INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

ATP, WHICH IS RELEASED FROM many cell types including epithelial tissue, functions as an autocrine/paracrine agonist by activating the two large families of P2X and P2Y receptors (41). The P2Y receptors belong to the superfamily of G protein-coupled receptors and are coupled to the inositol phosphate pathway resulting in the mobilization of Ca²⁺. A subset of P2Y receptors is activated by UTP to a similar degree to that by ATP and, in some instances, more potently (28, 41, 52).

The colon, the terminal part of the gastrointestinal tract, performs both the absorption and secretion of a variety of electrolytes by the epithelial transport systems. These transport systems are regulated by many kinds of endocrine, neurocrine, and paracrine agents and probably play an important role in maintaining fluid and electrolyte homeostasis in the whole body (1). Previous studies have suggested that P2-type receptor activation by ATP (and/or UTP) was also involved in regulating electrolyte transport in the colon. It has been reported that ATP indirectly stimulated colonic Cl secretion by acting on submucosal secretomotor neurons (9). More recently, ATP has been shown to induce an increase in intracellular Ca²⁺ concentration ([Ca²⁺]_i) and to stimulate Cl and mucin secretion in a colonic carcinoma cell line (12, 17, 18, 21, 23, 32, 37, 46, 47). In the cases of T84 and Caco-2 cells, UTP added to the luminal side also elicited anion secretion, implicating the role of the P2Y-type receptor residing on the luminal membrane (23, 46). In addition to anion secretion, the induction of electrogenic K⁺ secretion by luminal ATP and UTP through activation of the P2Y receptor has been reported in the rat colon (26). The epithelial P2Y receptor in the colon may not only participate in regulating transport functions, but may also play a role in regulating cell proliferation and apoptosis (21).

Electrogenic Na⁺ absorption, which involves an apical amiloride-sensitive Na⁺ channel and basolateral Na⁺, K⁺-ATPase/K⁺ channel, is one of the major pathways for Na⁺ absorption in the colon (1). Although this Na⁺ transport pathway is known to be stimulated by aldosterone and -adrenergic agonists and inhibited by vasopressin (1, 44, 51), the role of extracellular ATP and other nucleotides in regulating this process has not been reported. ATP and/or UTP, however, has been shown to inhibit electrogenic Na⁺ absorption from the luminal side in several other epithelia by activating, in most cases, P2Y receptors (3, 8, 10, 22, 24, 27, 34, 36, 42, 53). We, therefore, investigated whether a functional P2Y receptor exists on the apical membrane of the colonic epithelium, which would be coupled to the regulation on electrogenic Na⁺ absorption. To this end, the effect of ATP and UTP added to the luminal side on the short-circuit current (I_sc) and ²²Na⁺ flux was examined in an isolated guinea pig distal colon mounted in Ussing chambers. In addition, we examined the effect of applying a hypotonic luminal solution to clarify whether endogenous ATP may support the regulatory volume decrease by inhibiting Na⁺ uptake.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Tissue preparation. Hartley-strain, male guinea pigs weighing 250-700 g were used in the experiments. All animals were injected twice with 375 µg/kg body wt of aldosterone (1.39 mM in saline) to enhance the electrogenic sodium absorption, once by a subcutaneous injection the evening before the experiment, and then by an intramuscular injection 4 h before the start of the experiment. The animals were stunned by a blow to the head and bled to death. The distal colon was excised, opened along the root of the mesentery, and muscle and submucosal layers were removed with a glass microscopic slide (55). Procedures involving animals were approved by the Institutional Animal Care Board at the University of Shizuoka.

Solutions. The standard bathing solution contained (in mM) 119 NaCl, 21 NaHCO₃, 2.4 K₂HPO₄, 0.6 KH₂PO₄, 1.2 MgCl₂, 1.2 CaCl₂, and 10 glucose. The Ca²⁺-free solution was prepared by omitting CaCl₂ and adding 0.2 mM EGTA. The Ca²⁺-Mg²⁺-free solution omitted both CaCl₂ and MgCl₂, whereas 1 mM EDTA/5 mM Tris was added. The isotonic control solution for the mechanical stimulation experiment contained (in mM) 88 NaCl, 21 NaHCO₃, 2.4 K₂HPO₄, 0.6 KH₂PO₄, 1.2 MgCl₂, 1.2 CaCl₂, 10 glucose, and 60 sucrose (290 mosmol/KgH₂O). The hypotonic solution was prepared by omitting sucrose from the isotonic solution (230 mosmol/KgH₂O). All solutions were gassed with 95% O₂-5% CO₂ (pH 7.3-7.4).

I_sc measurements. I_sc and transmural tissue resistance (G_t) were measured in vitro in Ussing chambers as previously described (44, 55). Stripped mucosa was mounted vertically between acrylic resin chambers with an internal surface area of 0.5 cm². The bathing solution in each chamber was 10 ml, and its temperature was maintained at 37°C in a water-jacketed reservoir. The tissue was continuously short circuited, with compensation for the fluid resistance between the two potential-sensing bridges, by using a voltage-clamping amplifier (CEZ9100, Nihon Kohden, Tokyo, Japan). The transepithelial potential was measured through 1 M KCl-agar bridges connected to a pair of calomel half-cells, the transepithelial current being applied across the tissue through a pair of Ag-AgCl electrodes kept in contact with the mucosal and serosal bathing solutions through a pair of 1 M NaCl-agar bridges. The I_sc value is expressed as positive when the current flowed from the mucosa to serosa. G_t was measured by recording the current resulting from short-duration, square bipolar voltage pulses (±5 mV) imposed across the tissue and then calculated according to Ohm's law. In all experiments, 10 µM indomethacin was added to the serosal solution to prevent the effect of endogenously produced prostaglandins.

²²Na+ flux measurements. The unidirectional transmural flux of ²²Na⁺ was measured under short-circuit conditions. The mucosal-to-serosal (J_ms) and serosal-to-mucosal (J_sm) flux values were measured in adjacent tissues that had G_t values differing by <30%. Thirty minutes were allowed for the isotopic steady state to be reached after labeling either the serosal or mucosal side of the bathing solution with ²²Na⁺. Ten samples (0.5 ml each) were taken from the unlabeled side at 10-min intervals and replaced with an equal volume of the unlabeled solution. The medium samples were assayed for ²²Na⁺ by a -well spectrometer.

Materials. Indomethacin, amiloride, benzamil, bumetanide, ATP, UTP (Na⁺ salt), ADP (Na⁺ salt), adenosine, suramin, thapsigargin, and hexokinase were purchased from Sigma (St. Louis, MO). ATP-Na⁺ was used in the experiment to examine the effect of luminal Ca²⁺-Mg²⁺ removal, with ATP-Mg²⁺ being used in all other experiments. Ionomycin was purchased from Calbiochem (San Diego, CA). All other chemicals were purchased from Wako Pure Chemicals (Osaka, Japan). Each drug was applied from a concentrated stock solution dissolved in water or DMSO, the final volume of DMSO in an experimental solution always being 0.1%. ²²Na⁺ was purchased from Dupont New England Nuclear (Boston, MA).

Statistics analyses. Each data value is presented as the mean ± SE of number (n) of guinea pigs. Statistical comparisons between two means were made with Student's t-test (paired or unpaired, as appropriate). Significance was accepted at P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effect of luminal ATP on the electrogenic Na+ absorption. The addition of ATP (1 mM) to the mucosal solution resulted in a decrease in I_sc and G_t in the guinea pig distal colon (Fig. 1, A and C). In some cases, a small transient I_sc increase with a duration of <5 min could be observed before the decrease. After the tissue had been pretreated with the epithelial Na⁺ channel blocker amiloride (0.1 mM mucosal), no such ATP-induced I_sc and G_t decreases were apparent, although initial transient I_sc increase was enhanced (Fig. 1, B and D). These results suggest that the luminal ATP-induced reduction in I_sc and G_t was probably due to an inhibition of the amiloride-sensitive electrogenic Na⁺ absorption. On the other hand, when ATP was added to the serosal side (1 mM), there was a robust increase in I_sc and G_t without any I_sc decrease below the basal level (Fig. 2; I_sc = 110 ± 11 µA/cm², G_t = 2 ± 0.4 mS/cm², n = 4). Therefore, ATP added to the mucosal solution probably caused the decrease in I_sc through the activation of a receptor on the apical surface, and not a receptor on the basolateral surface, after penetrating into the serosal solution. Negative I_sc in the presence of amiloride (Fig. 1) is likely to be mainly due to an electrogenic K⁺ secretion, because it was largely suppressed by serosal bumetanide (see Figs. 5-7 and also Ref. 44). Whether the K⁺ secretion was also affected by luminal ATP remained to be determined.

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Effects of luminal ATP on short-circuit current (Isc) and transmural tissue resistance (Gt) in mucosa from the guinea pig distal colon. The guinea pigs used in the experiments reported in this paper were all pretreated with aldosterone (see MATERIALS AND METHODS). Typical time course traces are shown of the ATP-induced change in Isc when ATP was added before (A) and after (B) the amiloride treatment. Arrows indicate the time when ATP (1 mM) or amiloride (Amilo; 0.1 mM) was added to the mucosal solution. A summary is shown of the ATP-induced change in Isc and Gt when ATP was added before (C; n = 7) and after (D; n = 5) the amiloride treatment.

View larger version (6K): [in this window] [in a new window]   Fig. 2.   Effect of serosal ATP on Isc. A typical time course trace of Isc is shown with the arrow indicating when ATP (1 mM) was added to the serosal solution.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Effect of removing Ca2+ and Mg2+ from the mucosal side on the ATP-induced inhibition of amiloride-sensitive Isc. Concentration-response relationships for the inhibitory effect of ATP on amiloride-sensitive Isc were determined for the control solution () and luminal Mg2+-Ca2+-free solution () by using adjacent tissues (n = 7). The %inhibition of amiloride-sensitive Isc was calculated from the Isc decrease induced by cumulative concentrations of luminal ATP and finally by luminal amiloride (0.1 mM). * P < 0.05; ** P < 0.01 compared with the control value.

Agonists and antagonists. The potency of different nucleosides or nucleotides for reducing amiloride-sensitive I_sc was then studied. This and the subsequent experiments were conducted in the presence of serosal bumetanide (10 µM) to inhibit the I_sc components due to electrogenic K⁺ and Cl secretions. Under this condition, basal I_sc is almost totally amiloride-sensitive (44). As is shown in Fig. 4, the concentration-response relationship for the inhibitory effect of luminal ATP on amiloride-sensitive I_sc in the presence of bumetanide was essentially similar to that observed in the absence of serosal bumetanide (Fig. 3). Luminal UTP inhibited amiloride-sensitive I_sc with a similar potency to that of ATP, suggesting the involvement of a P2Y receptor (28, 41, 52). The threshold concentration of ADP was ~10-fold higher and that of adenosine was ~100-fold higher than the threshold concentration of ATP or UTP. Thus the order of potency appears to have been ATP = UTP > ADP > adenosine.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Concentration dependence of the inhibitory effect of nucleotides and a nucleoside on amiloride-sensitive Isc. The %inhibition of amiloride-sensitive Isc was calculated from the Isc decrease induced by cumulative concentrations of luminal nucleotides or a nucleoside and finally by luminal amiloride (0.1 mM) [ATP (; n = 4), UTP (, n = 4), ADP (; n = 5), and adenosine (; n = 5)]. The %inhibition of amiloride-sensitive Isc by luminal ATP at a concentration of 1 mM was 45 ± 5% from these measurements, whereas it was 50 ± 5% (n = 5) when determined by a single concentration. The concentration-response relationship, determined by this cumulative addition protocol, was therefore not seriously affected by a possible downregulation process.

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Effect of suramin on the ATP-induced inhibition of amiloride-sensitive Isc and Gt. A: typical time course trace; arrows indicate when suramin (1 mM), ATP (1 mM), and amiloride (0.1 mM) were added to the mucosal solution. B: summary of the inhibition of amiloride-sensitive Isc induced by luminal ATP (0.1 or 1 mM) in the absence (open column) and presence (hatched column) of suramin (1 mM). * P < 0.05 compared with the control values (n = 5 for 0.1 mM ATP, and n = 5 for 1 mM ATP).

Role of [Ca²+]_i. The activation of P2Y receptors is, in most cases, coupled to the Ca²⁺-signaling pathway. The next series of experiments was designed to investigate the role of the increase in [Ca²⁺]_i in the ATP-induced inhibition of Na⁺ absorption. We first examined whether or not an increase in [Ca²⁺]_i by itself could inhibit amiloride-sensitive I_sc in the guinea pig colon by using the Ca²⁺ pump inhibitor thapsigargin and the Ca²⁺ ionophore ionomycin. Both thapsigargin and ionomycin caused a decrease in I_sc (Fig. 6). When the tissue had been pretreated with luminal amiloride (0.1 mM), neither thapsigargin nor ionomycin had any effect on I_sc (data not shown). In addition, the I_sc decreases induced by thapsigargin and ionomycin were remarkably attenuated when the tissue was bathed with a serosal Ca²⁺-free solution (see Fig. 7, C and D). These findings indicate that an increase in [Ca²⁺]_i was sufficient to cause a decrease in the amiloride-sensitive Na⁺ absorption.

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Inhibition of the electrogenic Na+ absorption by thapsigargin and ionomycin. Typical time course traces are shown when thapsigargin (A) or ionomycin (B) was added to the serosal side and then amiloride was added to the mucosal side. Arrows indicate when thapsigargin (serosal, 10 µM), ionomycin (mucosal, 1 µM), ATP (mucosal, 100 µM), and amiloride (mucosal, 0.1 mM) were added. C: summary is given of the Isc values 20 min after the treatment of thapsigargin or ionomycin (n = 7 for thapsigargin and n = 5 for ionomycin).

View larger version (29K): [in this window] [in a new window]   Fig. 7.   Role of intracellular Ca2+ in mediating the ATP-induced inhibition of amiloride-sensitive Isc. Typical time course traces are shown of the ATP-induced Isc change under the following conditions: control solution (A), serosal Ca2+-free solution (B), serosal Ca2+-free solution in the presence of serosal thapsigargin (C; 10 µM), and serosal Ca2+-free solution in the presence of mucosal ionomycin (D; 1 µM). Summaries are given of the Isc values (E) and %inhibition of amiloride-sensitive Isc (F). The Isc measurement under each experimental condition was done simultaneously, with the control measurement by using adjacent tissue. ATP was applied to both control and experimental tissue 20 min after thapsigargin or ionomycin was applied to the experimental tissue. The percentage inhibition of amiloride-sensitive Isc induced by ATP was not significantly different from each other (n = 7 for normal, n = 10 for Ca2+-free solution, n = 6 for Ca2+-free solution containing thapsigargin, and n = 4 for Ca2+-free solution containing ionomycin).

ATP release to the luminal solution. In the final series of experiments, we investigated the physiological source of ATP acting on the apical membrane receptor coupled with the inhibition of electrogenic Na⁺ absorption in the colon. We first examined the effect on basal I_sc of adding hexokinase to the luminal solution. If sufficient ATP is released to the luminal solution to induce inhibition under the basal condition, the addition of hexokinase would evoke an increase in I_sc by scavenging ATP as a result of respectively converting ATP and the glucose to ADP and glucose-6-phosphate (30). However, as shown in Fig. 8A, hexokinase (2.5 U) added to the luminal solution failed to increase I_sc, and instead slightly decreased it (by 4.5 ± 1.2%, n = 10). In the presence of hexokinase at this concentration, the inhibition of I_sc induced by luminal 10 µM ATP was completely abolished (data not shown, n = 6), and that induced by 100 µM ATP was greatly suppressed (Fig. 8A; 4.8 ± 2.1% inhibition, n = 3; compare with Fig. 4). It is thus unlikely under these experimental conditions that sufficient ATP to inhibit electrogenic Na⁺ absorption would have been released to the luminal solution.

View larger version (17K): [in this window] [in a new window]   Fig. 8.   ATP release to the luminal solution. A: effects of hexokinase on the ATP-induced inhibition of Isc. Typical time course traces of Isc are shown in the absence (solid line) and presence (broken line) of hexokinase (2.5 U) in the luminal solution. ATP was added to the luminal solution at a concentration of 100 µM. B: typical traces of the experiment in which the effect of mechanical stress on the ATP-induced inhibition of amiloride-sensitive Isc was examined are shown. To apply the mechanical stress to the mucosa, the isotonic (Iso) solution on the mucosal side was replaced by another Iso solution (solid line), hypotonic (Hypo) solution (dashed line), or Hypo solution containing hexokinase (Hypo + Hexokinase, bold solid line) at the point indicated by "replacement." This replacement was achieved by repeated evacuation of the luminal solution and changing with a fresh solution 3 times within 1 min. C: summary of the experiment as shown in B. Basal values were obtained just before the replacement procedure, and the after-replacement values were obtained 15 min after the replacement procedure (n = 12 for Iso, n = 6 for Hypo, and n = 6 for Hypo + Hexokinase). * P < 0.05 compared with the basal value.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The results of our present study on the guinea pig distal colon treated with aldosterone suggest that a P2Y-type purinergic receptor is likely to be present on the apical membrane that is coupled to the inhibition of electrogenic Na⁺ absorption. ATP added to the mucosal side caused decreases in I_sc (and G_t) and ²²Na⁺ absorption of essentially similar magnitude, and the I_sc and G_t decreases induced by ATP were suppressed in the presence of amiloride. UTP, which is an agonist for a subset of P2Y receptor, also partially inhibited amiloride-sensitive I_sc. In the present preparation, luminal ATP caused, in addition to the ultimate decrease, a transient increase in I_sc, which was not inhibited by amiloride (Fig. 1), suggesting the stimulation of Cl secretion. Whether this response would also be mediated by a P2Y receptor remains to be determined. In addition, the present preparations exhibited electrogenic K⁺ secretion (Fig. 1). Whether this process is also affected by luminal ATP remains to be examined. Previous studies on gastrointestinal tissues have suggested that stimulation of the apical P2Y receptor induced Cl and K⁺ secretion (7, 23, 26, 46).

Multiple subtypes of P2Y receptors have been found (28, 41, 52), and it has been suggested that the P2Y-receptor subtype in the apical membrane is P2Y₂ for colonic K⁺ secretion (26) and P2Y₄ for intestinal Cl secretion (7, 31). To determine the subtype of the apical membrane P2Y receptor responsible for inhibiting Na⁺ absorption in the colon, the potency order of different nucleotides or nucleosides for suppressing amiloride-sensitive I_sc was examined, the result being ATP = UTP > ADP > adenosine. In addition, suramin significantly attenuated the inhibitory effect of luminal ATP on amiloride-sensitive I_sc. These findings are at least consistent with the P2Y₂ receptor contributing to the inhibition of Na⁺ absorption (5, 14, 33, 41, 52). However, it has recently been reported that the P2Y₄ receptor in the rat and mouse, but not human, was activated equipotently by ATP and UTP (2, 31, 48). In addition, suramin, although only weakly, inhibited the rat, but not human or mouse, P2Y₄ receptor (2, 6, 48). Thus the possible involvement of the P2Y₄ receptor cannot be entirely excluded. It has been said that the agonist potency determined from a functional assay on the native or recombinant P2 receptor was difficult to interpret for several reasons (29, 52). For example, it may be necessary to determine the effect of an ectonucleotidase, possibly existing on the apical surface of the colon (see the next paragraph for further discussion). In addition, more than one subtype of P2 receptor might be present on the apical membrane and contribute to the ATP-induced inhibition of Na⁺ absorption (see final paragraph). Further studies are therefore clearly needed to determine the receptor subtype responsible for the luminal ATP-induced inhibition of Na⁺ absorption.

When both Ca²⁺ and Mg²⁺ were removed from the mucosal side, the potency of the inhibitory effect of ATP on amiloride-sensitive I_sc was enhanced (Fig. 3). One possible explanation for this is that the active form of the agonist that induced the inhibition of Na⁺ absorption was ATP⁴ and not MgATP² or CaATP² (56). However, it is more likely that this enhanced effect by Ca²⁺ and Mg²⁺ removal was due to the inhibition of ectonucleotidases on the apical surface. The activities of certain ectonucleotidases depend on the presence of Ca²⁺ or Mg²⁺ or both (19, 40). The activity of such ectonucleotidases that degrade nucleotides could skew the concentration-response curve by generally shifting it to the right. In fact, the potency of the inhibition of Na⁺ absorption by luminal ATP in the present study was shifted to the right by at least one order of magnitude compared with the potency on the cloned P2Y receptors in the expression system (2, 14, 33). Although a previous study has demonstrated that cultured colonic epithelial cells possessed ecto-ATPase activities (12, 16, 47), whether or not such an ectonucleotidase was present on the apical surface of the present preparation remains to be determined.

The activation of P2Y receptors is known to lead to an increase in [Ca²⁺]_i through activation of the phospholipase C inositol 1,4,5-trisphosphate pathway (41, 52). In addition, in the present colon preparation, the amiloride-sensitive Na⁺ absorption could be inhibited by an increase in [Ca²⁺]_i, because it was inhibited by the addition of ionomycin or thapsigargin (Fig. 6). It seems, therefore, reasonable to assume that an increase in [Ca²⁺]_i mediated the ATP-induced inhibition of Na⁺ absorption. However, the removal of Ca²⁺ from the serosal side or the addition of thapsigargin or ionomycin in the absence of serosal Ca²⁺ failed to affect the ATP-induced inhibition of amiloride-sensitive I_sc (Fig. 7). Under these conditions, a receptor activation-coupled [Ca²⁺]_i increase was presumably attenuated by depletion of the [Ca²⁺]_i store (4, 39, 49). Therefore, an intracellular signaling pathway other than (or in addition to) the [Ca²⁺]_i increase was activated and may mainly have been responsible for the ATP-induced inhibition of electrogenic Na⁺ absorption. It has previously been shown in several tissues that an elevated [Ca²⁺]_i level was not required for the responses induced by activation of the P2Y receptor (15, 17, 22, 25, 27, 35).

Extracellular ATP released by mechanical or hypotonic stress has been suggested to have importance in the autocrine/paracrine control of epithelial cell functions (11, 13, 16, 20, 38, 43, 45, 50, 54). The present results for the colon suggest that a hypotonic challenge on the mucosal side caused a release of ATP, which, in turn, inhibited amiloride-sensitive Na⁺ absorption, probably via interaction with the apical P2Y receptor (Fig. 8). The inhibition of Na⁺ entry would conceivably lead to a reduction in the epithelial cell volume, thereby facilitating a regulatory volume decrease of the cell after the swelling induced by the hypotonic challenge. Facilitation of the regulatory volume decrease mediated by a released nucleotide has been demonstrated in gastrointestinal and airway epithelial cells (11, 43, 54). In addition, the inhibition of Na⁺ absorption in concert with the stimulation of Cl secretion, which is probably also induced by luminal ATP (Fig. 1), might help in flushing out noxious agents from the colonic lumen. In many other epithelial cells, luminal ATP has been reported to inhibit electrogenic Na⁺ absorption and to stimulate electrogenic Cl secretion through acting on the apical P2Y receptor (8, 10, 22, 24, 36, 42, 53). It is now clear in several epithelial cells and is likely to also be the case in the colon that, after mechanical/hypotonic stress, nucleotides are released to both the mucosal and serosal sides (28, 43, 50). In addition, multiple subtypes of the purinergic receptors are generally expressed on both apical basolateral membranes (7, 12, 23, 27, 36, 46). Autocrine/paracrine regulation of the intestinal transport function by nucleotides is thus a very complex issue and warrants further study.