Functional study on TRPV1-mediated signalling in the mouse small intestine: involvement of tachykinin receptors

Authors


Joris G. De Man, Laboratory of Experimental Medicine and Pediatrics, Division of Gastroenterology, Faculty of Medicine, University of Antwerp, Universiteitsplein 1, B-2610 Antwerp, Belgium.
Tel: +32 3820 2636; fax: +32 3 820 2567;
e-mail: joris.deman@ua.ac.be

Abstract

Abstract  Afferent nerves in the gut not only signal to the central nervous system but also provide a local efferent-like effect. This effect can modulate intestinal motility and secretion and is postulated to involve the transient receptor potential of the vanilloid type 1 (TRPV1). By using selective TRPV1 agonist and antagonists, we studied the efferent-like effect of afferent nerves in the isolated mouse jejunum. Mouse jejunal muscle strips were mounted in organ baths for isometric tension recordings. Jejunal strips contracted to the TRPV1 agonist capsaicin. Contractions to capsaicin showed rapid tachyphylaxis and were insensitive to tetrodotoxin, hexamethonium, atropine or l-nitroarginine. Capsaicin did not affect contractions to electrical stimulation of enteric motor nerves and carbachol. Tachykinin NK1, NK2 and NK3 receptor blockade by RP67580, nepadutant plus SR-142801 reduced contractions to capsaicin to a similar degree as contractions to substance P. The effect of the TRPV1 antagonists capsazepine, SB-366791, iodo-resiniferatoxin (iodo-RTX) and N-(4-tertiarybutylphenyl)-4-(3-cholorphyridin-2-yl)tetrahydropyrazine-1(2H)-carbox-amide (BCTC) was studied. Capsazepine inhibited contractions not only to capsaicin but also those to carbachol. SB-366791 reduced contractions both to capsaicin and carbachol. Iodo-RTX partially inhibited the contractions to capsaicin without affecting contractions to carbachol. BCTC concentration-dependently inhibited and at the highest concentration used, abolished the contractions to capsaicin without affecting those to carbachol. From these results, we conclude that activation of TRPV1 in the mouse intestine induces a contraction that is mediated by tachykinins most likely released from afferent nerves. The TRPV1-mediated contraction does not involve activation of intrinsic enteric motor nerves. Of the TRPV1 antagonists tested, BCTC combined strong TRPV1 antagonism with TRPV1 selectivity.

Abbreviations
β-A-NKA

[beta-Ala8]-neurokinin A (4–10)

BCTC

N-(4-Tertiarybutylphenyl)-4-(3-cholorphyridin-2-yl)tetrahydropyrazine-1(2H)-carbox-amide

EFS

electrical field stimulation

Iodo-RTX

5′-iodo-resiniferatoxin

RP67580

7,7-Diphenyl-2 [1-imino-2 (2-methoxy-phenyl)-ethyl] perhydroisoindol-4-one (3 aR, 7 aR)

l-NOARG

l-nitroarginine

TRPV1

transient receptor potential ion channel of the vanilloid type 1

TTX

tetrodotoxin

Introduction

The transient receptor potential ion channel of the vanilloid type 1 (TRPV1) is a non-selective cation channel that is a polymodal sensor of nociceptive stimuli such as low pH, noxious heat and excessive pressure. TRPV1 is abundantly expressed on afferent nerves and is regarded as the molecular integrator of painful chemical and physiological stimuli in sensory neurons. TRPV1 receptors are sensitised during inflammation. Due to the important role of afferent nerves in pathophysiological signalling in the gut, TRPV1 is an evident target for pharmacological intervention.1

Apart from signalling to the central nervous system, afferent nerves also provide a local efferent-like effect in the gastrointestinal tract by releasing neuropeptides from their nerve endings.2–5 This efferent action of afferent nerves may play a role in controlling peripheral effector mechanisms6 and in the modulation of gastrointestinal motility and secretion,7–9 in particular in pathological conditions. Release of proinflammatory neuropeptides from peripheral sensory nerve terminals may play an important role in the initiation of neurogenic inflammation in the gut wall.9 However, evidence for an efferent action of afferent nerves mainly comes from studies in the guinea pig intestine while similar mechanisms in other species remain scarcely studied (recently reviewed in10).

A natural ligand of TRPV1 is capsaicin, the pungent ingredient of chili peppers. In the isolated guinea pig small intestine, capsaicin induces a contraction that is rapidly followed by a relaxation. The contraction to capsaicin is mediated by tachykinins released from sensory nerves activating cholinergic enteric nerves5,11,12 whereas the relaxation to capsaicin is mediated primarily by calcitonin gene-related peptide (CGRP).13,14 The intestinal relaxation in response to capsaicin may however also involve non-TRPV1 pathways.15,16

Functional studies on the effect of TRPV1 activation have been hindered by the lack of availability of selective TRPV1 antagonists. Capsazepine is widely used as TRPV1 antagonist although it shows moderate potency and limited selectivity as TRPV1 blocking agent in different species. The extensive research on TRPV1 receptors in recent years has however led to the discovery of new compounds with TRPV1 antagonist properties. These compounds contribute to a better understanding of the role of TRPV1 in physiological and pathophysiological conditions, but their effect on the actions of capsaicin in the isolated intestine remains largely unstudied.

TRPV1 is abundantly expressed on peripheral nerve endings of afferent nerves in the mouse small intestine17 but the efferent action of afferent nerves has not been documented in this tissue. The biological actions of vanilloids and the pharmacological profile of TRPV1 also show striking species-related differences7,18,19 and many questions on the exact role of TRPV1 in the gut remain unanswered.1,20 It is therefore relevant to determine the efferent role of afferent nerves in the mouse intestine, also because this species is widely used in experimental studies on the role of TRPV1 on afferent nerve signalling. In the present study, we have characterized the pharmacology of capsaicin and TRPV1 antagonists and studied their effect on the efferent-like role of afferent nerves in the isolated mouse jejunum.

Material and methods

Tissue preparation

Swiss OF1 mice (25–30 g) were fasted for 24 h with free access to water prior to experimentation. Mice were anaesthetised with diethyl ether and exsanguinated from the carotid artery. The small intestine was rapidly removed and put in ice-cold aerated Krebs-Ringer solution (118.3 mmol L−1 NaCl, 4.7 mmol L−1 KCl, 1.2 mmol L−1 MgSO4, 1.2 mmol L−1 KH2PO4, 2.5 mmol L−1 CaCl2, 25 mmol L−1 NaCHO3, 0.026 mmol L−1 CaEDTA and 11.1 mmol L−1 glucose). A ∼15-cm long segment of the jejunum, located ∼5 cm from the ligament of Treitz, was used for further preparation. All experimental procedures received approval of the Committee for Medical Ethics of the University of Antwerp.

Pharmacological studies: tissue preparation and isometric tension recording

The jejunal segment was gently flushed with Krebs-Ringer solution and opened longitudinally along the mesenteric border. The mucosa was removed and muscle strips were cut in the longitudinal direction. A silk thread was attached at the upper and lower end of the muscle strips after which they were mounted in organ baths (volume 5 mL) filled with Krebs-Ringer solution (37 °C, aerated with 5% CO2/95% O2). The muscle strips were positioned in organ baths containing two platinum ring electrodes (distance in between rings: 10 mm, diameter of rings: 3 mm) for electrical stimulation of the tissue. The lower end of the muscle strip was fixed and the other end connected to a strain gauge transducer (Scaime transducers, Annemasse, France) for recording of isometric tension. After a 30-min equilibration period, strips were contracted with carbachol (0.1 μmol L−1). Carbachol was washed away from the organ bath and strips were stretched (increments of 2.5 mN). When the tone of the strips stabilised, carbachol was added again. This procedure was repeated until the contraction to carbachol was maximal. This point was taken as the point of optimal length–tension relationship.21 The tissues were allowed to equilibrate for 60 min before starting the experiment. During the equilibration period, the preparations were washed every 15 min with fresh Krebs-Ringer solution.

Experimental protocols

In a first series of experiments, concentration–response curves to capsaicin (0.01–10 μmol L−1) were constructed to find a concentration that induced submaximal contraction. To avoid that desensitisation influenced the contraction, every individual muscle strip was challenged only once with an individual concentration of capsaicin. In a second series of experiments, the effects of tetrodotoxin (TTX), hexamethonium, atropine, l-nitroarginine and tachykinin NK1, NK2 and NK3 receptor antagonists were investigated on the contractions in response to capsaicin (1 μmol L−1). The effect of the tachykinin receptor antagonists was also investigated on a dose–response curve to substance P (1–300 nmol L−1). The incubation time of the antagonists was 20 min.22 In a third series of experiments, the effect of capsaicin (1 μmol L−1) was investigated on contractions induced by electrical field stimulation (1–8 Hz, pulse duration: 1 ms, pulse train: 10 s) of enteric motor nerves and by carbachol (0.01–1 μmol L−1). In a fourth series of experiments, the effect of the TRPV1 antagonists capsazepine (1–10 μmol L−1,23), SB-366791 (3–10 μmol L−1,24,25), 5′-iodo-resiniferatoxin (iodo-RTX, 0.1 100 nmol L−1,26–28) and N-(4-tertiarybutylphenyl)-4-(3-cholorphyridin-2-yl)tetrahydropyrazine-1(2H)-carbox-amide (BCTC; 10–100 nmol L−1,29,30) was investigated on contractions to capsaicin (1 μmol L−1) and on the concentration–response curves to carbachol. The incubation time of the TRPV1 antagonists was 45 min to allow optimal penetration of the compounds in the tissue.

Solutions and drugs

The following drugs were used: [beta-Ala8]-neurokinin A (4–10) (β-A-NKA), iodo-RTX, RP67580, SB-366791 (Tocris Bioscience, Bristol, UK); BCTC (Biomol International, Tebu bio, Boechout, Belgium); septide, senktide (Calbiochem, Nottingham, UK); atropine sulphate, capsaicin, carbachol, hexamethonium (Sigma-Aldrich, St. Louis, MO, USA); capsazepine (Ascent Scientific, Weston-Super-Mare, UK). Nepadutant was kindly provided by Dr. Criscuoli, Menarini, Florence, Italy; SR 142801 was kindly provided by Sanofi-Synthelabo Recherche, Chilly-Mazarin, France. Capsaicin and nepadutant were dissolved in 50% dimethyl sulfoxide (DMSO). BCTC, capsazepine, iodo-RTX, SR 142801, RP67580 were dissolved in 100% DMSO. All other drugs were dissolved in water. Solutions were injected in the 5 mL organ bath in volumes of 5 μL. The final volumes of DMSO in the organ bath did not exceed 0.1% and this did not influence the contractions of the muscle strips.

Presentation of results and statistical analysis

The amplitude of the contractions is expressed in mN. Results are shown as mean ± SEM. The number that is given by n refers to the number of muscle strips that was used in each set of experiments. Each muscle strip in separate sets of experiments was obtained from different mice and therefore n also refers to the number of mice that was used in each set of experiments. For statistical analysis, Student’s t-test for paired or unpaired values or one way anova followed by Dunnett’s posthoc testing was used. P-values ≤0.05 were considered to be significant.

Results

Concentration response curves to capsaicin

Preliminary experiments revealed that contractions in response to capsaicin showed the typical tachyphylaxis on repeated application: a muscle strip that contracted to capsaicin (1 μmol L−1) did not respond a second time to the TRPV1 agonist, even after a prolonged resting period of 120 min. Concentration–response curves to capsaicin were therefore constructed by challenging individual muscle strips once with one singular concentration. Jejunal muscle strips contracted in response to capsaicin in a concentration-dependent manner (0.01–10 μmol L−1, Fig. 1A,B). Contractions to capsaicin developed slowly and without delay until reaching a plateau phase after which the muscle strip slowly lost tension again (Fig. 1A).

Figure 1.

 (A) Typical tracings showing the contraction in response to capsaicin (1 μmol L−1) in a control muscle strip (treated with saline) and in a muscle strip treated with iodo-resiniferatoxin (iodo-RTX), note the delayed response to capsaicin). (B) The concentration–response curve to capsaicin in isolated muscle strips of the mouse jejunum. Contractions are expressed in mN and shown as mean±SEM, n = 6–8.

Unlike observations in the guinea pig intestine, capsaicin did not induce relaxations of mouse jejunal muscle strips even when the muscle strips were precontracted with 0.3 μmol L−1 carbachol (results not shown).

Effect of capsaicin on contractions to enteric nerve stimulation and to carbachol

Electrical field stimulation of jejunal muscle strips induced frequency-dependent contractions that were abolished by the blocker of neuronal conductance TTX (3 μmol L−1) and by atropine (1 μmol L−1) (results not shown). Capsaicin (1 μmol L−1) did not affect these contractions to stimulation of enteric motor nerves. Also the direct smooth muscle contractions to carbachol, which were unaffected by TTX and blocked by atropine (results not shown), were not affected by capsaicin (Table 1).

Table 1.   Contractions to EFS of enteric motor nerves and direct smooth muscle contractions to carbachol before (control) and after treatment of jejunal muscle strips with capsaicin (1 μmol L−1)
Contraction to EFS of enteric nerves (n = 7)
  1. EFS, electrical field stimulation. Contractions are expressed in mN and shown as mean ± SEM. Student’s t-test for paired observations shows no significant differences.

 1 Hz2 Hz4 Hz8 Hz
Control2.9 ± 0.43.8 ± 0.45.1 ± 0.46.5 ± 0.5
Capsaicin2.3 ± 0.33.6 ± 0.34.6 ± 0.25.9 ± 0.3
Contraction to carbachol (n = 5)
 0.03 μmol L−10.1 μmol L−10.3 μmol L−11 μmol L−1
Control1.0 ± 0.12.1 ± 0.24.0 ± 0.46.1 ± 0.4
Capsaicin0.9 ± 0.22.6 ± 0.44.2 ± 0.36.0 ± 0.3

Effect of blockade of neurogenic, cholinergic and nitrergic pathways on the contraction to capsaicin

As explained above, contractions in response to capsaicin showed rapid desensitisation. Therefore, the effect of antagonists on the response to capsaicin was determined by comparing responses between different muscle strips that were treated with saline (controls) or the antagonist under study.

Muscle strips that were incubated with TTX (3 μmol L−1) contracted in response to capsaicin (1 μmol L−1) and these contractions were of similar amplitude as those obtained in control muscle strips that were incubated with saline (Table 2). Also incubation of the muscle strips with either hexamethonium (100 μmol L−1, nicotinic receptor antagonist), atropine (1 μmol L−1, muscarinic receptor antagonist) or l-nitroarginine (300 μmol L−1, blocker of NO synthase) did not affect the contractions to capsaicin compared with control muscle strips that were incubated with saline (Table 2).

Table 2.   Effect of treatment of muscle strips with either saline (control), tetrodotoxin (3 μmol L−1), hexamethonium (100 μmol L−1), atropine (1 μmol L−1) and L-NOARG (300 μmol L−1) on contractions in response to capsaicin (1 μmol L−1) in the mouse small intestine
 ControlTetrodotoxinHexamethoniumAtropineL-NOARG
  1. l-NOARG, l-nitroarginine. Contractions are expressed in mN and shown as mean±SEM, n = 5. One-way anova did not show significant differences.

Capsaicin4.5 ± 0.84.3 ± 0.74.9 ± 0.84.0 ± 0.84.8 ± 0.6

Effect of blockade of tachykinin receptors on the contraction to capsaicin

Given the evidence for the involvement of tachykinins in the response to capsaicin in the guinea pig ileum,11,12 we evaluated the effect of the tachykinin NK1, NK2 and NK3 receptor antagonists RP67580, nepadutant and SR 142801, respectively, on the response to capsaicin in the mouse small intestine. The optimal concentration of the tachykinin receptor antagonists was first assessed. Jejunal muscle strips contracted in response to the tachykinin NK1 receptor agonist septide (0.1 μmol L−1, Fig. 2A) and the tachykinin NK2 receptor agonist β-A-NKA (0.1 μmol L−1, Fig. 2B) but not to the tachykinin NK3 receptor agonist senktide (0.1–0.3 μmol L−1). Contractions to septide were abolished by RP67580 (2 μmol L−1) but unaffected by nepadutant (1 μmol L−1) and SR 142801 (0.3 μmol L−1) (Fig. 2A). Contractions to β-A-NKA (0.1 μmol L−1) were abolished by nepadutant (1 μmol L−1) but unaffected by RP67580 (2 μmol L−1) or SR 142801 (0.3 μmol L−1) (Fig. 2B).

Figure 2.

 Bargraphs showing the contractions in response to (A) the NK1 agonist septide (0.1 μmol L−1) and (B) the NK2 agonist β-A-NKA (0.1 μmol L−1) before (control) and after treatment of the jejunal muscle strip with the NK1 antagonist RP67580 (2 μmol L−1), the NK2 antagonist nepadutant (1 μmol L−1) and the NK3 antagonist SR 142801 (0.3 μmol L−1). Contractions are expressed in mN and shown as mean ± SEM, n = 5. *P ≤ 0.05, Student’s t-test for paired observations.

Contractions to capsaicin were significantly lower in muscle strips that were incubated with the tachykinin NK1 receptor antagonist RP67580 compared with control muscle strips incubated with saline (Fig. 3A). The NK3 antagonist SR 142801 by itself showed a tendency for inhibition while nepadutant had no effect per se (Fig. 3A). The RP67580-induced inhibition of the contraction to capsaicin was more pronounced in the presence of the NK3 antagonist SR 142801 but not of the NK2 antagonist nepadutant (Fig. 3A). The combined blockade of tachykinin NK1, NK2 plus NK3 receptors inhibited the contractions to capsaicin by ∼69% (Fig. 3A). Because complete blockade was not achieved, the effect of the tachykinin receptor antagonists was investigated on contractions to substance P. Jejunal muscle strips contracted to substance P (1–300 nmol L−1) in a dose-dependent manner (Fig. 3B). The combined blockade of NK1,2,3 receptors with RP67580 (2 μmol L−1) plus nepadutant (1 μmol L−1) plus SR 142801 (0.3 μmol L−1) shifted the dose–response curve to substance P to the right resulting in a significant decrease of the Emax (from 5.9 ± 0.4 mN to 2.7 ± 0.4 mN, P = 0.0025, n = 6) and the LogEC50 (from −7.53 ± 0.15 to −6.84 ± 0.15, P = 0.0188, n = 6) (Fig. 3B). Fig. 3 also shows that contractions of comparable amplitude to 1 μmol L−1 capsaicin and 0.1 μmol L−1 substance P were reduced to a similar extent (∼70%) by the NK1,2,3 receptor blockers.

Figure 3.

 Bargraphs showing (A) contractions in response to capsaicin (1 μmol L−1) in jejunal muscle strips treated with either saline (control) or with the NK1 antagonist RP67580 (RP, 2 μmol L−1), the NK2 antagonist nepadutant (Nep, 1 μmol L−1) and the NK3 antagonist SR 142801 (SR, 0.3 μmol L−1), either alone or in combination. Contractions are expressed in mN and shown as mean ± SEM, n = 4–6. *P < 0.05, **P < 0.01, One way anova followed by Dunnett’s posthoc test. (B) The effect of the NK1,2,3 antagonists on a dose–response curve to substance P (1–300 nmol L−1) (O, control; ▪ NK1,2,3 receptor blockade). Contractions are expressed in mN and shown as mean ± SEM, n = 6. *P < 0.05, Student’s t-test for paired observations.

Effect of TRPV1 antagonists on the contraction to capsaicin

We have compared the effect of the widely used TRPV1 antagonist capsazepine with that of the more recently developed TRPV1 antagonists SB-366791, iodo-RTX and BCTC on contractions to capsaicin and on direct smooth muscle contractions in response to carbachol.

Treatment of muscle strips with capsazepine (1–10 μmol L−1) resulted in significantly smaller contractions to capsaicin compared with muscle strips that were treated with saline (Fig. 4A). This effect of capsazepine was concentration-dependent. However, capsazepine also significantly inhibited the direct smooth muscle contractions to carbachol (Fig. 5A). Contractions of comparable amplitude in response to capsaicin (1 μmol L−1) and carbachol (0.3 μmol L−1) were inhibited by capsazepine (3 μmol L−1) by 68.9% and 43.4% respectively indicating a more pronounced effect on TRPV1-mediated contractions than on direct smooth muscle contractions to carbachol.

Figure 4.

 Bargraphs showing contractions in response to capsaicin (1 μmol L−1) in jejunal muscle strips treated with either saline (control) or with different concentrations of (A) capsazepine (CZP, 1–10 μmol L−1), (B) SB-366791 (3–10 μmol L−1), (C) iodo-RTX (0.1–100 nmol L−1) and (D) BCTC (10–100 nmol L−1). Contractions are expressed in mN and shown as mean ± SEM, n = 6–8. *P ≤ 0.05, One-way anova followed by Dunnett’s posthoc test.

Figure 5.

 Concentration–response curves to carbachol in jejunal muscle strips before (control) and after treatment of the strips with (A) capsazepine (3 μmol L−1), (B) SB-366791 (10 μmol L−1), (C) iodo-RTX (100 nmol L−1) and (D) BCTC (100 nmol L−1). Contractions are expressed in mN and shown as mean ± SEM, n = 6. *P ≤ 0.05, Student’s t-test for paired observations.

In the presence of SB-366791 (3–10 μmol L−1), contractions to capsaicin were slightly smaller compared with control muscle strips treated with saline (Fig. 4B). A comparable inhibitory effect of SB-366791 was however noted on the contractions in response to carbachol (Fig. 5B).

Treatment of muscle strips with iodo-RTX (0.1–100 nmol L−1) resulted in significantly smaller contractions to capsaicin compared with control contractions obtained in the presence of saline (Fig. 4C). This effect of iodo-RTX was concentration-dependent but complete blockade of the contraction to capsaicin was not achieved (Fig. 4C). Iodo-RTX also changed the shape of the contraction to capsaicin from an immediate into a delayed response (Fig. 1A). Iodo-RTX did not affect the concentration-response curve to carbachol (Fig. 5C).

Finally, in the presence of BCTC (10–100 nmol L−1), contractions to capsaicin were significantly smaller compared with the contraction to capsaicin in control muscle strips treated with saline (Fig. 4D). The highest concentration of BCTC that was tested (100 nmol L−1) virtually abolished the contractions to capsaicin. BCTC did not affect the direct smooth muscle contractions to carbachol (Fig. 5D).

Discussion

In this study we showed that isolated muscle strips of the mouse small intestine contract in response to capsaicin. These contractions are tachykininergic in origin and are mediated by TRPV1 but they do not involve activation of intrinsic enteric motor nerves. Of the TRPV1 antagonists tested, BCTC showed pronounced and selective inhibition of TRPV1 mediated responses in the mouse small intestine.

The contraction to capsaicin in the mouse isolated intestine showed rapid desensitisation, a typical phenomenon for TRPV1 activation that results from a decreased neurotransmitter release from sensory nerve terminals.7 Capsaicin did not affect the contractions to electrical field stimulation or carbachol. The contractions to electrical field stimulation were tetrodotoxin and atropine sensitive confirming that they are mediated by acetylcholine released from enteric cholinergic motor nerves.21 Our findings suggest that capsaicin does not affect smooth muscle contractility per se or intrinsic cholinergic neurotransmission. The latter finding is in contrast to the guinea pig intestine where capsaicin inhibits the twitch contractions to electrical field stimulation albeit through a TRPV1 independent mechanism.31

We characterized the nature of the contractile response to capsaicin. Given previous findings in the guinea pig small intestine,5,11,12,32 we investigated the involvement of tachykinin NK1, NK2 and NK3 receptors. Efficacy studies showed that contractions to the NK1 and NK2 receptor agonists septide and β-A-NKA respectively were abolished by RP67580 and nepadutant, antagonists of NK1 and NK2 receptors respectively. The NK3 receptor agonist senktide did not induce contractions per se preventing efficacy testing of the NK3 antagonist SR 142801. However, SR 142801 in the concentration that we used showed its efficacy on tachykinin NK3 receptors in intestinal preparations.33,34

NK1 receptor blockade inhibited the contraction to capsaicin and this effect was more pronounced after blockade of NK3 receptors. Blockade of NK2 receptors had no additional effect or any effect per se. The involvement of NK1 and NK3 receptors in the response to capsaicin agrees with previous findings in the guinea pig small intestine.31,33,34 The combined blockade of tachykinin NK1,2,3 receptors significantly inhibited but did not abolish the contraction to capsaicin. We therefore investigated the effect of tachykinin receptor blockade on a dose–response curve to substance P. Blockade of NK1,2,3 receptors shifted the dose–response curve to substance P to the right and reduced contractions of comparable amplitude to substance P and capsaicin to a similar extent. The effect of higher antagonist concentrations was not studied because we noticed loss of receptor subtype specificity for RP67580 (above 5 μmol L−1) and SR 142801 (above 0.5 μmol L−1). We previously observed a similar incomplete inhibition of contractions to substance P in circular muscle strips of the mouse jejeunum.22 Although the reason of this incomplete blockade is not clear, the comparable inhibitory effect of tachykinin receptor antagonists on contractions to substance P and capsaicin suggests that the main mediator of the capsaicin-induced contraction is a tachykinin, most likely substance P.

The mechanisms of the capsaicin-induced contraction were studied by blockade of nerve conductance and of cholinergic and nitrergic pathways. Tetrodotoxin, a blocker of Na+-dependent neuronal conductance, did not affect the capsaicin-induced contraction. This is in line with the finding that capsaicin-sensitive sensory nerves express high levels of tetrodotoxin-resistant Na+ currents.35 Our results are however discrepant from findings in the isolated guinea pig and rat small intestine. In these tissues, capsaicin induces the release of a sensory neurotransmitter that on its turn activates enteric motor nerves resulting in intestinal contraction and this effect is blocked by tetrodotoxin.36,37 The tetrodotoxin-insensitive contractions to capsaicin in the mouse small intestine suggest that intrinsic enteric nerves do not mediate this contraction but that the mediator released by capsaicin directly acts on smooth muscle cells.

This suggests that the NK3 receptor modulating the contraction to capsaicin is not located on enteric nerves as this signalling pathway would be blocked by tetrodotoxin. In the intestine, NK3 receptors are involved in ganglionic neuro-neuronal transmission of enteric nerves.38,39 The absence of a smooth muscle response to the NK3 receptor agonist senktide indicates that NK3 receptors are not located on smooth muscle cells in the mouse small intestine and that NK3 receptor activation by itself does not provoke neurotransmitter release from intrinsic enteric nerves. Therefore, the NK3-mediated modulation of the contraction to capsaicin in the mouse small intestine may result from activation of NK3 receptors on afferent nerve terminals in the gut wall allowing automodulation of tachykinin release from sensory nerves. This is in accordance with findings in the rat showing an NK3 receptor-mediated modulation of substance P release from spinal afferent nerve terminals.40 In addition, there is evidence that NK3 receptors on sensory neurons modulate the tachykininergic nociceptive response of rat spinal primary afferents.41,42 The TRPV1 immunoreactive nerves in the mouse small intestine are also of spinal rather than vagal origin.17

The contraction to capsaicin in the mouse intestine is also not mediated by cholinergic enteric pathways given the lack of effect of hexamethonium and atropine. This also is in contrast to guinea pig intestinal muscle where capsaicin induces tetrodotoxin-sensitive acetylcholine release from myenteric nerves.31,43 Accordingly, guinea pig intestinal contractions to capsaicin are strongly inhibited by atropine11,34,35 suggesting that the capsaicin-induced release of substance P from sensory nerves in the guinea pig intestine activates enteric excitatory cholinergic nerves. Our results in the mouse small intestine suggest that substance P, released from afferent nerves by capsaicin, directly activates smooth muscle cells without involvement of cholinergic enteric nerves.

Blockade of nitric oxide synthase inhibits the relaxation to capsaicin in mouse colon and human small intestine.44,45 Although we never observed relaxations to capsaicin in mouse intestinal muscle strips, even when strips were precontracted and after blockade of NK1,2,3 receptors, we found that the NO synthase blocker l-nitroarginine did not affect the contraction to capsaicin suggesting that NO is not involved in this response.

To study the involvement of TRPV1 in the contraction to capsaicin, we compared the effect of the widely used TRPV1 antagonist capsazepine23 with that of other TRPV1 antagonists. To study the non-specific actions on smooth muscle contractility, the effect of the TRPV1 antagonists was also investigated on direct smooth muscle contractions to the muscarinic receptor agonist carbachol.

Capsazepine inhibited contractions to capsaicin but also reduced those to carbachol. This latter effect was not as pronounced as that on responses to capsaicin confirming the value of capsazepine as TRPV1 receptor blocker. Its non-selective effects however, for instance on calcium and cholinergic pathways,46,47 limit its use in TRPV1 research. In mice, its value may even be restricted to in vitro models only.19

SB-366791 is reported to be a high potency TRPV1 antagonist24 inhibiting the capsaicin-induced release of substance P from rat tracheal sensory nerves.25 SB-366791 slightly inhibited the contraction to capsaicin in mouse intestinal muscle strips but a similar inhibitory effect was noted on the contractions to carbachol. Higher concentrations of SB-366791 (>10 μmol L−1) induced more pronounced effects on contractions to capsaicin and carbachol. The inhibitory effect of SB-366791 in the mouse intestine thus mainly resulted from a non-specific effect on smooth muscle contractile pathways.

Iodo-resiniferatoxin is proposed as a high-affinity TRPV1 antagonist26–28 blocking the capsaicin-induced activation of the cloned mouse TRPV1 receptor.19 In this study, iodo-RTX inhibited the contractions to capsaicin in the mouse intestine without affecting the responses to carbachol. Iodo-RTX however did not abolish the contraction to capsaicin. Muscle strips that were treated with iodo-RTX also showed a remarkable delayed response to capsaicin, an effect that was not observed in muscle strips treated with saline.

BCTC is an orally effective TRPV1 antagonist that is reported to have superior pharmacokinetic properties29,30 and that concentration-dependently blocks the capsaicin-induced activation of the cloned mouse TRPV1.19 In our experiments, BCTC concentration-dependently inhibited the contractions to capsaicin and, at the highest concentration used, virtually abolished the response to capsaicin without affecting the direct smooth muscle contractions to carbachol. In our hands, BCTC combined efficacious TRPV1 blocking activity with TRPV1 selectivity. BCTC also blocks the cold sensor TRPM8, a transient receptor potential channel sharing functional and pharmacological properties with TRPV1.48 TRPM8 is present in subsets of vagal afferent neurons projecting to the stomach49 but is not activated by capsaicin50,51 and thus not likely to interfere in our experimental set up.

In conclusion, our results show that capsaicin induces a tachykinin-mediated contraction in the isolated mouse small intestine. Contractions to capsaicin were abolished by the TRPV1 antagonist BCTC and inhibited but not abolished by iodo-RTX. Unlike observations in the guinea pig, the TRPV1-mediated contraction in the mouse small intestine does not involve intrinsic enteric nerve activation but most likely results from activation of smooth muscle cells by tachykinins released from sensory nerve endings. Our findings add further evidence for the involvement of TRPV1 in a local efferent-like effect of afferent nerves.

Acknowledgments

This work was supported by grant nr P5/20 of the Interuniversity Attraction Pole of the Belgian Federal Science Policy and by grant nr G.0200.05 from the Fund for Scientific Research – Flanders (FWO-Vlaanderen). This study was presented during the annual meeting of the American Gastroenterological Association in San Diego, CA, in May 2007.

Competing interest

The authors have no competing interests.

Ancillary