Abstract
The intestinal secretory actions of the proinflammatory peptide kallidin (lysyl-bradykinin) are mediated partially by enteric neurons. We hypothesized that kallidin produces neurogenic anion secretion through opioid- and cannabinoid-sensitive enteric neural pathways. Changes in short-circuit current (Isc) across sheets of porcine ileal mucosa-submucosa mounted in Ussing chambers were measured in response to kallidin (1 μM) or drugs added to the contraluminal bathing medium. Kallidin transiently increased Isc, an effect reduced after inhibition of neuronal conduction by 0.1 μM saxitoxin, cyclooxygenase inhibition by 10 μM indomethacin, or kinin B2 receptor blockade by 1 μMd-arginyl-l-arginyl-l-prolyl-trans-4-hydroxy-l-prolylglycyl-3-(2-thienyl)-l-alanyl-l-seryl-d-1,2,3,4-tetrahydro-3-isoquinolinecarbonyl-l-(2α,3β,7αβ)-octahydro-1H-indole-2-carbonyl-l-arginine (HOE-140). Its action was dependent upon extracellular Cl−or HCO
The intestinal mucosa has evolved a diverse array of innate and acquired mechanisms to protect the vast surface area it encompasses from infection. In the early stages of infection and tissue injury, the proinflammatory peptide kallidin (lysyl-bradykinin) is produced by the kallikrein-catalyzed cleavage of low molecular weight kininogen (Kaplan et al., 2002). Kallikrein-type proteases have been localized in mast cells and goblet cells along the length of the intestinal tract (Hinterleitner and Powell, 1991). Kallidin and its des-lysyl homolog bradykinin act as agonists at kinin B1 and B2 receptors, which are members of the G protein-coupled receptor superfamily (Regoli et al., 2001). The kinin B2 receptor is constitutively expressed, whereas the B1 receptor is induced by proinflammatory cytokines released in the course of tissue injury (Couture et al., 2001). These peptides produce pain, vasodilatation, and in the digestive tract, evoke active, transepithelial chloride secretion in the small intestine and colon (Gaginella and Kachur, 1989; Cuthbert and Huxley, 1998) or transepithelial bicarbonate secretion in gallbladder and duodenum (Baird and Margolius, 1989; Chen et al., 1997). The actions of kinins on ion transport are attributed to their combined effects on enteric neurons and non-neuronal cells, including intestinal epithelial cells. Moreover, these peptides can induce the formation of arachidonic acid metabolites that in turn act upon both neurons and enterocytes. For example, kinins activate primary afferent nerves in peripheral tissues, including the small intestine, through direct effects on neurons and the formation of eicosanoids such as prostaglandin E2 and 12-lipoxygenase metabolites (Maubach and Grundy, 1999; Shin et al., 2002).
Natural and synthetic opioids can alleviate diarrhea and produce constipation, actions that have been attributed in part to their intestinal antipropulsive and antisecretory actions. In most species examined, the latter effect is mediated by inhibitory δ-opioid receptors expressed in submucosal neuronal circuits that are linked to active anion secretion (DeLuca and Coupar, 1996). Indeed, δ-opioid receptor immunoreactivity has been colocalized with immunoreactivity to the cholinergic neural marker choline acetyltransferase in submucosal neurons and nerve fibers of the porcine ileum. Subpopulations of these δ-opioid receptor-positive neurons also express immunoreactivities for the sensory neural markers calcitonin gene-related peptide and vanilloid VR1 receptor (Poonyachoti et al., 2002). In addition, the selective δ-opioid agonist [d-Pen2,5]-enkephalin (DPDPE) inhibits active, neurogenic anion secretion mediated by type 2 proteinase-activated receptors, H1-histamine receptors, and serotonin receptors in muscle-stripped sheets of porcine ileal mucosa (Green et al., 2000; Poonyachoti and Brown, 2001; Green and Brown, 2002). These studies suggest that submucosal δ-opioid receptors may function to limit neurogenic secretion associated with intestinal inflammation.
In the present investigation, we addressed the hypothesis that kallidin-induced anion secretion, like that evoked by histamine, serotonin, or trypsin, is mediated by opioid-sensitive enteric neural circuits in the porcine ileum. Cannabinoids produce antipropulsive actions in the intestine that are similar to opioids, but their ability to alter intestinal secretion has not been clearly defined (Izzo et al., 2001). Because immunoreactivity for cannabinoid CB1 receptors has been detected in the porcine submucosal neurons (Kulkarni-Narla and Brown, 2000), it was of additional interest to compare the effects of the cannabinoid agonist HU-210 with those of the selective δ-opioid agonist DPDPE on kallidin-stimulated neurogenic ion transport in the porcine ileum.
Materials and Methods
Drugs and Chemicals.
Kallidin,d-arginyl-l-arginyl-l-prolyl-trans-4-hydroxy-l-prolylglycyl-3-(2-thienyl)-l-alanyl-l-seryl-d-1,2,3,4-tetrahydro-3-isoquinolinecarbonyl-l-(2α,3β,7αβ)-octahydro-1H-indole-2-carbonyl-l-arginine (HOE-140), and [des-Arg9,Leu8]-bradykinin (DALBK) were obtained from Bachem (Torrance, CA). Atropine, bumetanide, carbamylcholine chloride (carbachol), 4,4′-diisothiocyanatostilbene-2,2′-disulfonic acid (DIDS), histamine, indomethacin, naltrindole, and saxitoxin were obtained from Sigma-Aldrich (St. Louis, MO). DPDPE was purchased from Peninsula Laboratories (Belmont, CA). (6aR)-trans-3-(1,1-dimethylheptyl)-6a,7,10,10a-tetrahydro-1-hydroxy-6,6-dimethyl-6H-dibenzo[b,d]pyran-9-methanol (HU-210) was purchased from Tocris Cookson Inc. (Ballwin, MO). HU-210 and indomethacin were solubilized in dimethyl sulfoxide (DMSO); this solvent had no effect on baseline or kallidin-stimulated intestinal ion transport. DPDPE was solubilized in 0.01 M acetic acid with 0.1% bovine serum albumin, aliquoted at stock concentrations of 100 μM, and stored until use at −65°C. All other drugs and reagents were dissolved in distilled water and added to the contraluminal bathing medium unless otherwise noted.
Animals and Tissue Preparation.
Intestinal tissues were obtained from Yorkshire pigs (6–10 weeks of age; 10–18 kg b.wt.) of each sex that were not fasted before sacrifice. Animals were sedated with an intramuscular injection of tiletamine hydrochloride-zolazepam (Telazol, 8 mg/kg; Fort Dodge Laboratories, Fort Dodge, IA), in combination with xylazine (8 mg/kg). The animals were subsequently euthanized by barbiturate overdose in accordance with approved University of Minnesota Institutional Animal Care and Use Committee protocols. A midline laparotomy was performed to expose the intestine and a portion of the ileum, identified by its attachment to the ileo-cecal ligament, was removed and placed in an oxygenated organ preservation solution (118 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl2, 0.5 mM MgCl2, 25 mM NaHCO3, 1.0 mM NaH2PO4, and 11 mMd-glucose, pH 7.4).
Ileal segments were stripped of underlying smooth muscle layers, and sheets of mucosa-submucosa were mounted between two Lucite half-chambers with a surface area of 2 cm2. Mucosal sheets were bathed on both sides with a salt solution that approximated the composition of the porcine extracellular fluid (130 mM NaCl, 6 mM KCl, 3 mM CaCl2, 0.7 mM MgCl2, 20 mM NaHCO3, 0.29 mM NaH2PO4, and 1.3 mM Na2HPO4) at pH 7.4 and gassed continuously with 5% CO2 in O2 at 39°C (porcine core temperature). In anion substitution experiments, gluconic acid was substituted for chloride ion and HEPES substituted for bicarbonate ion at equimolar concentrations. d-Glucose and mannitol (10 mM) were added to the contraluminal and luminal bathing media, respectively.
Measurement of Transepithelial Ion Transport.
Short-circuit current (Isc, in microamperes per square centimeter) across each mucosa-submucosal sheet was monitored continuously by an automatic voltage-clamp apparatus (model TR100; JWT Engineering, Overland Park, KS; model EVC-4000, WPI, Sarasota, FL). Experiments were initiated after the basal Ischad stabilized (approximately 25–35 min). Tissue conductance [Gt, in milliSiemens (mS) per square centimeter] was calculated by Ohm's law from the current change produced by the periodic delivery of a bipolar 5-mV pulse measured throughout each experiment. Data were acquired using a PowerLab data acquisition unit and analyzed with Chart data analysis software (AD Instruments, Grand Junction, CO). Both Isc and Gt were determined immediately before drug administration and at the peak of drug action. At the end of each experiment, when Isc had returned to baseline, mucosal Isc responses to 10 μM carbachol (contraluminal addition) and 10 mM glucose (luminal addition) were measured in each tissue to assess tissue viability.
Kallidin was added to the contraluminal bathing medium to achieve a final bath concentration of 1 μM. This concentration was chosen because it is in the upper range of the kallidin concentration-effect relationships determined in previous studies in vitro with intestinal mucosa preparations (Gaginella and Kachur, 1989). In some experiments, receptor blockers or ion transport inhibitors were added to either the luminal or contraluminal bathing medium 15 min before kallidin addition. Some tissues were pretreated contraluminally with 0.1 μM DPDPE or 1 μM HU-210 10 min before kallidin addition; in some experiments with DPDPE, saxitoxin (0.1 μM) or the selective δ-opioid antagonist naltrindole (1 μM) was added to the contraluminal bathing medium 5 min before DPDPE addition.
pH Stat Titration.
Tissue sheets were mounted in Ussing chambers under short-circuit conditions and bathed luminally or contraluminally in physiological salt solution and a salt solution with an equimolar substitution of sodium gluconate for sodium bicarbonate and sodium phosphate on the opposite side. The salt solutions were maintained at 39°C and gassed continuously with 5% CO2 in O2 or 100% O2, respectively. In a second set of experiments, tissues bathed with HCO
Data Analysis.
Data are expressed as mean Isc or Gt under baseline conditions, or as mean changes in peak Iscelevations occurring in response to kallidin or other substances. Data from tissues that, at the end of the experimental period, did not respond to carbachol and glucose with elevations in Isc were omitted. In the case that the tissues serving as controls manifested poor responses to glucose and carbachol, data from all tissues obtained from the donor animal were excluded from further analysis. Statistical analyses of data were performed using the PRISM computer software program (version 3.0; GraphPad Software Inc., San Diego, CA). Comparisons between a control mean and a single treatment mean were made with a two-tailed, paired, or unpaired Student's t test when appropriate. Comparisons of a control mean with multiple treatment means were made by one-way analysis of variance followed by Dunnett's test. In all cases, the limit for statistical significance was set at P < 0.05.
Results
Mediators of Kallidin Action.
Baseline Isc and Gt in isolated sheets of ileal mucosa-submucosa averaged 4 ± 3 μA/cm2 and 23 ± 1 mS/cm2 (n = 224 tissues from 37 pigs). At a contraluminal concentration of 1 μM, kallidin produced a monophasic rise in Isc that attained a peak elevation of 80 ± 6 μA/cm2 relative to mean baseline values (n = 97 tissues from 37 pigs). The duration of kallidin action on Isc was 11.2 ± 0.5 min (8 tissues from four pigs). The kallidin-induced increase in Isc was 70% of that produced by 10 μM carbachol (ΔIsc produced by carbachol = 114 ± 10 μA/cm2, n = 65 tissues from 28 pigs). Tissue conductance nearly doubled to 43 ± 3 mS/cm2 when determined at the time of peak change in Isc produced by kallidin (n = 65 tissues from 28 pigs).
To verify that the action of kallidin was mediated by enteric neurons, mucosal Isc responses to kallidin were examined in tissues pretreated with 0.1 μM saxitoxin, a neuronal Na+ channel blocker. The neurotoxin significantly increased baseline Gt from 21 ± 5 to 26 ± 6 mS/cm2 (15 tissues from 7 pigs;P < 0.05, paired t test). In the presence of saxitoxin, Isc elevations produced by kallidin decreased by 67% relative to those in toxin-untreated tissues (Fig.1).
To determine whether the activation of muscarinic cholinergic receptors was involved in kallidin action, tissues were pretreated with the muscarinic cholinergic antagonist atropine. At a contraluminal concentration of 0.1 μM, atropine did not significantly alter baseline Isc or Gt, and mucosal Isc responses to kallidin were not significantly altered in atropine-treated tissues (Fig. 1). In tissues pretreated with atropine, Isc responses to the contraluminal addition of 10 μM carbachol were abolished (data not shown).
Because kallidin action may depend upon the formation of eicosanoids, mucosal responses to kallidin were measured in seven tissues pretreated with indomethacin, a cyclooxygenase inhibitor. At a contraluminal concentration of 10 μM, indomethacin did not significantly alter baseline Isc or Gt. However, it significantly decreased kallidin-induced Isc responses by 68% of peak Isc elevations in untreated tissues (Fig. 1). Because DMSO was used to solubilize indomethacin and a few other substances, the effect of this solvent on kallidin-induced Isc elevations was examined as well. At a contraluminal concentration of 0.1% (v/v), it did not significantly change baseline Isc or Gt, or alter significantly mucosal responses to 1 μM kallidin (mean ΔIsc after kallidin in DMSO-pretreated tissues = 101 ± 31 μA/cm2;P = 0.84 versus response in DMSO-untreated tissues; Student's t test, n = 8 tissues from five pigs).
The type of kinin receptor mediating mucosal responses to kallidin was determined in tissues pretreated with the kinin B1 receptor antagonist DALBK or the B2 receptor antagonist HOE-140. At a contraluminal concentration of 1 μM, HOE-140, but not DALBK, significantly decreased mucosal Isc responses to kallidin (Fig. 2).
Ionic Basis for Kallidin-Evoked Short-Circuit Current.
Previous reports have indicated that kinins induce anion secretion across the intestinal epithelium (Gaginella and Kachur, 1989; Cuthbert and Huxley, 1998). Therefore, tissues were pretreated with bumetanide, a blocker of the Na+/K+/Cl−cotransporter that represents one Cl− entry pathway in intestinal epithelial cells that is of importance in active anion secretion. Bumetanide did not significantly change baseline Isc or Gt after its contraluminal addition at 10 μM. At this concentration, it did not significantly decrease the peak Isc elevations occurring in response to kallidin (mean ΔIscafter kallidin in untreated and bumetanide-pretreated tissues = 107 ± 14 and 97 ± 31 μA/cm2, respectively; P > 0.05, Student's t test,n = 26 and 5 tissues from 13 and 3 pigs, respectively).
Anion substitution experiments were undertaken to assess the dependence of the Isc response to kallidin on extracellular Cl−- and HCO
The peak elevations in Isc produced by kallidin in tissues bathed in HCO
In tissues bathed in Cl−-deficient media, saxitoxin significantly reduced kallidin-induced elevations in Isc (Fig. 3A). In contrast, Isc responses to kallidin in tissues bathed in HCO
Some tissues were treated luminally with 0.3 mM DIDS before kallidin addition. Luminal DIDS did not alter kallidin action in tissues bathed in either Cl−-or HCO
pH-stat titration experiments were undertaken to examine further the role of HCO
Alkalinization measurements using bicarbonate-containing physiological salt solution bathing both sides of mucosal sheets were undertaken as well. Tissues were maintained at a baseline pH of 7.43 ± 0.02 in luminal buret experiments and a pH of 7.44 ± 0.01 in contraluminal buret experiments. The contraluminal addition of 1 μM kallidin produced a rapid alkalinization of both the luminal and contraluminal bathing media, with the former action being significantly greater than the latter (Fig. 4). The subsequent contraluminal administration of 10 μM carbachol also produced both luminal and contraluminal alkalinization (nanomoles of base equivalents delivered luminally and contraluminally after carbachol = 120 ± 20 and 50 ± 20, respectively;P > 0.05, unpaired Student's t test,n = 8 and 6 tissues from 7 and 6 pigs, respectively). Histamine, in contrast, produced little or no luminal alkalinization after its contraluminal addition at 2 μM (Fig. 4).
Effects of Opioid and Cannabinoid Receptor Agonists on Kallidin-Stimulated Ion Transport.
The δ-opioid agonist DPDPE (0.1 μM, contraluminal addition) did not produce significant changes in baseline Isc or Gt. However, it decreased by 53% the peak Iscelevation produced by kallidin. Tissues pretreated with both saxitoxin and DPDPE seemed to display an additional decrease in Isc responses to kallidin. The selective δ-opioid antagonist naltrindole prevented the inhibitory action of DPDPE (Fig. 5).
At a contraluminal concentration of 1 μM, the nonselective cannabinoid agonist HU-210 did not produce significant changes in baseline Isc or Gt, but significantly decreased subsequent mucosal responses to kallidin (Fig.5).
Discussion
Consistent with the results of previous investigations on the intestinal secretory actions of kinins (Gaginella and Kachur, 1989), kallidin transiently increased Isc in mucosa-submucosa sheets from porcine ileum after its contraluminal administration. This effect was blunted by saxitoxin and indomethacin, providing evidence that it is mediated in part by enteric neurons and cyclooxygenase metabolites, respectively. The enteric neural circuit(s) mediating kallidin-induced secretion does not seem to contain muscarinic cholinergic receptors because of the insensitivity of kallidin action to atropine. This result is in contrast to that obtained by Diener et al. (1988), who reported that atropine decreased significantly mucosal Isc responses to bradykinin in the rat distal colon. However, our previous studies in porcine ileal mucosa that examined the secretory effects of other inflammatory mediators, including serotonin (Green and Brown, 2002), proteinases (Green et al., 2000), and histamine (Poonyachoti and Brown, 2001), showed a similar pattern of atropine resistance. These results in combination suggest that proinflammatory mediators stimulate active anion secretion in the porcine intestine through a neural mechanism that does not involve cholinergic secretomotor neurons. Kinin B2 receptors seem to mediate the effect of kallidin on transepithelial ion transport, because kallidin-induced elevations in Isc were sensitive to the selective B2 receptor blocker HOE-140, but not to the kinin B1 antagonist DALBK. This result again is consistent with studies in other intestinal segments and in other mammalian species as well as in intestinal tissues from B2 receptor knockout mice, which demonstrate that kinin B2 receptors mediate the secretory effects of kinins (Gaginella and Kachur, 1989; Cuthbert and Huxley, 1998). The present results show that the porcine ileal mucosa seems to be generally similar to analogous intestinal preparations from other species with respect to the kinin receptor type and the neural and eicosanoid influences contributing to kallidin activity in the intestinal mucosa.
The kallidin-induced elevation in Isc across the porcine ileal epithelium bathed in media containing Cl− and HCO
Saxitoxin decreased Isc responses to kallidin in tissues bathed in Cl−-deficient media, but had no significant effect on responses to kallidin in tissues bathed in HCO
Because the neurogenic actions of kallidin seemed to involve electrogenic HCO
Enteric δ-like opioid receptors seem to modulate electrogenic ion transport evoked by transmural stimulation of submucosal neurons in the porcine ileum (Poonyachoti et al., 2001). Moreover, Isc elevations produced by either trypsin, histamine, serotonin, or an immediate hypersensitivity reaction to a food allergen in the porcine ileal mucosa are attenuated by the selective δ-opioid agonist DPDPE (Green et al., 2000; Poonyachoti and Brown, 2001; Green and Brown, 2002). We tested the hypothesis that δ-opioid receptors are expressed in a common enteric neuronal circuit that mediates secretory responses to inflammatory mediators, including kallidin. In support of this hypothesis, DPDPE markedly attenuated mucosal Isc responses to kallidin, and its effects were prevented by the selective δ-opioid antagonist naltrindole. When tissues were pretreated with both saxitoxin and DPDPE, there seemed to be a small additional decrease in kallidin action. This may be due to a minor action of kallidin on additional enteric neural pathways that do not express δ-opioid receptors. In previous studies, the combination of the neurotoxin and DPDPE was without any additional effect on mucosal Iscresponses to histamine or serotonin (Poonyachoti and Brown, 2001; Green and Brown, 2002). Based on this result and our previous data, we postulate that the indomethacin-sensitive portion of HCO
Proinflammatory kinins seem to stimulate neurogenic HCO
Footnotes
-
↵1 Present address: United States Department of Agriculture, Agricultural Research Service, Clay Center, NE 68933-0166.
-
This study was funded in part by National Institutes of Health Grant DA-10200. B.T.G. was a predoctoral trainee supported by National Institutes of Health Training Grant T32 DA-07239.
-
DOI: 10.1124/jpet.102.047829
- Abbreviations:
- DPDPE
- [d-Pen2,5]-enkephalin
- CB
- cannabinoid
- HOE-140
- d-arginyl-l-arginyl-l-prolyl-trans-4-hydroxy-l-prolylglycyl-3-(2-thienyl)-l-alanyl-l-seryl-d-1,2,3,4-tetrahydro-3-isoquinolinecarbonyl-l-(2α,3β,7αβ)-octahydro-1H-indole-2-carbonyl-l-arginine
- DALBK
- [des-Arg9,Leu8]-bradykinin
- DIDS
- 4,4′-diisothiocyanato-stilbene-2,2′-disulfonic acid
- HU-210
- (6aR)-trans-3-(1,1-dimethylheptyl)-6a,7,10,10a-tetrahydro-1-hydroxy-6,6-dimethyl-6H-dibenzo[b,d]pyran-9-methanol)
- DMSO
- dimethyl sulfoxide
- Received December 23, 2002.
- Accepted February 4, 2003.
- The American Society for Pharmacology and Experimental Therapeutics