Peripheral Activity of a New NK₁ Receptor Antagonist TAK-637 in the Gastrointestinal Tract

Materials and Methods

Experimental Animals.

Male Dunkin-Hartley guinea pigs (350–400 g body weight) purchased from Harlan Sprague-Dawley, Inc. (Indianapolis, IN) were used in the experiments. The animals were housed two per cage under environmentally controlled conditions (21°C, 12-h light/dark cycle) with free access to food and water. All animals were allowed to acclimatize to the environment in the animal facility for at least 1 week prior to experiments. The animals were killed by sectioning the carotid arteries under isoflurane anesthesia, and intestinal tissue was harvested immediately. A total of 35 guinea pigs were used in the study. The experimental protocol was approved by the Oklahoma City Veterans Affairs Medical Center Animal Care Committee.

Isometric Contractile Activity in Isolated Intestinal Preparations.

Segments of the mid portion of the colon, the ileum (20 cm proximal to the ileocecal junction), or the terminal ileum (5–15 cm proximal to the ileocecal junction), as well as the taenia coli, were isolated and placed in ice-cold Krebs-bicarbonate solution that was continuously aerated with 95% O₂ and 5% CO₂. Muscle strips (10–12-mm long) containing both muscle layers were dissected from the mucosa-submucosa layer and oriented in the direction of the longitudinal or circular muscle to record isometric contractile activity. Strips (approximately 15-mm long), consisting of the longitudinal muscle with the myenteric plexus attached were dissected from ileal intestinal segments. The taenia coli muscle strips used were 4- to 5-mm long. In one of the experimental series, whole segment preparations (approximately 25-mm long) were dissected from the terminal ileum and mounted vertically for recording of contractile activity in the direction of the longitudinal muscle layer. Each preparation was allowed to equilibrate at zero tension for 20 min, followed by consecutive loading with 0.20-g increments until a level of optimal resting tension (To) was achieved. Resting tension was considered optimal when the contractile response to carbachol (1 μM) or 5-hydroxytryptamine (5-HT, 10 μM) ceased to increase with loading. All experiments were performed at To and isometric contractions were monitored via isometric transducers (Radnoti Glass Technology Inc., Monrovia, CA) and recorded on a Grass-7 polygraph (Grass Instrument Co., Quincy, MA).

Electrical Field Stimulation.

Neurally mediated responses were elicited by electrical field stimulation (EFS) applied via pairs of platinum wire electrodes placed parallel to the muscle strips. Submaximal contractile responses, allowing the measurement of both reduction and potentiation of the neurally mediated contractions, were considered an appropriate model to study the effects of TAK-637. Electrical stimuli were generated by a Grass-88 stimulator (Grass Instrument Co.). Contractile responses of isolated longitudinal colonic muscles were evoked by rectangular pulses (0.5 ms pulse duration) applied in 5-s train at 5- to 10-min intervals. Pulse frequencies within the train were increased from 1 to 16 Hz to induce frequency-dependent responses. An electromotor force of 50 V was found to induce submaximal (80–95% of carbachol maximum) contraction of the longitudinal colonic muscle to stimulation frequency of 16 Hz. Consequently, a constant voltage of 50 V was used with all frequencies of stimulation. Identical parameters of stimulation were used to elicit nonadrenergic, noncholinergic (NANC) responses in colonic longitudinal muscles treated with atropine (1 μM) and guanethidine (10 μM). Cholinergic neurally mediated responses were elicited in ileal longitudinal muscle-myenteric plexus strips by single-pulse stimulation (0.5 ms pulse duration) applied at a frequency of 0.1 Hz. Submaximal twitch contractions of the longitudinal muscle were obtained by increasing the voltage by 5-V increments (starting at 10 V) until a maximal response was obtained. The voltage was then reduced by 5 V. Under these conditions, the evoked contractions were equal to 70 to 80% of the maximum.

Experimental Design.

Intestinal muscle preparations that functionally express tachykinin NK₁, NK₂, or NK₃ receptors were used to characterize antagonist potency and receptor specificity of TAK-637 effects in the gastrointestinal tract. To achieve equilibrium with NK₁ receptors, each concentration of TAK-637 was added into the bath solution 30 min prior to construction of an agonist concentration-response curve or EFS. The effects of TAK-637 on concentration-response curves induced by selective tachykinin receptor agonists were investigated in the following preparations: 1) longitudinal colonic muscle showing NK₁receptor-mediated contractions to [Sar⁹,Met(O₂)¹¹]-SP or GR 73632; 2) circular colonic muscle showing NK₂ receptor-mediated contractions to GR 64349 (Maggi et al., 1994); and 3) the taenia coli muscle showing NK₃ receptor-mediated contractions to senktide (Croci et al., 1995). The blockade of nerve conduction by tetrodotoxin (TTX, 1 μM) was used to distinguish between a neuronal and non-neuronal (smooth muscle) site of action of TAK-637. The effects of TAK-637 on neurally mediated responses were evaluated in the longitudinal muscle of the colon. The action of TAK-637 at cholinergic neuromuscular junctions was studied using electrically induced twitch contractions in longitudinal muscle-myenteric plexus strips isolated from the guinea pig ileum. Finally, the effects of TAK-637 were studied in isolated segments of terminal ileum, where capsaicin (10 μM) was applied to stimulate primary afferents and induce local efferent motor responses (Bartho et al., 1982, 1999).

Data Analysis and Statistics.

Contractions were measured in millinewtons and normalized for the cross-sectional area of the muscle strip (millimeters squared). The cross-sectional area was calculated as a function of the length of the muscle at To (millimeters), the wet weight (milligrams), and smooth muscle density of 1.05 mg/mm³ (Gordon and Siegman, 1971). In the experiments designed to study the action of TAK-637 on neurally mediated or capsaicin-induced contractions of isolated intestinal segments, responses were expressed as percentage of the contraction induced by carbachol (1 μM) applied to each preparation at the beginning of the experiment. Differences between the maximal effect of the agonist in the presence and absence of TAK-637 were tested using a paired or unpaired t test, after confirming that the data showed normal distribution with equal variances. Concentration-response curves were constructed in the absence and presence of increasing concentrations of TAK-637 and analyzed fitting the data to a sigmoid dose-response curve by a nonlinear regression using the PRIZM program from GraphPad Software, Inc. (San Diego, CA). The obtained dose ratios (DRs) were further subjected to Schild analysis. An equilibrium dissociation constant (K_b) for TAK-637 was obtained as the antilog of pA₂derived from the linear regression of mean values of the log (DR−1) plotted against the negative log of TAK-637 concentration. To investigate whether the effect of TAK-637 is neurally mediated,K_b values were estimated in control or TTX-pretreated colonic longitudinal muscles using concentration-response curves to GR 73632 obtained in the absence and presence of a single concentration of 100 nM TAK-637.K_b values based on a single shift of the agonist dose-response curve were calculated by the equation:where a is the antagonist concentration and DR is the ratio of equieffective agonist concentrations (EC₅₀) determined in the presence and absence of the antagonist.

The results were presented as mean ± S.E. Statistical significance was tested using unpaired or paired t test or the nonparametric Mann-Whitney test when appropriate. Differences were considered significant at p values less than 0.05. Statistical analysis was performed using InStat (GraphPad Software, Inc.).

Solutions and Drugs.

The modified Krebs-bicarbonate buffer used in the experiments was of the following composition: 120 mM NaCl, 6 mM KCl, 1.2 mM MgCl₂, 1.2 mM NaH₂PO₄, 2.5 mM CaCl₂, 14.4 mM NaHCO₃, and 11.5 mM glucose. The following substances were purchased from Sigma-RBI (Natick, MA): [Sar⁹,Met(O₂)¹¹]-substance P, GR 73632, GR 64349, succinyl-[Asp6,N-Me-Phe8]-fragment 6–11(senktide), and MEN-10376. The putative NK₁ receptor antagonist TAK-637 was kindly provided by Takeda Chemical Company (Osaka, Japan). The peptide compounds [Sar⁹,Met(O₂)¹¹]-SP, GR 73632, GR 64349, and senktide were dissolved in distilled water, and stock solutions of 1 mM concentration were stored at −20°C. Working dilutions were prepared fresh for each experiment. TAK-637 and MEN-10376 were dissolved in dimethyl sulfoxide (DMSO). The final concentration of DMSO in the bathing solution did not exceed 0.01% and had no significant effect on muscle contractions.

Results

Antagonist Potency of TAK-637 against NK₁Receptor-Mediated Contractions.

Longitudinal muscles isolated from the guinea pig colon were studied in the presence of atropine (1 μM) to prevent cholinergically mediated contractions. Under the influence of atropine, the preparations would not respond to carbachol; therefore, the optimal tension was adjusted by examining the contractile response induced by 5-HT (10 μM). Under these experimental conditions, the selective NK₁receptor agonists [Sar⁹,Met(O₂)¹¹]-SP or GR 73632 was added cumulatively at concentrations increasing from 0.1 nM to 1 μM to induce concentration-dependent contractions. The contractile responses showed no tachyphylaxis and were reproducible after washing. However, it was interesting to discover that GR 73632-induced maximal contraction was lower in amplitude compared with [Sar⁹,Met(O₂)¹¹]-SP. Complete concentration-response curves for both [Sar⁹,Met(O₂)¹¹]-SP and GR 73632 were obtained in preparations pretreated for 30 min with the vehicle (DMSO) or increasing concentrations of TAK-637 (1–100 nM). Application of DMSO (0.1% v/v) had no significant effect on spontaneous contractile activity or the concentration-effect curves for [Sar⁹,Met(O₂)¹¹]-SP or GR 73632. TAK-637 showed no changes of spontaneous activity and did not significantly reduce the maximal responses to [Sar⁹,Met(O₂)¹¹]-SP or GR 73632 (Table 1). Nonetheless, TAK-637 induced a concentration-dependent shift of the dose-response curves to higher concentrations of [Sar⁹,Met(O₂)¹¹]-SP (Fig. 1a) and GR 73632 (Fig.2a). The EC₅₀values obtained in the absence and presence of TAK-637 are given in Table 1. Schild plot of the rightward shifts of the concentration-effect curves to [Sar⁹,Met(O₂)¹¹]-SP yielded a straight line (linear coefficient 0.97) with a slope of 1.09 ± 0.14 (95% CL, 0.81–1.23) and aK_b value of 4.7 nM (Fig. 1b). The Schild analysis of the effect of TAK-637 on GR 73632 responses yielded a straight line (linear coefficient 0.99) with a slope of 0.80 ± 0.08 (95% CL, 0.62–1.12) and a K_bvalue of 1.8 nM (Fig. 2b).

A separate set of experiments was performed to establish whether TAK-637 had a neuronal or smooth muscle site of action. The antagonist potency of TAK-637 against contractions induced by GR 73632 was studied in the presence or absence of TTX (1 μM) in colonic longitudinal muscles pretreated with atropine. A single concentration of 100 nM TAK-637 induced similar rightward shifts of the dose-response curves for GR 73632 in the untreated muscles and in the muscles treated with TTX (Fig. 3). TheK_b values of TAK-637 estimated by a single shift of the concentration-response curves obtained in the absence and presence of TTX were 3.7 and 3.5 nM, respectively.

Receptor Specificity of Effects of TAK-637.

To establish whether TAK-637 is a selective NK₁ receptor antagonist, concentration-response curves for the contractile responses induced by selective activation of NK₂ or NK₃ receptors were studied in the absence and presence of TAK-637. In colonic circular muscle that functionally expresses NK₂ receptors, TAK-637 (100 nM) had no significant effect on the concentration-response curve for the contractile effect of a selective NK₂ receptor agonist, GR 64349 (0.1 nM–1 μM). In contrast, the selective antagonist of NK₂ receptors, MEN-10376 (100 nM), caused a parallel shift of the concentration-response curve to higher concentrations with a DR of 2.6 (Fig.4a). The taenia coli preparation expresses NK₃ receptors that are activated by the selective NK₃ receptor agonist senktide, causing a contractile response. Pretreatment of the taenia coli with TAK-637 (100 nM) had no significant effect on the concentration-effect curves obtained for senktide (0.1 nM–1 μM) (Fig. 4b).

Effects of TAK-637 on Neurally Mediated Responses.

In the longitudinal muscle of the colon, EFS (0.5 ms pulse duration, 1–16 Hz pulse frequency, 5-s train duration) induced frequency-dependent responses, which were abolished by TTX (1 μM). Responses to stimulus frequencies of 1 to 4 Hz showed an initial inhibitory component developing during stimulation, which was followed by an “off” contractile response. High frequencies (8–16 Hz) of stimulation induced only contractile responses. Since the inhibitory component depends on the level of resting tension, which was relatively low (0.2–0.5 g/mm²) in colonic longitudinal muscle, the effect of TAK-637 was investigated only against the contractile response. TAK-637 (100 nM) did not alter the contractile responses to all frequencies of stimulation (Fig. 5a). In a separate set of experiments, colonic muscles were pretreated with atropine (1 μM) and guanethidine (10 μM), and EFS was applied to induce NANC responses. NANC contractile responses were of lower amplitude, compared with untreated controls, and increased with the increase in stimulus frequency. TAK-637 (100 nM) did not significantly change the NANC contractions, except for responses to the highest stimulus frequencies (8–16 Hz), which showed a modest but significant (p < 0.05) reduction by TAK-637 (Fig. 5b).

In longitudinal muscle strips isolated from the guinea pig ileum, EFS with single electrical pulses (0.5 ms) applied at a frequency of 0.1 Hz induced twitch-like contractions. These contractions are considered a classical model of cholinergic motor responses since they are abolished by atropine. Treatment with increasing concentrations of TAK-637 (100 nM or 1 μM) did not alter the amplitude of EFS-induced responses (data not shown).

Effects of TAK-637 on Capsaicin-Induced Contractions.

Capsaicin is a sensory neuron stimulant that releases SP and depletes primary sensory afferent neurons of the small nonmyelinated type associated with pain sensation. In isolated segments of terminal ileum, the initial application of capsaicin (10 μM) induced a contractile response that involves the effect of endogenous tachykinins. To avoid the effect of SP depletion, a single application of capsaicin was tested in each preparation. In untreated control preparations, the contractile response to capsaicin was 68 ± 7% of carbachol contraction, whereas pretreatment with TAK-637 (100 nM) completely prevented the development of capsaicin-induced contractions (Fig.6). In contrast, the response to capsaicin was significantly (p < 0.001) reduced, but not abolished by atropine (1 μM). The development of capsaicin-induced contraction was completely blocked by pretreatment with TTX (1 μM).

Discussion

Our results demonstrate that TAK-637 has peripheral activity acting as a selective and competitive antagonist of NK₁ receptors in the gastrointestinal tract. In the isolated longitudinal muscle of the guinea pig colon, TAK-637 showed high antagonist affinity against contractions induced by [Sar⁹,Met(O₂)¹¹]-SP or GR 73632 with K_b values in the nanomolar range. However, as indicated by theK_b values (4.7 nM versus 1.8 nM), the antagonist affinity of TAK-637 was 2.6 times higher when calculated against the contractile responses to GR 73632 compared with [Sar⁹,Met(O₂)¹¹]-SP contractile responses. Such a difference in the affinity of a single antagonist is not uncommon when comparing the effects of an NK₁ receptor antagonist against C-terminal truncated (GR 73632) and full-length ([Sar⁹,Met(O₂)¹¹]-SP) analogs of SP (Patacchini et al., 1995; Jenkinson et al., 1999). In addition, although GR 73632 and [Sar⁹,Met(O₂)¹¹]-SP had similar potency (EC₅₀ values were 12.1 and 17.0 nM, respectively) in the longitudinal colonic muscle, the amplitude of the maximal response to GR 73632 was about 2 times higher than the response to [Sar⁹,Met(O₂)¹¹]-SP. Taken together, the results suggest that TAK-637, similar to other nonpeptide antagonists (Longmore et al., 1994; Zeng and Burcher, 1994;Patacchini et al., 1995), may show variable affinity for a single NK₁ receptor population, depending on the agonist with which it competes. In the broader context of drug-receptor interaction, these discrepancies suggest that GR 73632 and [Sar⁹,Met(O₂)¹¹]-SP bind differentially to the NK₁ receptor in the nanomolar range.

In addition to inducing a direct contractile effect, tachykinins have been shown to regulate intestinal activity via neuronal NK₂ (Zagorodnyuk et al., 1995), NK₃ (Maggi et al., 1993a), and NK₁ (Johnson et al., 1998; Lecci et al., 1999) receptors. In the current study, experiments were designed to differentiate between the neuronally mediated and direct effects of TAK-637. The DR and K_b calculated for a single shift of the dose-response curves to GR 73632 produced by TAK-637 (100 nM) in colonic longitudinal muscles pretreated with TTX were similar to the respective values obtained in untreated preparations. These findings imply that in the colonic longitudinal muscle, TAK-637 antagonizes the effect of a non-neuronal NK₁ receptor. Moreover, we have demonstrated that TAK-637 is a selective antagonist interacting with NK₁ but not NK₂ or NK₃ receptors. TAK-637 was ineffective against the contractile responses of circular colonic muscle to a NK₂ receptor agonist GR 64349 or against the NK₃ receptor-mediated contractions of the taenia coli induced by senktide. In summary, TAK-637 was defined as a competitive antagonist interacting selectively with smooth muscle NK₁ receptors causing contraction in the guinea pig colon.

Electrical field stimulation was used to further characterize the peripheral action of TAK-637 in the gastrointestinal tract by investigating its effect on neuromuscular motor responses. However, despite the fact that TAK-637 antagonizes the effect of exogenous SP in the longitudinal muscle of the colon, a significant inhibitory effect of TAK-637 was found only against NANC contractile responses induced by high frequencies of EFS. Also, the cholinergic twitch contractions induced by EFS in the longitudinal muscle-myenteric plexus strips were unaffected by TAK-637. The discrepancy between the effect of TAK-637 on responses induced by exogenous SP and the lack of a complete inhibitory effect on motor neurally mediated responses could be explained by the method of intramural nerve stimulation. Responses elicited by EFS are the net effect of stimulation of all intramural nerve terminals sensitive to the applied stimulus, and the muscle response reflects exclusively the immediate release of neurotransmitters at the neuroeffector, i.e., neuromuscular junction. Thus, the negative results obtained with TAK-637 suggest that TAK-637 has no effect on the motor neurotransmission at cholinergic and NANC neuromuscular junctions, at which SP is not an immediate neurotransmitter. A small portion of the contraction in response to NANC stimulation at higher frequency was blocked by TAK-637, indicating that TAK-637 may selectively inhibit the component of neuromuscular motor responses that is mediated by NK₁ receptors.

A capsaicin-stimulated component of intestinal sensory-motor reflexes is maintained in the isolated intestine and was used as a model to study the effects of TAK-637. In the guinea pig ileum, capsaicin stimulates primary afferent nerve terminals, which in turn activate cholinergic (and to a smaller extent noncholinergic) motor neurons, causing a net contractile response (Bartho et al., 1982). Since the capsaicin-induced contraction was followed by a rapid desensitization of capsaicin-sensitive sensory neurons, a single response was induced in either control or TAK-637-pretreated segments of terminal ileum. The striking lack of response in TAK-637-treated preparations compared with the presence of contractile response in vehicle-treated control preparations clearly indicated that TAK-637 acts as an inhibitor of a sensory-motor response in the ileum, which requires an intact NK₁ receptor mediated pathway. However, the source of receptive input to the capsaicin-sensitive fibers, i.e., the functional role of the reflex(es) that could be modulated by TAK-637, requires further investigation. In addition, we found that cholinergic motor neurons blocked by atropine are responsible for a major portion of the capsaicin-induced contractile response, since the capsaicin-induced contraction was reduced by approximately 80% in the presence of atropine. A small portion (approximately 20%) of the contractile responses to capsaicin, which was also inhibited by TAK-637, is due to the effect of a noncholinergic motor transmitter(s) released simultaneously with acetylcholine. The nature of the excitatory noncholinergic response was not identified in the present experiments. In contrast to a recent finding by Bartho et al. (1999), we did not find a “nontachykininergic” component of the capsaicin-induced excitatory response of the small intestine. The ability of TAK-637 to inhibit capsaicin-induced response supports the suggestion that NK₁ receptor antagonists have a therapeutic potential to inhibit effects mediated by enteric neurons and to block local sensory responses at the periphery. These effects of TAK-637 may be of particular benefit in the treatment of functional gastrointestinal disorders.