Subepithelial trypsin induces enteric nerve-mediated anion secretion by activating proteinase-activated receptor 1 in the mouse cecum
- First Online:
- Cite this article as:
- Ikehara, O., Hayashi, H., Waguri, T. et al. J Physiol Sci (2012) 62: 211. doi:10.1007/s12576-012-0198-7
- 145 Views
Serine proteases are versatile signaling molecules and often exert this function by activating the proteinase-activated receptors (PAR1–PAR4). Our previous study on the mouse cecum has shown that the PAR1-activating peptide (AP) and PAR2-AP both induced electrogenic anion secretion. This secretion mediated by PAR1 probably occurred by activating the receptor on the submucosal secretomotor neurons, while PAR2-mediated anion secretion probably occurred by activating the receptor on the epithelial cells. This present study was aimed at using trypsin to further elucidate the roles of serine proteases and PARs in regulating intestinal anion secretion. A mucosal–submucosal sheet of the mouse cecum was mounted in Ussing chambers, and the short-circuit current (Isc) was measured. Trypsin added to the serosal side increased Isc with an ED50 value of approximately 100 nM. This Isc increase was suppressed by removing Cl− from the bathing solution. The Isc increase induced by 100 nM trypsin was substantially suppressed by tetrodotoxin, and partially inhibited by an NK1 receptor antagonist, by a muscarinic Ach-receptor antagonist, and by 5-hydroxytryptamine-3 (5-HT3) and 5-HT4 receptor antagonists. The Isc increase induced by trypsin was partially suppressed when the tissue had been pretreated with PAR1-AP, but not by a pretreatment with PAR2-AP. These results suggest that the serine protease, trypsin, induced anion secretion by activating the enteric secretomotor nerves. This response was initiated in part by activating PAR1 on the enteric nerves. Serine proteases and PARs are likely to be responsible for the diarrhea occurring under intestinal inflammatory conditions.
KeywordsSerine proteaseInflammationDiarrheaEicosanoid5-HydroxytryptamineSubstance P
Intestinal fluid secretion is mainly derived from electrogenic anion secretion with its accompanying obligate of Na+ and water. It is important to the physiology of digestion and absorption of nutrients and also for surface lubrication to enable the luminal contents to pass through smoothly. It is also believed to play a role in pathological conditions by flushing out noxious luminal agents . Intestinal anion secretion is controlled by a variety of luminal and subepithelial substances (or by conditions such as mechanical distortion) that may have originated exogenously or be derived from the host itself. They may directly affect the epithelium, but their effects may also be mediated by various neurocrine, paracrine, and endocrine systems. The enteric nervous system, particularly the submucosal nerves, plays a vital role in the neurocrine regulation of intestinal anion secretion [1–3].
Proteases are not merely protein-degrading enzymes, but are now viewed as important signaling molecules that have fundamental roles in a wide variety of physiological processes, and are also associated with multiple disease conditions, including inflammation and cancer [4–9]. Nearly one-third of mammarian proteases are serine proteases, a class that includes trypsin, thrombin, kallikrein, and also type II transmembrane serine protease [8, 9]. Several previous studies have demonstrated in vitro in an Ussing chamber that trypsin and thrombin induced intestinal electrogenic anion secretion [10–14].
The signaling functions of serine proteases are often mediated by the proteinase-activated receptors (PAR1–PAR4). The PAR family members, particularly PAR1 and PAR2, are expressed throughout the gastrointestinal tract on epithelial cells, smooth muscle cells, and enteric nerves [15–17]. The role of PAR1 and PAR2 in regulating intestinal anion secretion has been demonstrated in several isolated intestinal mucosa by using the PAR1-activating peptide (AP) and PAR2-AP [10–13, 18, 19]. In addition, the activation of PAR2 has been implicated in trypsin-induced anion secretion in the mouse distal colon, rat jejunum, and human colon [10, 12, 13]. However, the precise role of PARs in the serine protease-mediated regulation of intestinal transport had remained obscure and was thus investigated in the present study.
The cecum is the largest segment of the large intestine in rodents and many other mammals, and is highly active in luminal fermentation by bacteria as well as in epithelial transport [20, 21]. We have recently reported that both PAR1- and PAR2-APs added to the serosal side induced anion secretion in the mouse cecum in vitro in an Ussing chamber, whereby PAR1-AP-mediated anion secretion occurred by activating submucosal secretomotor neurons, whereas PAR2-AP-mediated anion secretion was by the direct activation of epithelial cells. The activation of anion secretion that was not mediated by the enteric nerve has been demonstrated in the same tissue by applying 10 μM trypsin to the serosal side . Based on these findings, the present study was designed to further elucidate the mechanism for serosal trypsin to regulate intestinal anion secretion in the mouse cecum, and particularly the involvement of PAR1, PAR2, and the enteric nerves. The results suggest that, in contrast to 10 μM trypsin just mentioned, a moderate concentration of trypsin (100 nM) added to the serosal side induced anion secretion by stimulating the submucosal secretomotor nerve, this effect being at least partially due to the activation PAR1, but not PAR2. We also investigated the role of such neurocrine and paracrine mediators as Ach, 5-hydroxytryptamine (5-HT), substance P, and arachidonate metabolites in the 100 nM trypsin-induced anion secretion.
Materials and methods
All procedures used in this study were performed in accordance with the “Guiding Principles for Care and Use of Animals” approved by the Physiological Society of Japan and by the Institutional Animal Care Board at the University of Shizuoka. Male mice (30–40 g, Std:ddY; Japan SLC, Hamamatsu, Japan) were fed with standard food and water ad libitum until the time of the experiments. The animals were then killed by cervical dislocation and the cecum was excised. The resulting tissue was opened into a flat sheet, and the musculature was removed by blunt dissection to obtain the mucosa–submucosa preparation. The tissue was divided into four pieces of approximately equal size. One of these pieces was used in most experiments for determining the trypsin-induced response under control conditions, while the others were used for determining the trypsin-induced response under various treatment conditions.
Each piece was then mounted vertically between Ussing-type chambers that provided an exposed area of 0.2 cm2. The volume of the bathing solution on each side was 5 ml, and the solution temperature was maintained at 37°C in a water-jacketed reservoir. The bathing solution contained (mM) NaCl, 119; NaHCO3, 21; K2HPO4, 2.4; KH2PO4, 0.6; CaCl2, 1.2; MgCl2, 1.2; and glucose, 10 (pH 7.4). A Cl-free solution was provided by, respectively, using 119 mM Na-gluconate, 1.2 mM Mg-(gluconate)2 and 8 mM Ca-(gluconate)2 in place of 119 mM NaCl, 1.2 mM MgCl2, and 1.2 mM CaCl2. Each solution was bubbled with 95%O2/5%CO2.
The experiments were performed under short-circuit conditions. The short-circuit current (Isc) and transmucosal conductance (Gt) were measured by using an automatic voltage-clamping device that compensated for the solution resistance between the potential-measuring electrodes (CEZ9100; Nihon Kohden, Tokyo, Japan) as previously described .
Bumetanide, procaine, 3-tropanyl-3,5-dichlorobenzoate, SB-204070 hydrochloride, atropine, hexamethonium, indomethacin, nordihydroguaiaretic acid (NDGA), thrombin from bovine plasma, trypsin from porcine pancreas, soybean trypsin inhibitor, and a P8340 protease inhibitor cocktail in DMSO were purchased from Sigma (St. Louis, MO, USA). Tetrodotoxin (TTX) was purchased from Calbiochem (La Jolla, CA, USA), and L-703,606 and 5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB) were purchased from Research Biochemical International (Natick, MA, USA). SKF-525A was purchased from Biomol Research Laboratories (Plymouth Meeting, PA, USA). Mouse PAR1 activating peptide (AP) SFFLRN-NH2 and mouse PAR2 AP SLIGRL-NH2 were purchased, respectively, from Yanaihara Institute (Fujinomiya, Japan) and Bachem (Bubendaf, Switzerland). Indomethacin was dissolved in 21 mM NaHCO3. Bumetanide, 3-tropanyl-3,5-dichlorobenzoate, NDGA, and SKF-525A were dissolved in dimethyl sulfoxide, before being administered to the bathing solution, the final concentration of dimethyl sulfoxide being 0.1%. The other chemicals were each applied from an aqueous stock solution.
Data and statistical analysis
Results are expressed as a percentage of the control response determined for each animal. Each value is presented as the mean ± SE, with n representing the number of animals. Statistical comparisons were made by Student’s paired t test, significance being accepted at P < 0.05.
Electrical responses to serosal trypsin
The mucosa–submucosa preparation of the mouse cecum in the Ussing chamber exhibited a basal short-circuit current (Isc) value of 92 ± 6 μA cm−2 and a transmural conductance (Gt) value of 20.8 ± 0.4 mS cm−2 (n = 78) 20–30 min after starting the incubation. The Isc value, but not Gt, decreased thereafter at the rate of approximately 10% per 10 min.
Ionic basis for the serosal trypsin-induced Isc increase
Involvement of the enteric nerves and other mediators
We then elucidated the involvement of 5-HT by examining the effects of the 5-HT3-receptor antagonist, 3-tropanyl-3,5-dichlorobenzoate, and of the 5-HT4-receptor antagonist, SB-204070. The Isc and Gt increases induced by 100 nM serosal trypsin was significantly suppressed when the tissues had been pretreated with both the 5-HT3 and 5HT4 receptor antagonists (Fig. 4; see also Fig. 3).
We then examined the role of neurotransmitters Ach and tachykinin in the 100 nM serosal trypsin-induced anion secretion by using antagonists for the musocarinic and nicotinic acetylcholine receptors (atropine and hexamethonium) and that for the tachykinin NK1 receptor (L-703,606). Atropine significantly reduced the increase in Isc, but not Gt, induced by 100 nM serosal trypsin, while hexamethonium had a minor influence on the response (Fig. 4; also see Fig. 3). In contrast, the trypsin-induced increases in Isc and Gt were both considerably decreased by L-703,606 (Fig. 4; also see Fig. 3). These results suggest that trypsin stimulated the release of substance P, resulting in activation of the NK1 receptor, and some Ach release to activate the muscarinic receptor.
Involvement of eicosanoids
Role of the proteinase-activated receptors
Effect of trypsin
The present results for the mouse cecum demonstrate that trypsin added to the serosal side induced electrogenic anion secretion with an ED50 value of 50–100 nM. Previous studies on the mammalian small and large intestines in an Ussing chamber have demonstrated that serosal trypsin could stimulate anion secretion with an ED50 value of 10–100 nM in most cases, this being similar to the value obtained in the present study [10–14]. However, the mechanism for stimulating anion secretion by trypsin appears to be diverse according to the intestinal segment, the dose of trypsin, and probably animal species. For example, the response in the mouse distal colon and porcine ileum was abolished by TTX [11, 12], like the response to 100 nM trypsin in the mouse cecum that was observed in the present study. On the other hand, the anion secretion induced by serosal trypsin has not been inhibited by TTX in the human rectum . In addition, our previous study on the mouse cecum has shown that, in sharp contrast to the moderate concentration used in the present study (100 nM), the activation of anion secretion induced by a high concentration of trypsin (10 μM) was not inhibited by TTX, indicating a dose-dependent difference in the action of trypsin . Trypsin would be likely to induce the inhibition of Na and/or Cl absorption, in addition to the stimulation of anion secretion, since the stimulants of anion secretion in the intestines often simultaneously inhibit NaCl absorption to facilitate a net fluid secretion . However, this likelihood has not yet been addressed.
Role of PARs
PAR1 and PAR2 have previously been demonstrated morphologically and functionally to be expressed in epithelial cells and/or enteric nerves and have been suggested to play a role in regulating intestinal secretion [10–14, 18, 19, 28, 29]. The results of the present study show that the Isc increase induced by trypsin was partially suppressed by pretreating with 30 μM PAR1-AP, which had been shown to almost completely desensitize PAR1 . It is therefore likely that trypsin induced anion secretion at least in part by activating PAR1, probably on the submucosal enteric nerve. Indeed, PAR1-AP has been shown to stimulate enteric nerve-mediated anion secretion , and also to be expressed in submucosal nerves . It is intriguing that the role of PAR1 in the enteric nerve appears to be complex, since PAR1-AP has been reported to inhibit the anion secretion evoked by stimulating enteric nerves in the mouse proximal colon, a segment next to the cecum .
We have previously shown that PAR2-AP stimulates anion secretion that is not inhibited by TTX. Here, 100 nM serosal trypsin evoked almost no anion secretion when the tissue was treated with TTX (Fig. 4), thus excluding a major role for PAR2 in trypsin-induced anion secretion. In agreement with this notion, desensitizing PAR2 failed to reduce the subsequent response to trypsin (Fig. 8). Other pathways not involving PAR1 or PAR2 would therefore also be involved in the trypsin-induced anion secretion. The failure of 100 nM trypsin to activate PAR2 and induce anion secretion in the mouse cecum was an unexpected finding, since PAR2 was expressed in this preparation and PAR2-AP can evoke anion secretion that is not mediated by enteric nerves . In fact, a high concentration of trypsin (10 μM) can induce TTX-insensitive anion secretion . Possibly, it was difficult, and therefore required a high concentration for large trypsin molecules to gain access to PAR2 residing deep in the tissue, probably on epithelial cells. The involvement of PAR2 in trypsin-induced anion secretion has been shown in the mouse distal colon and rat jejunum [10, 12, 13]. Taking this information together, PARs are likely to be present on various cell types and to play important roles in trypsin or other serine proteases regulating intestinal ion transport, but their precise roles are probably diverse according to the intestinal segment and animal species.
Neurocrine and paracrine mediators
The anion secretion induced by 100 nM serosal trypsin was partially inhibited by the muscarinic receptor antagonist, atropine, suggesting the involvement of Ach release from submucosal cholinergic secretomotor nerves [1–3, 30, 31]. The present results also suggest that activation of the NK1 receptor by substance P, which was probably released from afferent nerves, would also be responsible in part for the anion secretion reflex induced by trypsin [14, 32–37]. NK1 receptors have recently been shown to be abundantly expressed in subepithelial fibroblasts, although the relevance of this intriguing finding to the anion secretion induced by trypsin needs to be more fully investigated .
The anion secretion stimulated by 100 nM serosal trypsin was partially inhibited in the presence of both the 5-HT3 receptor and 5-HT4 receptor antagonists, suggesting the involvement of 5-HT release in this response [2, 3, 39–41]. 5-HT is mainly present in the epithelial enterochromaffin cells, but is also contained in the enteric nerve and mucosal mast cells [3, 39–41]. The mechanism by which trypsin stimulates any of these cells to release 5-HT remains for future study.
Arachidonic acid released from membrane phospholipids can be metabolized to generate various eicosanoids via three enzymatic routes: the cyclooxygenase, lipoxygenase and cytochrome P-450 epoxygenase pathways [42, 43]. The trypsin-induced anion secretion in the present study was substantially suppressed by NDGA (a lipoxygenase inhibitor) and SKF-525A (a cytochrome P-450 inhibitor), but not significantly affected by indomethacin (a cyclooxygenase inhibitor). This suggests that metabolites of the lipoxygenase pathway, hydroxyeicosatetraenoic acids (HETEs) and leukotrienes, and those of epoxygenase pathway, HETEs and epoxyeicosatrienoic acids (EETs), were involved in the response, but not that of the metabolites on the cyclooxygenase pathway, prostanoids. The intestinal mucosa has been shown to produce lipoxygenase metabolites [43, 44], which have been found to activate or inhibit anion secretion [45–50]. P450 expoxygenases and the production of HETEs and EETs have also been demonstrated in the intestinal tissues, although their role in regulating anion secretion is not known [42, 43, 51–53]. The cell types involved in releasing the lipoxygenase and epoxygenase metabolites of arachidonic acid in response to trypsin remain to be determined. Although prostanoids have hardly been involved in trypsin-induced anion secretion in the cecum, the anion secretion stimulated by PAR1-AP or PAR2-AP has been shown to be mediated via prostaglandin production in certain intestinal tissues [16, 43, 54–56].
The serine protease, trypsin, can induce anion secretion by activating the enteric secretomotor nerves in the mouse cecum. This response was initiated in part by activating PAR1, probably on the enteric nerves. Other pathways not involving PAR1 or PAR2 would also have been involved, but their precise action remains for future investigation. Since serine proteases are known to be recruited and/or activated under certain pathological conditions, they are likely to play a role in the development of diarrhea that is often observed under such conditions [1, 57–63]. This reaction may be important for protecting the intestine, particularly the cecum, by washing harmful microbes and their toxic products out of the lumen . The natural proteases working in situ and the precise mode of their actions in regulating intestinal fluid secretion may vary under specific conditions and warrant further investigation.
We thank Tony Innes of Link Associates for helping to edit the English text.
Conflict of interest
No conflicts of interest, financial or otherwise, are declared by the authors.