). Adding L-NAME to TTX did not reduce
Figure 2 Concentration–response curves indicating the
effects of (A) APA (5 · 10)7 mol L)1) and APA + ODQ
10)5 mol L)1); (B) ODQ (10)5 mol L)1) and ODQ + APA
(5 · 10)7 mol L)1) on the amplitude (top) and duration
(bottom) of concentration-dependent NO-induced relaxations
in precontracted (methacholine; 10)4 mol L)1) circular muscle
strips of rat distal colon. Data are mean ± SEM of n ¼ 7–8.
*P < 0.05, **P < 0.001: response in the presence of one drug vs
control. DP < 0.05, DDP < 0.001: response in the presence of
two drugs vs control. P < 0.05, P < 0.001: response in the
presence of two drugs vs response in the presence of one drug
(repeated measures ANOVA followed by a Bonferroni corrected
t-test).
86
2004 Blackwell Publishing Ltd
Volume 16, Number 1, February 2004
Nitrergic–purinergic interaction in rat colon
Figure 3 Amplitude and duration of relaxations induced by 10)4 and 10)3 mol L)1 ATP in precontracted (methacholine;
10)4 mol L)1) circular muscle strips of rat distal colon in control conditions and in the presence of (A) APA (5 · 10)7 mol L)1) and
APA + ODQ (10)5 mol L)1); (B) ODQ (10)5 mol L)1) and ODQ + APA (5 · 10)7 mol L)1); (C) APA (3 · 10)8 mol L)1) and
APA + ODQ (10)5 mol L)1); (D) ODQ (10)5 mol L)1) and ODQ + APA (3 · 10)8 mol L)1). Data are mean ± SEM of n ¼ 7–8.
*P < 0.05, **P < 0.001: response in the presence of one drug vs control. DP < 0.05, DDP < 0.001: response in the presence of two
drugs vs control. P < 0.05, P < 0.001: response in the presence of two drugs vs response in the presence of one drug (repeated
measures ANOVA followed by a Bonferroni corrected t-test).
2004 Blackwell Publishing Ltd
87
K. Van Crombruggen & R. A. Lefebvre
Neurogastroenterology and Motility
Figure 4 Amplitude and duration of relaxations induced by 10)4 and 10)3 mol L)1 ATP in precontracted (methacholine;
10)4 mol L)1) circular muscle strips of rat distal colon in control conditions and in the presence of (A) L-NAME (3 · 10)4 mo L)1)
and L-NAME + APA (5 · 10)7 mol L)1); (B) Reactive Blue 2 (RB2, 3 · 10)5 mol L)1) and RB2 + L-NAME (3 · 10)4 mol L)1);
(C) L-NAME (3 · 10)4 mol L)1) and L-NAME + RB2 (3 · 10)5 mol L)1); (D) APA (5 · 10)7 mol L)1) and APA + RB2
(3 · 10)5 mol L)1). Data are mean ± SEM of n ¼ 6–8. *P < 0.05, **P < 0.001: response in the presence of one drug vs control.
DP < 0.05, DDP < 0.001: response in the presence of two drugs vs control. P < 0.05: response in the presence of two drugs vs
response in the presence of one drug (repeated measures ANOVA followed by a Bonferroni corrected t-test).
the relaxations induced by 10)4 mol L)1 and
10)3 mol L)1 ATP more than TTX alone (Fig. 5B).
Adding RB2 to TTX further reduced the duration of
the relaxations induced by 10)3 mol L)1 ATP (Fig. 5C).
The combination APA + TTX resulted into a more
pronounced reduction in amplitude of the relaxations
induced by 10)4 mol L)1 ATP than with APA alone;
the duration of these relaxations was not significantly
more reduced by the combination APA + TTX in
comparison with APA alone, because the response to
10)4 mol L)1 ATP was already very short after addition
of APA (Fig. 5D). APA + TTX could not reduce the
duration of the relaxations induced by 10)3 mol L)1
ATP more than APA alone and had no influence on the
amplitude (Fig. 5D).
The mean amplitude and duration of the relaxations
induced by 10)4 mol L)1 ATP-c-S were in the same
order as those induced by 10)3 mol L)1 ATP (amplitude:
0.83 ± 0.11 g mm)2; duration: 112.3 ± 13.3 s, n ¼ 7);
APA (3 · 10)8 mol L)1) and APA + ODQ influenced
the ATP-c-S-induced responses in a similar way as the
responses to 10)3 mol L)1 ATP (data not shown).
88
2004 Blackwell Publishing Ltd
Volume 16, Number 1, February 2004
Nitrergic–purinergic interaction in rat colon
Figure 5 Amplitude and duration of relaxations induced by 10)4 and 10)3 mol L)1 ATP in precontracted (methacholine;
10)4 mol L)1) circular muscle strips of rat distal colon in control conditions and in the presence of (A) tetrodoxin (TTX,
3 · 10)6 mol L)1) and TTX + APA (5 · 10)7 mol L)1); (B) TTX (3 · 10)6 mol L)1) and TTX + L-NAME (3 · 10)4 mol L)1); (C) TTX
(3 · 10)6 mol L)1) and TTX + RB2 (3 · 10)5 mol L)1); (D) APA (5 · 10)7 mol L)1) and APA + TTX (3 · 10)6 mol L)1). Data are
mean ± SEM of n ¼ 7–8. *P < 0.05, **P < 0.001: response in the presence of one drug vs control. DP < 0.05, DDP < 0.001: response
in the presence of two drugs vs control. P < 0.05, P < 0.001: response in the presence of two drugs vs response in the presence of
one drug (repeated measures ANOVA followed by a Bonferroni corrected t-test).
10)4 mol L)1 adenosine had no effect while the
relaxations by 10)3 mol L)1 adenosine were quantitatively similar to the relaxations by 10)4 mol L)1 ATP
(amplitude: 0.81 ± 0.20 g mm)2; duration: 22.5 ± 4.9 s,
n ¼ 6). RB2 and L-NAME had a similar influence on
the relaxations induced by 10)3 mol L)1 adenosine as
on the relaxations induced by 10)4 mol L)1 ATP, i.e.
RB2 nearly abolished the relaxations induced by
10)3 mol L)1 adenosine, while L-NAME only shortened
them (data not shown).
2004 Blackwell Publishing Ltd
Responses to EFS
EFS (40 V/0.05 ms/30 s trains at 0.5–4 Hz) did not yield
relaxations at 0.5 Hz, induced a very moderate response
at 1 Hz and induced frequency-dependent relaxations at
2–4 Hz (Figs 6 and 7) that were abolished by TTX
(3 · 10)6 mol L)1). The relaxations were not influenced
by L-NAME (3 · 10)4 mol L)1) alone (Fig. 7B). APA (5 ·
10)7 mol L)1) shortened the electrically induced relaxations, but did not influence their amplitude (Figs 6 and
89
K. Van Crombruggen & R. A. Lefebvre
Neurogastroenterology and Motility
Figure 6 Original traces showing the frequency-dependent responses of precontracted (methacholine; 10)4 mol L)1)
circular muscle strips from rat distal colon
to electrical field stimulation (EFS) under
control conditions (top), in the presence of
APA (5 · 10)7 mol L)1; middle) and
APA + L-NAME 3 · 10)4 mol L)1; bottom). Bold bars represent EFS (40 V,
0.05 ms, 30 s trains) at the indicated frequencies. .. indicates an interruption of
the trace; after the first interruption, the
paper speed was increased.
L-NAME (Fig. 8A, C). Adding ODQ to APA further
reduced the duration of the relaxation compared with
APA alone but still had no effect on the amplitude
(Fig. 8B). RB2 added to APA was not able to significantly reduce the relaxations more than APA alone
(Fig. 8D). Adding TTX to APA abolished the relaxations just like TTX alone did (Fig. 8E).
EFS (40 V/1 ms/30 s trains at 0.5–4 Hz) induced
consistent frequency-dependent relaxations from
0.5 Hz on. The influence of the drugs and drug-combinations was the same as described above for EFS with
stimuli at 0.05 ms duration, except for the influence of
TTX (results not shown). The EFS-induced responses
(40 V/1 ms/0.5–4 Hz) were indeed insensitive to
3 · 10)6 mol L)1 TTX; adding TTX (3 · 10)6 mol L)1)
7A). Adding L-NAME to APA nearly abolished the
relaxations (Fig. 7A). The same effect was obtained by
adding APA to L-NAME (Fig. 7B). A similar influence
of APA alone or in combination with L-NAME
on EFS-induced relaxations was observed when
3 · 10)8 mol L)1 APA was used (data not shown). In
further experiments with APA vs EFS, only
5 · 10)7 mol L)1 APA was used.
The following drug-combinations were only tested
vs relaxations induced by EFS at 4 Hz. ODQ
(10)5 mol L)1) and RB2 (3 · 10)5 mol L)1) alone did
not influence the relaxations induced by EFS (Fig. 8A,
C). ODQ + APA significantly reduced the duration of
the EFS-induced relaxation but did not significantly
influence the amplitude; the same applied to RB2 +
90
2004 Blackwell Publishing Ltd
Volume 16, Number 1, February 2004
Nitrergic–purinergic interaction in rat colon
cGMP analysis
The cGMP concentration of the non-ATP-treated control tissues was 14.75 ± 1.16 pmol g)1 tissue (n ¼ 4).
The cGMP concentration was 24.27 ± 3.31 (P < 0.01;
n ¼ 4) and 22.02 ± 1.24 pmol g)1 tissue (not significant
different from control; n ¼ 4) in tissues relaxed by
10)4 mol L)1 and 10)3 mol L)1 ATP respectively. In
tissues treated with 10)4 mol L)1 ATP in the presence
of L-NAME, the cGMP concentration was 3.26 ±
0.15 pmol g)1 tissue (P < 0.001 vs control; n ¼ 4).
DISCUSSION
The aim of our study was to investigate the nitrergic–
purinergic interactions in Wistar rat distal colon motility. The responses to exogenously applied NO (10)6 to
10)4 mol L)1), ATP (10)4 to 10)3 mol L)1), ATP-c-S
(10)4 mol L)1) and adenosine (10)3 mol L)1) and to EFS
(40 V/0.05 ms or 1 ms/30 s trains at 0.5–4 Hz) were
studied on methacholine precontracted circular muscle
strips of rat distal colon in the absence and presence of
various agents that interfere with the nitrergic–purinergic pathways. L-NAME, ODQ and TTX increase the
tone of the strips and/or the amplitude of the phasic
activity at Lo during the incubation period. This
suggests a tonic neurogenic release of NO, inhibiting
motility via cGMP, as was already described for Wistar
rat proximal colon circular muscle,20 as well as for
distal colon longitudinal8 and circular muscle;15 this is
also corroborated by our observation of a decreased
cGMP concentration in strips that were incubated with
L-NAME compared with control strips. Although APA
increases the amplitude of the phasic activity, RB2 had
no effect per se, which indicates that, in our experimental conditions, there is no evidence for continuously released ATP. All the above-mentioned drugs had
no effect on the methacholine-induced contractions.
The concentration of APA used was in general
5 · 10)7 mol L)1. However, to evaluate a possible difference in SK-channel subtype sensitivity towards
3 · 10)8 mol L)1 vs 5 · 10)7 mol L)1 APA, a set of
experiments with NO, ATP and EFS was performed with
3 · 10)8 mol L)1 APA. As 3 · 10)8 mol L)1 APA had the
same effect as 5 · 10)7 mol L)1 APA on the relaxant
responses to NO, ATP and EFS, non-specific effects of
5 · 10)7 mol L)1 APA by blocking SK-channels that are
not blocked by 3 · 10)8 mol L)1 APA can be excluded.
Figure 7 Frequency–response curves indicating the effects of
(A) APA (5 · 10)7 mol L)1) and APA + L-NAME
(3 · 10)4 mol L)1); (B) L-NAME (3 · 10)4 mol L)1) and
)7
L-NAME + APA (5 · 10
mol L)1) on the amplitude (top) and
duration (bottom) of relaxations induced by EFS (40 V,
0.05 ms, 0.5–4 Hz for 30 s) in precontracted (methacholine;
10)4 mol L)1) circular muscle strips of rat distal colon. The
x-axis shows the frequency on a log 2-scale. Data are
mean ± SEM of n ¼ 7–8. *P < 0.05: response in the presence of
one drug vs control. DP < 0.05, DDP < 0.001: response in the
presence of two drugs vs control. P < 0.05, P < 0.001:
response in the presence of two drugs vs response in the
presence of one drug (repeated measures ANOVA followed by a
Bonferroni corrected t-test).
to APA (5 · 10)7 mol L)1) did not reduce the duration of
the relaxation by EFS (40 V/1 ms/4 Hz) more than APA
alone and did not influence the amplitude either
(Fig. 8F).
2004 Blackwell Publishing Ltd
Responses to exogenously applied NO
Different transduction pathways have been described
for the inhibitory actions of NO. Firstly, NO was
91
K. Van Crombruggen & R. A. Lefebvre
Neurogastroenterology and Motility
Figure 8 Amplitude and duration of electrical field stimulation (EFS; 40 V, 0.05 ms, 4 Hz for 30 s)-induced relaxations in precontracted (methacholine; 10)4 mol L)1) circular muscle strips of rat distal colon in control conditions and in the presence of
(A) ODQ (10)5 mol L)1) and ODQ + APA (5 · 10)7 mol L)1); (B) APA (5 · 10)7 mol L)1) and APA + ODQ (10)5 mol L)1); (C) RB2
(3 · 10)5 mol L)1) and RB2 + L-NAME (3 · 10)4 mol L)1); (D) APA (5 · 10)7 mol L)1) and APA + RB2 (3 · 10)5 mol L)1); (E) APA
(5 · 10)7 mol L)1) and APA + TTX (3 · 10)6 mol L)1); (F) amplitude and duration of EFS (40 V, 1 ms, 4 Hz for 30 s)-induced
relaxations in precontracted (methacholine; 10)4 mol L)1) circular muscle strips of rat distal colon in control conditions and in the
presence of APA (5 · 10)7 mol L)1) and APA + TTX (3 · 10)6 mol L)1). Data are mean ± SEM of n ¼ 6–8. *P < 0.05, **P < 0.001:
response in the presence of one drug vs control. DP < 0.05, DDP < 0.001: response in the presence of two drugs vs control. P < 0.05,
P < 0.001: response in the presence of two drugs vs response in the presence of one drug (repeated measures ANOVA followed
by a Bonferroni corrected t-test).
via phosphorylation of cellular proteins of the contractile apparatus by cGMP-dependent protein kinases.
Independently of sGC activation, NO might ÔdirectlyÕ
increase the open probability of SK-channels, resulting
in a membrane hyperpolarization with decreased Ca2+
influx through voltage-operated calcium-channels. Our
data do not exclude that cGMP might to some extent
also ÔdirectlyÕ activate SK-channels or make them more
sensitive to the ÔdirectÕ action of NO.
While the effects of APA and ODQ in reducing the
duration of the relaxations by NO were additive, APA
had no effect per se on the amplitude, but reduced it
further in the presence of ODQ. This indicates an
interaction between the sGC- and SK-channel-related
pathways. Different SK-channels have been described.
Fujita et al.31 reported that SK3 and SK4, but not SK1and SK2-channels are expressed in ileum and colon of
the rat; as reverse transcription-polymerase chain
reaction (RT-PCR) analysis showed faint bands for
SK1- and SK2-channels, expression of these subtypes in
rat colon cannot be excluded. SK1- and SK4-channels
reported to activate sGC with increased production of
cGMP. These increased cGMP levels could induce
relaxations via phosphorylation of cellular proteins by
cGMP-dependent protein kinases.16 Secondly, it was
suggested that NO enhances the open probability of
APA-sensitive SK-channels20,21 or APA-resistant K+channels22–24 which leads to hyperpolarization, subsequent inhibition of voltage-operated calcium-channels
and finally to relaxation by the decreased Ca2+ influx.
The activation of these K+-channels could be mediated
directly by NO22,24–26 or via cGMP production.15,27,28
Finally, NO might induce relaxations by a mechanism
independent of changes in membrane potentials and
cGMP production.29,30
In the present study, we demonstrate that in rat
distal colon circular muscle both sGC and SK-channels
are involved in relaxations induced by exogenously
applied NO. The two pathways seem to act in parallel,
as a combination of ODQ and APA is required to block
the response to NO. sGC activated by NO will yield
increased cGMP levels, which might induce relaxation
92
2004 Blackwell Publishing Ltd
Volume 16, Number 1, February 2004
Nitrergic–purinergic interaction in rat colon
studied; the response to 10)4 mol L)1 ATP was abolished. This could be explained by the possibility that
a part of the relaxation induced by ATP via P2Y
receptors might not involve the PLC-IP3 pathway
with subsequent SK-channel activation. Other transduction pathways activated by P2Y receptor stimulation have been proposed, including phospholipase D,
phospholipase A2 and mitogen-activated protein kinase.40 However, the by far most relevant PLC-independent pathway involves P2Y receptors (i.e. P2Y11)
that are positively coupled to the adenylyl cyclase
(AC)-cyclic adenosine monophosphate (cAMP) pathway.42,43 Consequently, the part of the relaxation
induced by ATP via these receptors is insensitive to
the SK-channel blocker APA but still sensitive to the
P2Y receptor blocker RB2.
When first concentrating on the response to
10)4 mol L)1 ATP, the reduction of the duration by
TTX, L-NAME and ODQ, to which the effect of APA or
RB2 added up when combined with one of these agents,
suggests that 10)4 mol L)1 ATP releases neurogenic
NO that via activation of sGC and subsequent generation of cGMP, sensitises the SK-channels and renders
them more sensitive to the Ca2+ puffs generated by
10)4 mol L)1 ATP. These SK-channels probably correspond to those described above as being permanently
APA-sensitive as APA alone is able to reduce the
duration of the relaxations by 10)4 mol L)1 ATP. ATP
has already been suggested to stimulate presynaptic
release of NO in colonic circular muscle of Sprague–
Dawley rats44 and Syrian hamsters45 and in the
ileocolonic junction of the opossum 46 and the dog.47
The increase in cGMP in strips treated with
10)4 mol L)1 ATP confirms the cGMP generation by
ATP, and corroborates that ATP induces NO release.
There cannot be excluded, however, that tonically
released NO might to some extent contribute to the
sensitization of these SK-channels.
The amplitude of the relaxations induced by
10)4 mol L)1 ATP was also reduced by TTX per se
but not by ODQ and L-NAME. This suggests the
involvement of a non-nitrergic neurotransmitter that
makes the SK-channels involved in the amplitude
more sensitive for ATP. The release of this nonnitrergic neurotransmitter might also be induced by
ATP. The amplitude of the relaxation by 10)4 mol L)1
ATP was reduced by APA per se, in contrast to the
amplitude of the relaxation by exogenous NO, where
previous inhibition of sGC is required to observe an
effect with APA. This might be related to the mechanism of action of the non-nitrergic neurotransmitter.
The characterization of the non-nitrergic neurotransmitter and its mechanism of action need thus further
have been reported to be APA-insensitive while SK2and SK3-channels are APA-sensitive.32–34 However,
others reported SK1-channels to be APA-sensitive.35,36
The ambiguous results on SK1-channel sensitivity to
APA were explained as a change in folding/assembly of
the channels due to the different expression systems
used or by additional accessory subunits expressed in
either of the cell lines that may be able to combine
with the SK1-channel and change the properties of the
channels.36
Our data suggest that possibly a set of SK-channels,
involved in the amplitude of the NO-induced relaxations, is rendered less sensitive to APA by cGMP or by
one of the cGMP-activated proteins via conformational
changes in the tertiary structure of these SK-channels,
while a permanently APA-sensitive set is involved in
the duration of the NO-induced relaxation. The relaxation by 10)6 mol L)1 NO was abolished by ODQ
per se, while the relaxations by 10)5 and 10)4 mol L)1
NO were only partially reduced by ODQ per se;
addition of APA to ODQ further reduced these
responses. This indicates that higher NO concentrations are needed for direct SK-channel activation than
for sGC stimulation.
Responses to exogenously applied ATP
ATP has also been linked to activation of APAsensitive SK-channels. The inhibitory responses to
ATP appear to involve occupation of P2Y receptors,37,38 activation of phospholipase C (PLC),
increased production of IP3 and release of Ca2+ from
internal stores.39,40 It might seem contradictory that
Ca2+ release results in inhibitory responses as increased Ca2+ levels yield contractions. It is suggested
that Ca2+-release induced by the ATP-PLC-IP3 pathway
is highly directional and causes local Ca2+-transients
without significant changes in global cytoplasmatic
Ca2+ concentration, which could provide the link
between stimulation of the purinoceptors by ATP and
the activation of the SK-channels with subsequent
inhibition of voltage-operated calcium-channels.41
As RB2 clearly reduced the amplitude and the
duration of the relaxations by 10)4 mol L)1 ATP and
the duration of the relaxations by 10)3 mol L)1 ATP,
it is obvious that the effects of exogenous ATP in the
rat distal colon circular muscle are related to interactions with P2Y receptors. The effect of APA was
similar to that of RB2, illustrating that also APAsensitive SK-channels are involved. These SK-channels are expected to be in serial relation with the
purinoceptors. Still, the effect of RB2 added up with
that of APA in the series where APA + RB2 was
2004 Blackwell Publishing Ltd
93
K. Van Crombruggen & R. A. Lefebvre
Neurogastroenterology and Motility
investigation; this is however out of the scope of this
manuscript that concentrates on nitrergic–purinergic
interactions.
The duration of the relaxations induced by
10)3 mol L)1 ATP was influenced in a similar way
as 10)4 mol L)1 ATP by the different agents tested,
except that ODQ had no influence even when added
in the presence of APA. The reduction of duration by
both TTX and L-NAME suggests that also in this
condition, neurogenic NO favours the relaxation by
ATP. The non-effect of ODQ and thus the noninvolvement of cGMP in the functional response
suggests that NO directly activates SK-channels, in
contrast to the cGMP-mediated sensitization of SKchannels with 10)4 mol L)1 ATP. This may be related
to a larger amount of NO released by the presynaptic
action of 10)3 mol L)1 ATP on the nitrergic neurones.
This is corroborated by the observation that the
absolute reduction in the duration of the 10)3 mol L)1
ATP-induced relaxations by L-NAME is larger compared with the relaxations by 10)4 mol L)1. As mentioned above, higher NO levels are needed for direct
SK-channel activation than for sGC stimulation. The
effects of this direct SK-channel activation by NO
probably surpass those of the sensitization by cGMP,
explaining the non-effect of ODQ. This conclusion is
also consistent with the observation that the increase
in cGMP levels is not more pronounced after the
application of 10)3 mol L)1 ATP than after the application of 10)4 mol L)1 ATP. The mechanism behind
the ATP-induced NO release is not known and needs
further investigation but certainly does not involve
RB2-sensitive P2Y receptors as L-NAME in the presence of RB2 further reduces the duration of
10)3 mol L)1 ATP-induced relaxations. The amplitude
of the relaxation by 10)3 mol L)1 ATP was not
sensitive to any of the drugs under study, indicating
that the mechanism behind it might differ from that
of the relaxation by 10)4 mol L)1 ATP.
The possibility that the degradation product of
ATP, adenosine, is involved in the responses to
exogenous ATP can be excluded. Indeed, the metabolically stable analogue ATP-c-S-induced similar
responses to ATP, that were influenced in the same
way by APA, alone or in combination with ODQ.
This also illustrates that the effects of APA and
ODQ on the responses to ATP are not related to
interference with the degradation of ATP. Adenosine
itself only induced relaxations at 10)3 mol L)1, which
were nearly abolished by RB2. This suggests a nonspecific action of this high concentration of adenosine with P2Y rather than with adenosine preferring
P1 receptors.
Responses to EFS
The interplay between the purinergic and the nitrergic
components of NANC inhibitory responses in the GI
tract is not fully elucidated. In rat distal colon longitudinal muscle, it was suggested that NO and ATP
may act in parallel and independently of each other as
mediators in NANC inhibitory neurotransmission
because the combined treatment of a purinergic
Figure 9 Schematic representation summarizing the different
pathways that are suggested to be involved in the responses of
precontracted rat circular muscle strips to exogenously applied nitric oxide (NO, 10)6 to 10)4 mol L)1) and adenosine
5¢-triphosphate (ATP, 10)4 to 10)3 mol L)1), and to electrical
field stimulation (EFS; 40 V, 0.05 ms, 4 Hz for 30 s). At
postsynaptic level, soluble guanylate cyclase (sGC), activated
by exogenously applied NO increases guanosine 3¢,5¢-cyclic
monophosphate (cGMP) levels, which might induce relaxation via phosphorylation of cellular proteins of the contractile
apparatus by cGMP-dependent protein kinases (PKG) (1).
Independently of sGC activation, higher concentrations of NO
ÔdirectlyÕ increase the open probability of small conductance
Ca2+-dependent K+ (SK)-channels (2), resulting in a membrane
hyperpolarization with decreased Ca2+ influx through voltageoperated calcium-channels (not depicted for clarity) and subsequent relaxation. cGMP or one of the cGMP-activated proteins might also ÔdirectlyÕ activate these channels or make
them more sensitive to the ÔdirectÕ action of NO (3), probably
via changes in the tertiary structure of the SK-channels. These
conformational changes also result in a reduced sensitivity
towards APA for the set of SK-channels, that is, involved in
the amplitude. The duration on the other hand involves a set
of SK-channels, in which the conformational changes have no
impact on the APA sensitivity and that thus remain permanently sensitive to APA.
Exogenously applied ATP (10)4 to 10)3 mol L)1) stimulates
purinergic P2Y-receptors (4) with subsequent activation of
phospholipase C (PLC), increased production of IP3 and local
release of Ca2+ from the internal stores of the sarcoplasmatic
reticulum (SR) which activates SK-channels (5) and/or activation of adenylyl cyclase (AC), increased production of cAMP
with the subsequent activation of cAMP-dependent protein
kinases (PKA) which leads to relaxation (6). ATP induced
release of NO (7) sensitizes the SK-channels involved in the
duration to the Ca2+ puffs generated by ATP, via activation of
sGC and increase of the cGMP levels at 10)4 mol L)1 ATP
(cf. mechanism 3 for exogenous NO) and via a direct action at
10)3 mol L)1 ATP (cf. mechanism 2 for exogenous NO). ATP
might at presynaptic level also release a non-nitrergic neurotransmitter X (8) that sensitizes the SK-channels involved in
the amplitude towards the Ca2+ puffs generated by ATP at
postsynaptic level.
Both NO (9) and the non-nitrergic neurotransmitter X (10)
are released in the early phase of the EFS-induced relaxation,
while NO (9) and ATP (11) are released during the sustained
phase of the EFS-induced relaxation. Hereby, NO might
inhibit the presynaptic release of ATP (12) or inhibit the P2YPLC-IP3 induced release of Ca2+ from the SR (not depicted for
clarity) via a cGMP-dependent pathway.
94
2004 Blackwell Publishing Ltd
Volume 16, Number 1, February 2004
Nitrergic–purinergic interaction in rat colon
caecum.50 Others have claimed NO to be responsible
for the APA-sensitive component of IJP in Wistar rat
colon circular muscle,20,51 canine intestine circular
muscle21 or even to be involved in both the fast and the
slow IJP component as was described for hamster
ileum circular muscle.28
In our study, EFS at 0.05 ms duration induced
relaxations of neurogenic origin as they were abolished
by TTX. As the different drugs under investigation
showed distinct effects on the amplitude and duration
of the EFS-induced relaxations, the relaxations can be
considered as divided into two phases: the first one
(characterized by the amplitude) and a second one
(characterized by the duration), in analogy to the fast
and the slow IJP components in response to NANC
receptor antagonist and a NO-synthase inhibitor
caused a significantly greater blockade of K+-induced
relaxations than either compound alone.6 Still, the
inhibitory effect did not fully add up, suggesting that
both pathways interfere with each other.
Studies, in which neurogenic transient hyperpolarizations – known as inhibitory junction potentials
(IJP) – were studied, show a fast and a slow component
of hyperpolarization in response to NANC nerve
stimulation. Some authors proposed ATP as the mediator of the fast APA-sensitive IJP component, whereas
NO is proposed as the NANC neurotransmitter
responsible for the slow APA-insensitive component
in circular muscle strips of human colon,48 guinea pig
colon,49 Sprague–Dawley rat colon40 and Wistar rat
2004 Blackwell Publishing Ltd
95
K. Van Crombruggen & R. A. Lefebvre
Neurogastroenterology and Motility
NO.14 Very recently, Matsuyama et al.45 reported a
nitrergic prejunctional inhibition of purinergic neuromuscular neurotransmission in hamster proximal
colon. Alternatively, NO might also inhibit Ca2+release from IP3-sensitive stores via a cGMP-dependent
interaction with the P2Y-PLC- IP3 pathway.55,56 As the
interactions of NO with the ATP pathway (either
presynaptic or postsynaptic) are mediated via a cGMPdependent pathway, L-NAME and ODQ will inhibit
both the relaxation induced by the postsynaptic action
of NO and its inhibitory effect on presynaptic ATP
release and/or postsynaptic Ca2+-release from IP3-sensitive stores; the relaxant function of NO is than taken
over by ATP that elicits relaxations via P2Y receptor
activation.
EFS-induced relaxations with stimuli at 1 ms duration responded in a similar way to the drug combinations as described above for EFS with stimuli at
0.05 ms duration, except that they were TTX-insensitive. Relaxations induced by EFS at 1 ms duration
might activate a TTX-insensitive subset of neurones.
Browning and Lees57 characterized the myenteric
neurones of the rat descending colon and classified
them into three groups on the basis of distinct
electrophysiological properties. The first group of
neurones (51% of all neurones) fired TTX-sensitive
action potentials in response to direct somal depolarization. The second group (40%) of neurones fired
TTX-insensitive action potentials which were followed
by long-lasting membrane after hyperpolarizations.
The final group of neurones (9%) could never be made
to fire action potentials.
nerve stimulation. The first phase involves the release
of NO in view of the effect of L-NAME in the presence
of APA but cGMP seems not to be involved as ODQ
alone or in combination with APA does not have any
effect on the amplitude of the EFS-induced relaxations.
NO is not the sole transmitter as L-NAME should
influence per se the amplitude if this was the case. The
second neurotransmitter does not interact with purinergic P2Y receptors as neither RB2 nor RB2 + L-NAME
influences the amplitude. This unknown neurotransmitter, which corresponds to the non-nitrergic neurotransmitter, that is, also released by ATP (see above),
activates SK-channels that are rendered APA-sensitive
in the absence of NO in view of the effect of APA in the
presence of L-NAME.
The second phase involves SK-channels and the
release of NO and ATP. As APA alone clearly reduced
the duration of the EFS-induced relaxations, it is
obvious that permanently APA-sensitive SK-channels
are involved. The involvement of NO and ATP is
suggested as the combination of RB2 + L-NAME significantly reduced the duration, although both compounds alone did not show any effect. This indicates
that in rat distal colon, both the nitrergic and the
purinergic pathway must be blocked to inhibit maintained EFS-induced relaxations. Similar results were
already observed by Selemidis et al.23 in guinea-pig
taenia coli. It was suggested that although NO is
co-released with a non-NO, APA-sensitive transmitter
that could be ATP, it remains functionally ÔsilentÕ
unless the effects of the other transmitter are blocked or
alternatively, the APA-sensitive neurotransmitter causes relaxation of taenia coli as well as inhibition of the
release of NO via an APA-sensitive mechanism. However, this model cannot explain why RB2 alone has no
effect on the duration of the electrically induced
relaxations in our study on rat distal colon circular
muscle strips. D’Ascenzo et al.52 and Yoshimura
et al.53 reported that NO inhibits N-channel gating
via a cGMP signalling pathway in human neuroblastoma cells and rat dorsal root ganglion neurones, and
several reports suggest that NO might be involved in
the modulation of high-voltage-activated Ca2+-channels.54 NO may in such a way suppress different
neuronal functions, including neurotransmitter
release. The contribution of NO and ATP to electrically
induced sustained relaxation in rat distal colon might
be explained as follows: NO elicits relaxation via
activation of APA-sensitive SK-channels and possibly
also via cGMP-dependent phosphorylation of proteins
of the contractile apparatus. It simultaneously inhibits
presynaptic ATP release from purinergic neurones or
from NANC neurones which co-localize ATP and
CONCLUSION
In conclusion, we showed that both NO and ATP
together with an unknown neurotransmitter are
involved as mediators of relaxation in circular muscle
of Wistar rat distal colon.
The nitrergic and purinergic components display a
complex pattern of reciprocal interactions at the level
of the SK-channels (Fig. 9). The investigation of the
responses to exogenously applied NO and ATP, under
the experimental conditions used in this study, suggests the involvement of distinct sets of SK-channels in
the amplitude and the duration of the relaxations. NO
can directly activate these SK-channels; it also renders
them more sensitive to this direct action via activation
of sGC and subsequent generation of cGMP. NO
renders the SK-channels involved in the duration
more sensitive to the Ca2+ puffs generated by ATP
while a non-nitrergic neurotransmitter sensitizes
the SK-channels involved in the amplitude towards
96
2004 Blackwell Publishing Ltd
Volume 16, Number 1, February 2004
Nitrergic–purinergic interaction in rat colon
these Ca2+ puffs elicited by ATP (Fig. 9). The SKchannels involved in the duration are permanently
sensitive to APA, while the SK-channels involved in
the amplitude become less sensitive to APA upon
sensitization.
Nitrergic and purinergic interactions are also suggested at a presynaptic level: exogenous ATP increases
the release of neuronal NO, that is, involved in
sustaining the relaxations; the responses to EFS also
suggest that NO might inhibit the release of ATP via a
cGMP-dependent pathway (Fig. 9).
11 Niklasson LG, Fasth S, Hulten L, Delbro DS. Inhibition of
nitric oxide synthase causes excitation of the circular
muscle in rat distal colon. Acta Physiol Scand 1992; 144:
489–90.
12 Middleton SJ, Cuthbert AW, Shorthouse M, Hunter JO.
Nitric oxide affects mammalian distal colonic smooth
muscle by tonic neural inhibition. Br J Pharmacol 1993;
108: 974–9.
13 Bailey SJ, Hourani SM. Effects of purines on the longitudinal muscle of the rat colon. Br J Pharmacol 1992; 105:
885–92.
14 Belai A, Burnstock G. Evidence for coexistence of ATP and
nitric oxide in non-adrenergic, non-cholinergic (NANC)
inhibitory neurones in the rat ileum, colon and anococcygeus muscle. Cell Tissue Res 1994; 278: 197–200.
15 Mulè F, D’Angelo SD, Serio R. Tonic inhibitory action
by nitric oxide on spontaneous mechanical activity in
rat proximal colon: involvement of cyclic GMP and apamin-sensitive K+ channels. Br J Pharmacol 1999; 127:
514–20.
16 Hofmann F, Ammendola A, Schlossmann J. Rising behind
NO: cGMP-dependent protein kinases. J Cell Sci 2000;
113: 1671–6 (Review).
17 Lucas KA, Pitari GM, Kazerounian S et al. Guanylyl cyclases and signaling by cyclic GMP. Pharmacol Rev 2000;
52: 375–414 (Review).
18 Esplugues JV. NO as a signalling molecule in the nervous
system. Br J Pharmacol 2002; 135: 1079–95 (Review).
19 Kelm M, Schrader J. Control of coronary vascular tone by
nitric oxide. Circ Res 1990; 66: 1561–75.
20 Serio R, Mule F, Postorino A. Nonadrenergic, noncholinergic inhibitory junction potentials in rat proximal colon:
role of nitric oxide. Can J Physiol Pharmacol 1995; 73:
79–84.
21 Christinck F, Jury J, Cayabyab F, Daniel EE. Nitric oxide
may be the final mediator of nonadrenergic, noncholinergic inhibitory junction potentials in the gut. Can J Physiol
Pharmacol 1991; 69: 1448–58.
22 Koh SD, Campbell JD, Carl A, Sanders KM. Nitric oxide
activates multiple channels in canine colonic smooth
muscle. J Physiol 1995; 489: 735–43.
23 Selemidis S, Satchell DG, Cocks TM. Evidence that NO
acts as a redundant NANC inhibitory neurotransmitter in
the guinea-pig isolated taenia coli. Br J Pharmacol 1997;
121: 604–11.
24 Watson MJ, Bywater RA, Taylor GS, Lang RJ. Effects of
nitric oxide (NO) and NO donors on the membrane conductance of circular smooth muscle cells of the guinea-pig
proximal colon. Br J Pharmacol 1996; 118: 1605–14.
25 Bolotina VM, Najibi S, Palacino JJ, Pagano PJ, Cohen RA.
Nitric oxide directly activates calcium-dependent
potassium channels in vascular smooth muscle. Nature
1994; 368: 850–3.
26 Martins SLR, De Oliveira RB, Ballejo G. Rat duodenum
nitrergic-induced relaxations are cGMP-independent and
apamin-sensitive. Eur J Pharmacol 1995; 284: 265–70.
27 Shuttleworth CWR, Sweeney KM, Sanders KM. Evidence
that nitric oxide acts as an inhibitory neurotransmitter
supplying taenia from the guinea-pig caecum. Br J Pharmacol 1999; 127: 1495–501.
28 Matsuyama H, Thapaliya S, Takewaki T. Cyclic GMPassociated apamin-sensitive nitrergic slow inhibitory
ACKNOWLEDGMENTS
This work was financially supported by grant
G.0053.02 from the Fund of Scientific Research
Flanders, and by Interuniversity Pole of Attraction
Programme P5/20 (Federal Services for Scientific,
Technical and Cultural Affairs, Belgium). The
authors thank Mr Valère Geers for the technical
assistance.
REFERENCES
1 Rand MJ, Li CG. Nitric oxide as a neurotransmitter in
peripheral nerves: nature of transmitter and mechanism of
transmission. Annu Rev Physiol 1995; 57: 659–82.
2 Stark ME, Szurszewski JH. Role of nitric oxide in gastrointestinal and hepatic function and disease. Gastroenterology 1992; 103: 1928–49.
3 Burnstock G. Overview. Purinergic mechanisms. Ann N Y
Acad Sci 1990; 603: 1–17.
4 Takahashi O. Regional differences in the nitrergic innervation between the proximal and the distal colon in rats.
Gastroenterology 1998; 115: 1504–12.
5 Börjesson L, Nordgren S, Delbro DS. DMPP causes relaxation of rat distal colon by a purinergic and a nitrergic
mechanism. Eur J Pharmacol 1997; 334: 223–31.
6 Börjesson L, Nordgren S, Delbro DS. K(+)-induced neurogenic relaxation of rat distal colon. J Pharmacol Exp Ther
1999; 291: 717–24.
7 Qian YM, Jones RL. Inhibition of rat colon contractility
by prostacyclin (IP-) receptor agonists: involvement of
NANC neurotransmission. Br J Pharmacol 1995; 115:
163–71.
8 Suthamnatpong N, Hata F, Kanada A, Takeuchi T, Yagasaki O. Mediators of nonadrenergic, noncholinergic inhibition in the proximal, middle and distal regions of rat
colon. Br J Pharmacol 1993; 108: 348–55.
9 Maehara T, Fujita A, Suthamnatpong N, Takeuchi T, Hata
F. Differences in relaxant effects of cyclic GMP on skinned
muscle preparations from the proximal and distal colon of
rats. Eur J Pharmacol 1994; 261: 163–70.
10 Okishio Y, Niioka S, Takeuchi T, Nishio H, Hata F,
Takatsuji K. Differences in mediator of nonadrenergic,
noncholinergic relaxation of the distal colon between
Wistar-ST and Sprague-Dawley strains of rats. Eur J Pharmacol 2000; 388: 97–105.
2004 Blackwell Publishing Ltd
97
K. Van Crombruggen & R. A. Lefebvre
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Neurogastroenterology and Motility
44 Pluja L, Fernandez E, Jimenez M. Neural modulation of
the cyclic electrical and mechanical activity in the rat
colonic circular muscle: putative role of ATP and NO. Br J
Pharmacol 1999; 126: 883–92.
45 Matsuyama H, El-Mahmoudy A, Shimizu Y, Takewaki T.
Nitrergic prejunctional inhibition of purinergic neuromuscular transmission in the hamster proximal colon.
J Neurophysiol 2003; 89: 2346–53.
46 Matsuda NM, Feitosa RL Jr, Lemos MC, De Oliveira RB,
Ballejo G. In vitro and in vivo effects of nitric oxide synthase inhibitors and nitric oxide inactivators on the South
American opossum ileocolonic junction. Neurogastroenterol Motil 1998; 10: 245–52.
47 Boeckxstaens GE, Pelckmans PA, Bult H, De Man JG,
Herman AG, van Maercke YM. Evidence for nitric oxide as
mediator of non-adrenergic non-cholinergic relaxations
induced by ATP and GABA in the canine gut. Br J Pharmacol 1991; 102: 434–8.
48 Keef KD, Du C, Ward SM, McGregor B, Sanders KM.
Enteric inhibitory neural regulation of human colonic
circular muscle: role of nitric oxide. Gastroenterology
1993; 105: 1009–16.
49 Zagorodnyuk V, Maggi CA. Electrophysiological evidence
for different release mechanism of ATP and NO as inhibitory NANC transmitters in guinea-pig colon. Br J Pharmacol 1994; 112: 1077–82.
50 Serio R, Mule F, Postorino A, Vetri T, Bonvissuto F.
Apamin-sensitive and -insensitive components of inhibitory junction potentials in rat caecum: role of nitric oxide.
J Auton Pharmacol 1996; 16: 183–9.
51 Serio R, Mule F, Postorino A. Non-adrenergic, non-cholinergic inhibitory responses to nerve stimulation in rat
colonic circular muscle. Exp Physiol 1992; 77: 119–27.
52 D’Ascenzo M, Martinotti G, Azzena GB, Grassi C. cGMP/
protein kinase G-dependent inhibition of N-type Ca2+
channels induced by nitric oxide in human neuroblastoma
IMR32 cells. J Neurosci 2002; 22: 7485–92.
53 Yoshimura N, Seki S, de Groat WC. Nitric oxide modulates Ca(2+) channels in dorsal root ganglion neurons
innervating rat urinary bladder. J Neurophysiol 2001; 86:
304–11.
54 Carabelli V, D’Ascenzo M, Carbone E, Grassi C. Nitric
oxide inhibits neuroendocrine Ca(V)1 L-channel gating
via cGMP-dependent protein kinase in cell-attached
patches of bovine chromaffin cells. J Physiol 2002; 541:
351–66.
55 Clementi E, Sciorati C, Riccio M, Miloso M, Meldolesi J,
Nistico G. Nitric oxide action on growth factorelicited signals. Phosphoinositide hydrolysis and [Ca2+]i
responses are negatively modulated via a cGMP-dependent
protein kinase I pathway. J Biol Chem 1995; 270: 22277–82.
56 Harvey JS, Burgess GM. Cyclic GMP regulates activation
of phosphoinositidase C by bradykinin in sensory neurons.
Biochem J 1996; 316: 539–44.
57 Browning KN, Lees GM. Myenteric neurons of the rat
descending colon: electrophysiological and correlated
morphological properties. Neuroscience 1996; 73: 1029–47.
junction potential in the hamster ileum. Br J Pharmacol
1999; 128: 830–6.
Suthamnatpong N, Hosokawa M, Takeuchi T, Hata F,
Takewaki T. Nitric oxide-mediated inhibitory response of
rat proximal colon: independence of changes in membrane
potential. Br J Pharmacol 1994; 112: 676–82.
Takeuchi T, Kishi M, Ishii T, Nishio H, Hata F. Nitric
oxide-mediated relaxation without concomitant changes
is cyclic GMP content of rat proximal colon. Br J Pharmacol 1996; 117: 1204–8.
Fujita A, Takeuchi T, Saitoh N, Hanai J, Hata F. Expression of Ca(2+)-activated K(+) channels, SK3, in the interstitial cells of Cajal in the gastrointestinal tract. Am J
Physiol Cell Physiol 2001; 281: C1727–33.
Kohler M, Hirschberg B, Bond CT et al. Small-conductance, calcium-activated potassium channels from mammalian brain. Science 1996; 273: 1709–14.
Ishii TM, Maylie J, Adelman JP. Determinants of apamin
and D-tubocurarine block in SK potassium channels. J Biol
Chem 1997; 272: 23195–200.
Vergara C, Latorre R, Marrion NV, Adelman JP. Calciumactivated potassium channels. Review. Curr Opin Neurobiol 1998; 8: 321–9 (Review).
Grunnet M, Jensen BS, Olesen SP, Klaerke DA. Apamin
interacts with all subtypes of cloned small-conductance
Ca2+-activated K+ channels. Pflugers Arch 2001; 441:
544–50.
Shah M, Haylett DG. The pharmacology of hSK1 Ca2+activated K+ channels expressed in mammalian cell lines.
Br J Pharmacol 2000; 129: 627–30.
Bultmann R, Dudeck O, Starke K. Evaluation of
P2-purinoceptor antagonists at two relaxation-mediating
P2-purinoceptors in guinea-pig taenia coli (Naunyn Schmiedebergs) . Arch Pharmacol 1996; 353: 445–51.
Zagorodnyuk V, Maggi CA. Pharmacological evidence for
the existence of multiple P2 receptors in the circular
muscle of guinea-pig colon. Br J Pharmacol 1998; 123:
122–8.
Boyer JL, Downes CP, Harden TK. Kinetics of activation of
phospholipase C by P2Y purinergic receptor agonists and
guanine nucleotides. J Biol Chem 1989; 264: 884–90.
Dubyak GR, el-Moatassim C. Signal transduction via
P2-purinergic receptors for extracellular ATP and
other nucleotides. Am J Physiol 1993; 265: C577–606
(Review).
Bayguinov O, Hagen B, Bonev AD, Nelson MT, Sanders
KM. Intracellular calcium events activated by ATP in
murine colonic myocytes. Am J Physiol Cell Physiol 2000;
279: C126–35.
Communi D, Paindavoine P, Place GA, Parmentier M,
Boeynaems JM. Pharmacological characterization of the
human P2Y11 receptor. Br J Pharmacol 1999; 128: 1199–
201.
Qi AD, Kennedy C, Harden TK, Nicholas RA. Differential
coupling of the human P2Y(11) receptor to phospholipase C and adenylyl cyclase. Br J Pharmacol 2001; 132:
318–26.
98
2004 Blackwell Publishing Ltd