Effects of extracellular purines on ion transport across the integument of Hirudo medicinalis

Fresh water represents a hypo-osmotic environment to the leech, in which it has to cope with the loss of salts and uptake of water. The nephridia and the integument collaborate in accomplishing this osmoregulation. While the homeostatic mechanisms in the nephridia have been thoroughly investigated(Nieczaj and Zerbst-Boroffka,1993; Zerbst-Boroffka and Wenning, 1986; Zerbst-Boroffka et al., 1982), there is little information concerning integumental ion transport (Milde et al.,2001; Schnizler and Clauss,1998; Weber et al., 1993, 1995). The absorption of extracellular Na⁺ plays a prominent role in maintaining electrolyte balance. The majority of electrogenic Na⁺ absorption across the leech integument is likely to be via mechanisms analogous to those in vertebrate tight epithelia. Recent models of ion absorption processes across frog skin from fresh water have included energization of Na⁺ entry via apical Na⁺ channels by electrogenic H⁺V-ATPases (Ehrenfeld and Klein,1997; Nelson and Harvey,1999). Extracellular Na⁺ crosses the apical cell membrane through highly selective Na⁺ channels along an electrochemical gradient that is sustained by the basolateral Na⁺/K⁺-ATPases. It remains to be determined whether apical Na⁺ entry is driven by H⁺ V-ATPases in leech integument, but epithelial Na⁺ channels are, as in vertebrates,selectively blocked by amiloride and represent a rate-limiting target for the control of transcellular Na⁺ uptake (Weber et al., 1993, 1995).

In the last 40 years, vertebrate epithelia have been the preferred models for investigating Na⁺ transport in tight epithelia. Regulation of Na⁺ absorption occurs in several ways(Garty and Palmer, 1997), the major stimulus in vertebrates being the mineralocorticoid aldosterone. This steroid hormone mediates long-term adaptation by influencing the expression of regulatory factors at both transcriptional and translational levels. Peptides,for example antidiuretic hormone (ADH), bind to G-protein-coupled receptors and exert their regulatory effects within minutes. Invertebrates such as the leech also possess these phylogenetically old antidiuretic hormones, but they lack the mineralocorticoid system found in vertebrates(Oumi et al., 1994; Proux et al., 1987; Salzet et al., 1993, 1995; Satake et al., 1999). To investigate the processes underlying Na⁺ transport across tight epithelia in invertebrates, we used the integument of the leech Hirudo medicinalis for electrophysiological measurements(Milde et al., 2001; Schnizler and Clauss, 1998;Weber et al., 1993, 1995).

The regulatory properties of vertebrate epithelial Na⁺ channels have been subject to intensive study over the last 10 years, and regulation of these epithelial Na⁺ channels occurs by modulation of channel activity within the apical cell membranes as well as by control of the total number in surface-expressed channels(Garty and Palmer, 1997). High extracellular Na⁺ concentrations induce Na⁺self-inhibition, a mechanism that downregulates amiloride-sensitive apical Na⁺ entry (Palmer et al.,1998; Turnheim,1991). Furthermore, in a variety of epithelia, downregulation of apical Na⁺ conductances was observed when intracellular Na⁺ concentration increased. This autoregulative Na⁺feedback inhibition represents a short-term adaptive process, and current Na⁺ feedback inhibition models postulate an intracellular, as yet unidentified, Na⁺-sensing receptor that mediates, through G-protein-coupled mechanisms, a ubiquitin-protein ligase (Nedd4)-dependent endocytosis and degradation of the epithelial Na⁺ channels(Ishibashi et al., 1999; Komwatana et al., 1998). The involvement of G proteins was not confirmed in all cases, however(Hubner et al., 1999).

Living in fresh water and only occasionally entering brackish water(Herter, 1939), Hirudo medicinalis cannot be expected to be a euryhaline organism. Nevertheless,our previous studies showed that the leech integument reduces electrogenic Na⁺ uptake in the presence of high apical Na⁺concentrations. This short-term adaptive downregulation was maintained for several hours (as long as the experiments lasted) and quickly became inoperative after a return to low-[Na⁺] conditions(Weber et al., 1995). Adult leeches survive in water of up to 16 ‰ salinity for several months(Boroffka, 1968). There, they behave as hyperosmotic osmoconformers, with hyporegulated [Cl^-] in the blood but an accumulation of short-chain carboxylic acids(Nieczaj and Zerbst-Boroffka,1993). In the initial phase (within 4 h) of acclimation, osmotic loss of water and uptake of salt were the prominent passive events. Concentrations of Na⁺, K⁺, Cl^- and of organic anions in the blood were greatly increased(Nieczaj and Zerbst-Boroffka,1993). After a few days, extracellular volume was restored. Our interest in investigating whether there is long-term acclimation of transintegumental ion transport during extended exposures to high-saline conditions arose from these findings. We exposed leeches for several days to a high-salinity environment (200 mmol l^-1 NaCl) to elicit an osmoregulatory long-term adaptation of Na⁺ uptake; the annelids appeared to cope with this excessive physiological stress. The plasma Na⁺ concentration of Hirudo medicinalis is approximately 115 mmol l^-1 (Zerbst-Boroffka and Wenning, 1986), and an unhindered influx of Na⁺from the 200 mmol l^-1 Na⁺ environment would therefore change the physiological situation from absorption to excretion of excess Na⁺.

For comparative purposes, dorsal segments of these high-salt-adapted and of pondwater-adapted integuments were dissected for Ussing chamber experiments. The preparations were voltage-clamped, and the initial electrophysiological properties determined to expose any differences attributable to high-salt-adaptation at the level of integumental Na⁺ conductances. We measured total transepithelial Na⁺ uptake and the amiloride-sensitive Na⁺ current.

Extracellular nucleotides serve as agonists for a variety of membrane receptors. Besides ionotropic P2X-type receptors, which comprise intrinsic ion channels, two other large families of G-protein-coupled purine receptors, the P2Y- and the P1-receptors with a multitude of subtypes, have been characterized and cloned (Ralevic and Burnstock, 1998; Surprenant et al., 1995). Expression of these receptors has been detected in a variety of epithelia in which they function in the para- and/or autocrine control of transepithelial ion-transport processes(Ralevic and Burnstock, 1998). ATP, for example, may be released from epithelial cells under physiological conditions (Schwiebert, 1999). A reciprocal regulation, stimulation of Cl^- secretion accompanied by a reduction in Na⁺ absorption, by extracellular trinucleotides may be a common and cytosolic Ca²⁺-dependent principle of ion-transport control, e.g. in renal or pulmonary epithelia(Cuffe et al., 2000; Hayslett et al., 1995; McCoy et al., 1993). Nucleotides can regulate NaCl transport via P2Y- and P2X-type receptors (McCoy et al.,1999). The distribution of some epithelial purinoceptors is restricted to either the apical or the basolateral membrane, and this accounts for the `side-specific' regulatory effects of extracellular nucleotides(Casavola et al., 1996, 1997). In the present study, we evaluate the regulatory impact of intercellular nucleotide messengers on the control of ion transport across the leech integument. For this purpose, we measured the effects of adenosine, ATP and the pyrimidine UTP on Na⁺ transport and also investigated whether the removal of apical Ca²⁺, which is known greatly to stimulate transintegumental current(Prusch and Otter, 1977; Weber et al., 1995),interferes with purinergic control. Examination of pondwater- and high-salt-adapted integuments indicated different sensitivities to extracellular nucleotides.

Hirudo medicinalis L. were obtained from Zaug (Biebertal, Germany)and were kept without food for 4-10 days either in artificial pondwater (APW;1 mmol l^-1 NaCl) or in high-salinity water (HSW; 200 mmol l^-1 NaCl) at room temperature (18-24 °C). Before preparation,animals were made torpid by exposure to low temperature (0 °C).

A ventral incision was made in the leech body wall, and the intestine was detached and muscular layers were carefully scraped off. The dorsal integument of the subclitellar region was fixed on a needle-spiked ring. The Ussing chamber had an aperture of 0.5 cm². The edges of the tissue were sealed with silicone grease. During experiments, both compartments of the Ussing chamber were continuously perfused (apical, at approximately 7 ml min^-1; basolateral, at approximately 3 ml min^-1). All experiments were performed at room temperature (18-24°C). A preparation of the dorsal integument (subclitellar region) was mounted in an Ussing type chamber. After measurement of the initial potential, V_init, the transepithelial potential, V_T, was allowed to equilibrate and then clamped to 0 mV. The short-circuit current (I_sc) was recorded continuously,and the transepithelial resistance (R_T) was calculated from the effects of 20 mV pulses on I_sc. The amiloride-sensitive current (I_ami) was measured as the decrease in I_sc in the apical presence of apical 100μmol l^-1 amiloride. I_Na, the transepithelial Na⁺ current, was when apical NaCl was substituted with equimolar tetramethylammonium chloride (TMA-Cl). Readdition of apical Na⁺re-established the I_sc.

The basolateral Ringer's solution contained (in mmol l^-1): 115 NaCl, 4 KCl and 1.8 CaCl₂. In the apical solution, KCl was replaced with TMA-Cl. In Na⁺-free solutions, NaCl was substituted by equimolar concentrations of TMA-Cl. Ca²⁺-free solutions contained 0.5 mmol l^-1 EDTA. All solutions were buffered with 5 mmol l^-1 Hepes and adjusted to pH 7.4 with Tris (Trizma-base).

APW contained (in mmol l^-1): 1 NaCl, 0.05 KCl, 0.4 CaCl₂ and 0.2 NaHCO₃, pH 7.4. HSW contained (in mmol l^-1): 200 NaCl, 1 KCl, 0.4 CaCl₂ and 0.2 NaHCO₃, pH 7.4.

Adenosine, ATP, UTP, cyclic AMP [8-(4-chlorophenylthio)-cAMP, cpt-cAMP] and 3-isobutyl-1-methyl-xanthine (IBMX) were obtained from Sigma.

Ag/AgCl wires in 1 mol l^-1 KCl served as current- or voltage-measuring electrodes. For conductive connection to bathing compartments, 1 mol l^-1 KCl/agar bridges were used. During measurements, the transepithelial potential was voltage-clamped to 0 mV(voltage-clamp amplifier; Nagel, Munich, Germany). The short-circuit current(I_sc) was recorded continuously on a stripchart recorder and on computer (Apple IIsi, MacLab interface, chart recorder program: Analog Digital Instruments). R_T was calculated according to Ohm's law from changes in I_sc during superimposed 20 mV pulses.

Data are presented as means ± S.E.M. N is the number of experiments and animals. Statistical analyses were performed using paired or independent Student's t-tests (Microcal Origin).

After an equilibration period, the electrophysiological parameters V_T, I_sc and R_Twere measured from both groups of adapted integuments(Table 1). The values of most of the initial variables were significantly reduced in HSW-adapted compared with APW-adapted integuments. Interestingly, although I_scand total I_Na were greatly decreased by adaptation to 200 mmol l^-1 NaCl, I_ami was not reduced significantly. Removal of apical Ca²⁺ increased I_sc by a factor of approximately 1.4 in APW-adapted and approximately 2.4 in HSW-adapted integuments(Table 1). The contribution of I_Na to I_sc increased from 40 % in APW and 48 % in HSW preparations to 75 % and 70 %, respectively, under Ca²⁺-free conditions (Table 1); however, the amiloride-sensitive portion of total I_Na remained at approximately 60 % for both groups. The non-Na⁺ currents, calculated as the I_sc—I_Na, even increased slightly in HSW-adapted integuments but decreased in the APW-adapted integuments upon removal of apical Ca²⁺.

We tested different concentrations of ATP for their effect on transepithelial ion conductances of APW- and HSW-adapted integuments. ATP was applied to the basolateral side at concentrations of 10 μmol l^-1to 10 mmol l^-1 (Fig. 1). Maximal stimulation was observed in all preparations with 1 mmol l^-1 ATP, giving a concentration required for 50 % stimulation(EC₅₀) of approximately 80 μmol l^-1. Exposure to 1 mmol l^-1 ATP did not affect I_sc in APW-adapted integuments (98.7±6.9 %; N=5)(Fig. 2). In HSW-adapted integuments ATP induced a twofold (214.7±42.1 %) increase in I_sc, which resulted from the activation of Na⁺conductances (Fig. 2). This, in turn, could be attributed to a more than threefold (364±103.9 %)increase in the amiloride-sensitive current. In comparison, in APW-adapted integuments, I_ami increased only slightly (by 187.5±33 %) at the cost of other Na⁺ conductances (not shown). The ATP-induced I_ami was larger than the subsequently determined net Na⁺ transport (I_Na; Table 2). This paradoxical finding may be explained by additional effects of apical Na⁺-free conditions on, for example, Cl^- conductances. Higher concentrations, such as 10 mmol l^-1 ATP, partially deactivated I_sc again. This downregulation phenomenon in the presence of 10 mmol l^-1 ATP was observed in all experiments (data not shown). Apical application of ATP (1 mmol l^-1) also greatly stimulated I_sc, a result that was not further investigated.

Under apical Ca²⁺-free conditions, ATP had no effect on I_sc across APW-adapted integuments (112.8±6.8 %). In contrast, upregulated I_sc across HSW-adapted integuments under Ca²⁺-free conditions responded to 1 mmol l^-1 ATP with a significant increase to 136.3±6.4 %. ATP preferentially stimulated Na⁺ uptake and had very little effect on non-Na⁺ conductances. This effect of ATP on transepithelial Na⁺ transport in HSW-adapted integuments was largely due to an upregulation of I_ami by a factor of 235.9±49.5 %,whereas I_ami increased to 185.4±27.9 % in APW-adapted integuments (not shown).

Certain types of P2Y receptor selectively bind uridine nucleotides but not adenosine nucleotides (Lazarowski and Boucher, 2001; Ralevic and Burnstock, 1998). UTP (1 mmol l^-1) was added either to the apical or basolateral compartment and tested for its effect on macroscopic transepithelial ion conductances across APW- and HSW- adapted integuments. Apical treatment with 1 mmol l^-1 UTP left I_sc, I_Na and its amiloride-sensitive component I_ami unaffected in both integuments (data not shown). While I_ami of APW-adapted integuments responded somewhat inconsistently to basolateral UTP, the I_sc was reduced to 87 % compared with the control value and total I_Na was not significantly affected. Under apical Ca²⁺-free conditions, UTP had no effect on I_sc. In addition, the basolateral side of HSW- adapted integuments was insensitive to 1 mmol l^-1 UTP under both normal and Ca²⁺-free conditions.

We examined the effects of extracellular adenosine in the apical and the basolateral compartments. I_sc and Na⁺ currents across APW- and HSW-adapted integuments were insensitive to apical adenosine(data not shown). While I_sc, I_Na and I_ami in APW-adapted integuments were largely unaffected by basolateral application of adenosine, the I_sc of HSW-adapted integuments increased in a concentration-dependent manner(Fig. 3). A maximal, stable and non-transient activation of I_sc was observed in the presence of 10 mmol l^-1 adenosine (204±35 % of control amplitude; N=5; EC₅₀=0.17 mmol l^-1). In contrast to the effect of ATP described above, the net increase in I_sc was due to a stimulation of non-Na⁺conductances (Fig. 4). Interestingly, in HSW-adapted integuments, adenosine increased I_ami current to 188.1±18.7 % without changing total I_Na (Fig. 4). Basolateral treatment with 1 mmol l^-1 adenosine,after upregulation of Na⁺ currents by Ca²⁺-free conditions, had no significant effect on I_sc or on the component of I_sc carried by Na⁺. I_ami, however was slightly reduced to 79.5±5.6 % in APW-adapted integuments, whereas it increased in HSW-adapted integuments to 175.6±65.2 % in response to adenosine. The size of this effect varied substantially, but it was observed in all preparations tested.

Table 2 summarizes the currents and the effects of apical Ca²⁺-free conditions in the basolateral presence of 1 mmol l^-1 ATP, UTP and adenosine. Transepithelial ion conductances have been reported to be regulated by extracellular adenosine and through cyclic AMP/protein kinase A (PKA)signalling mechanisms. We treated APW- and HSW-adapted integuments in the absence of apical Ca²⁺ with membrane-permeant 100 μmol l^-1 cpt-cAMP in combination with 1 mmol l^-1 IBMX (an inhibitor of phosphodiesterases). However, this effect of cyclic AMP was abolished under Ca²⁺-free conditions. Both I_Naand the non-Na⁺ current were unaltered by cyclic AMP(Fig. 5). The amiloride-sensitive part of I_Na did not change in APW-adapted (94±6 %) or HSW- adapted integuments (102±8 %). The apical epithelial Na⁺ channels were therefore not affected.

Hirudo medicinalis proved to be reasonably tolerant of transfer to a high-salinity environment (Boroffka,1968; Nieczaj and Zerbst-Boroffka, 1993). The initial phase, consisting of loss of water and accumulation of Na⁺, Cl^- and organic anions,is followed by long-term acclimation, as for a hyperosmotic osmoconformer(Nieczaj and Zerbst-Boroffka,1993). After several days, plasma volume is restored, while intracellular water content is not(Nieczaj and Zerbst-Boroffka,1993). How this physiological acclimation affects transintegumental ion transport was the subject of the present study.

Pondwater presents the leech integument with the problem of absorbing ions against considerable electrochemical gradients. The apical transport of Na⁺ requires energy and may be driven by proton-motive forces from,for example, H⁺ V- ATPases(Ehrenfeld and Klein, 1997; Nelson and Harvey, 1999). In high-salinity conditions the situation for overall ion regulation is reversed. Extracellular 200 mmol l^-1 Na⁺ exceeds plasma concentrations (115-140 mmol l^-1)(Prusch and Otter, 1977; Zerbst-Boroffka and Wenning,1986) and should therefore result in increased levels of body salts. Shutdown of apical Na⁺ conductances and extrusion of excess salt by, for example, nephridia may then balance total Na⁺ uptake(Zerbst-Boroffka and Wenning,1986). To determine when long-term (transcriptional and/or translational) regulation of ion transport sets in, one would have to perform a series of time-course experiments that specify time required for changes in electrical properties of the leech integument. However, such an approach is difficult because this epithelium shuts down ion transport in the presence of high apical Na⁺ concentrations. Na⁺ uptake recovers soon after a return to low-[Na⁺] conditions(Weber et al., 1995; M. Schnizler, personal observations). The recovery of Na⁺ transport after a shift from apical high-Na⁺ conditions to low-Na⁺conditions is repeatably reproducible and the phenomenon remains unchanged even after exposure to high-Na⁺ conditions for more than 5 h. We therefore decided to extend the high-salt challenge over several days, by which time long-term adaptive processes could be expected to have occurred(Nieczaj and Zerbst-Boroffka,1993). Recent models attribute triggering of the autoregulative feedback inhibition of Na⁺ transport in tight epithelia to an elevation in intracellular [Na⁺](Palmer et al., 1998; Turnheim, 1991). The phenomenon is a short-term adaptive process that starts immediately and is reproducible for several cycles of transfer from low to high salinity. There is evidence for Na⁺-sensing receptors that mediate the endocytosis of epithelial Na⁺ channels and their subsequent degradation(Hubner et al., 1999; Ishibashi et al., 1999; Komwatana et al., 1998). The feedback inhibition is therefore due to a reduction in the total number of surface-expressed channels. We expected the high-salt exposure of several days not only drastically to reduce transcellular Na⁺ transport but also to affect the number of new vesicle-stored channels available for apical membrane insertion. Long-term adaptation to high-salt conditions reduced V_T and I_sc(Table 1). However, although transepithelial Na⁺ uptake was downregulated, the amiloride-sensitive transcellular Na⁺ current(I_ami) was not affected(Table 1).

High extracellular Na⁺ concentrations are generally believed to elevate cytosolic Ca²⁺ concentrations which, in turn, inhibit the amiloride-sensitive Na⁺ channels, either directly or by protein-kinase-C-mediated mechanisms(Garty and Palmer, 1997). Despite persistent evidence for the importance of Ca²⁺ in short-term Na⁺ feedback inhibition, their direct involvement could not be verified (Abriel and Horisberger,1999). It is not clear whether the downregulation of transcellular Na⁺ uptake by apical Ca²⁺ results from interactions with elements in the apical membranes or by Ca²⁺ entering the cells and increasing cytosolic Ca²⁺ concentrations(Abriel and Horisberger, 1999; Palmer et al., 1998; Turnheim, 1991). In addition to a direct and possibly non-specific action upon Na⁺ channels or regulatory factors, G-protein-coupled Ca²⁺-sensing mechanisms may come into play (Komwatana et al.,1998). Surface expression of such Ca²⁺ receptors has been reported for several tissues (Brown et al., 1993; Riccardi,2000), but their presence in the leech integument must first be verified.

One objective of our study was to investigate whether longterm acclimation to high-salt conditions affects this short-term regulation of transcellular Na⁺ uptake and its sensitivity to apical Ca²⁺. Both high-salt- and pondwater-adapted integument responded to the removal of apical Ca²⁺ with an increased I_sc to values of similar magnitude, and this was mainly due to an upregulation of I_ami (Table 1). Thus, the amiloride-sensitive Na⁺ conductances are partly inhibited by the extracellular presence of Ca²⁺. This stimulating effect of Ca²⁺ removal in the leech skin may be correlated with downregulated Na⁺ transport since only under apical high Na⁺-conditions could I_sc be stimulated by the removal of apical Ca²⁺(Weber et al., 1995).

Transcellular transport of Ca²⁺ occurs in a variety of epithelia, and apical entry of Ca²⁺ greatly affects intracellular Ca²⁺ concentrations (Hoenderop et al., 2000). In vertebrates, for example, there are epithelial Ca²⁺ channels that lose their selectivity and transport monovalent cations in the absence of extracellular Ca²⁺(Vassilev et al., 2001). Total removal of apical Ca²⁺ can be expected to reduce cytosolic[Ca²⁺]. Basolateral electrogenic 3Na⁺/1Ca²⁺exchange processes are considerably downregulated in this situation, and this may explain upregulation of transcellular Na⁺ transport in Ca²⁺-free conditions (Blaustein and Lederer, 1999; Hoenderop et al., 2000). Interestingly, the stimulating effect of a removal of Ca²⁺ was greater in HSW-adapted than in APW-adapted integuments,but the final upregulated currents from HSW-adapted integuments were equal to those from APW-adapted integuments (Table 1). We do not know the extent to which vesicular translocation of new channels contributed to the effect of Ca²⁺, but as a result of long-term adaptation, one would expect a marked reduction in de novosynthesis of channels rather than the formation of channel-containing membrane vesicles. The rapid upregulation observed indicated in situactivation of silent, but surface-expressed, Na⁺ channels.

Locally released ATP has a direct paracrine and/or autocrine effect modulating epithelial transporters and channels. Reciprocal regulation of Cl^- secretion and Na⁺ absorption by extracellular nucleotides appears to be a common feature of mammalian respiratory and renal epithelia (Cuffe et al., 2000; Mall et al., 2000; McCoy et al., 1999). ATP and UTP stimulate Cl^- secretion, whereas apical Na⁺ entry is downregulated (Cuffe et al.,2000; Inglis et al.,1999). In frog skin epithelium, the classic model for NaCl absorption from fresh water, ATP and UTP affect the short-term regulation of transepithelial ion transport (Brodin and Nielsen, 2000b). However, in this case, transcellular amiloride-sensitive Na⁺ transport increased in response to extracellular nucleotides, but Cl^- conductances seemed not to be affected (Brodin and Nielsen,2000a). Testing whether trinucleotides mediate transintegumental ion regulation in our limnic invertebrate, we found a different situation. We were unable to detect any significant ATP sensitivity of I_sc in APW-adapted integuments. In contrast, adaptation to high-salt conditions increased the sensitivity of I_sc to 1 mmoll^-1 ATP applied from the basolateral side(Fig. 2). Adaptation to high-salt conditions resulted in the recruitment of ATP-triggered transduction mechanisms that are not present or remain silent in a pondwater environment. I_ami was upregulated in integuments adapted to both conditions although the HSW-adapted preparations responded more strongly(Fig. 2, Table 2). The inhibition of this basolateral effect of ATP by amiloride clearly identified epithelial Na⁺ channels localized in the apical membrane as the upregulated target, suggesting the involvement of transduction mechanisms that trigger the release of diffusible second messengers. The time course of upregulation of I_sc was a typical response to ATP(Fig. 1), supporting the idea of a slower messenger-mediated process.

The principal agonists for P2X receptors appear to be ATP and its derivatives, but not UTP. Neither basolateral nor apical application of 1 mmoll^-1 UTP had a dramatic effect on I_sc or I_Na in integuments adapted to both conditions(Table 2), which suggests the existence of P2X-mediated processes. Following activation by ATP, ligand-gated P2X receptors open their intrinsic non-selective cation channel, allowing Na⁺ and Ca²⁺ to pass from the extracellular fluid into the cell. Cellular release of ATP therefore provides one mechanism for controlling intracellular Ca²⁺ concentration and downstream transduction processes (Bean,1992; Dubyak and el-Moatassim,1993). In general, P2X- and P2Y-receptor-induced regulation of transepithelial Na⁺ absorption and/or Cl^- secretion is mediated by an increase in cytosolic [Ca²⁺](Cuffe et al., 2000; Mall et al., 2000; McCoy et al., 1999) as a result of extracellular Ca²⁺ entering the cell via P2X receptors or a P2Y-induced release of Ca²⁺ from intracellular stores. However, the absence of Ca²⁺ from apical solution did not prevent the stimulating effect of basolateral ATP on Na⁺conductances in HSW-adapted integuments(Table 2). I_ami was again upregulated, although the response was less pronounced. A significant contribution of Ca²⁺ influx to the transduction of purinergic stimulation of I_ami cannot be ruled out since, for example, the previously observed upregulation of I_sc or the total I_Na in HSW-adapted integuments by ATP are reduced or in some cases even prevented after removal of apical Ca²⁺. A further explanation may be that Ca²⁺greatly prestimulates amiloride-sensitive Na⁺ conductances, leaving fewer quiescent apical Na⁺ channels to be activated in situ by further, i.e. purine-induced, mechanisms. Hydrolysis or the conversion of exogenous nucleotides by ecto-nucleotidases(Harden et al., 1997) may be one explanation for the requirement of concentrations as high as 1 mmoll^-1 ATP to evoke the maximal effect.

At present, no agonists or antagonists are available that discriminate effectively between the families and subtypes of P2X and P2Y receptors(Norenberg and Illes, 2000; Ralevic and Burnstock, 1998; von Kugelgen and Wetter, 2000)and little is known about invertebrate orthologues and their pharmacology. Thus, it is difficult to characterize the basolateral leech ATP receptors and to prove a correlation between their relative expression levels and environmental salinity using northern blots or quantitative reverse transcriptase/ polymerase chain reactions.

Adenosine has been reported to have pleiotropic actions in epithelia(McCoy et al., 1993; Olah and Stiles, 1992). Na⁺ uptake by tight epithelia, e.g. in the medullary collecting duct cells of the rat kidney, is controlled by extracellular adenosine(Yagil, 1994). In A6 cell monolayers (cells from Xenopus laevis nephron), A1 receptors are located on the apical surface and regulate Cl^- secretion or inhibition of transepithelial Na⁺ transport viastimulation of phospholipase C activity. On the basolateral surface, there are A_2A receptors, which stimulate Na⁺ uptake (Casavola et al., 1996, 1997). This positive effect is believed to be due to upregulation of cyclic AMP production followed by stimulation of Na⁺/H⁺ exchange activity. The restricted distribution of receptors to the apical or basolateral membrane and the distinct post-receptor mechanisms are both responsible for the dual-control regulation by adenosine of transepithelial ion conductances.

In the leech integument, addition of adenosine to the apical compartment failed to have any effect on macroscopic I_sc. Instead, we found that activation of basolateral adenosine receptors only stimulated I_sc in HSW-adapted integuments and not in pondwater-adapted controls. Net Na⁺ transport was not upregulated,and we assume that Cl^- conductances are targets for this effect of adenosine (Fig. 4). The slope of the current traces indicated that this activation of HSW-adapted integuments was a monophasic but rather slow process(Fig. 3). Interestingly, the amiloride-sensitive portion of I_Na was enlarged at the expense of other Na⁺ conductances. This effect was abolished under Ca²⁺-free conditions.

After ligand-binding by certain purinoceptors, the signal is propagated in some cases by cyclic AMP/PKA-mediated pathways(Ralevic and Burnstock, 1998). Under normal conditions, membrane-permeant cyclic AMP activates amiloride-sensitive currents across leech integuments(Weber et al., 1995). We applied cpt-cAMP after the effect of the removal of Ca²⁺ had concluded, but could detect no convincing effect of this compound on transintegumental ion conductances, however(Fig. 5). I_ami across APW-adapted integuments was slightly reduced rather than stimulated, and I_ami across HSW-adapted integuments was unchanged. Interestingly, the overall Na⁺ current, I_Na was unaltered by the addition of cpt-cAMP(Fig. 5), possibly indicating that apical Ca²⁺-free conditions had stimulated the Na⁺conductances to their maximum. High concentrations of adenosine were required to induce the maximal response, so mechanisms other than binding to metabotropic receptors must be taken into consideration. The nucleosides may have entered the cells and initiated further processes.

In conclusion, the leech integument provides a useful model for investigating the regulation of ion transport in an invertebrate tight epithelium. Adaptation of leeches to high-salinity conditions altered most of the initial electrophyiological properties of the integuments but left the amiloride-sensitive Na⁺ current unaffected. Absence of apical Ca²⁺ stimulated the bioelectrical activities in integuments adapted to high-salinity conditions and pondwater to a similar extent. Long-term exposure to high (200 mmoll^-1) NaCl concentration conferred ATP-sensitivity to the I_sc and the I_Na, although there was little effect on control integuments. Our findings provided no evidence that UTP has any relevance in the control of ion-transport processes across the leech integument. Adenosine upregulated transepithelial non-Na⁺ currents in HSW-integuments,while APW-adapted preparations were largely unaffected by this nucleoside. The surface expression of adenosine/P1 receptors was restricted to the basolateral membrane. These disparities in sensitivity to extracellular purines may be caused by differentially expressed receptors or regulatory factors during adaptation to high-salt conditions. Finally, one can argue that high-salt stress is a situation that a leech will never face under natural conditions. Nevertheless, exposure to such physiological challenges provides an opportunity for detecting variances of macroscopic variables, for instance transepithelial ion conductances, which may indicate assimilative processes in a whole-animal model and initiate more fine-tuned approaches in future projects.

This study was supported by the Deutsche Forschungsgemeinschaft DFG. We thank Margret Hündt, Siegfried Kristek and Bernhardt Kahnert for excellent technical assistance.