Intracellular Free Calcium Measured with Calcium-Selective Microelectrodes From Salivary Gland Cells of the Snail Planorbis Corneus

The importance of Ca²⁺ for the process of secretion by gland cells was first recognized by Douglas & Rubin (1961). Since then the idea that an increase in the intracellular concentration of free Ca²⁺ ([Ca²⁺]_i) within gland cells is an event essential for secretion has become accepted generally (Case, 1984). However, the source of the Ca²⁺ and the means by which [Ca²⁺]_i is regulated are often controversial. In the case of the exocrine pancreas, for example, it has been proposed that increases in [Ca²⁺]_i arise through an influx of Ca²⁺ across the contraluminal membrane (e.g. O’Doherty & Stark, 1982; Kanno et al. 1984), the release of Ca²⁺ from intracellular stores (e.g. Hunter, Smith & Case, 1983; Ochs, Korenbrot & Williams, 1983; Streb, Irvine, Berridge & Schulz, 1984) or a combination of these two processes (e.g. Petersen & Maruyama, 1983; Iwatsuki & Nishiyama, 1984). One reason why such questions concerning [Ca²⁺]_i are often difficult to answer is that the measurement of [Ca²⁺]_i in the cytoplasm of living cells is a challenging technical problem (Thomas, 1982; Tsien & Rink, 1983), particularly in gland cells because of their generally small size.

This paper presents the results of an investigation into the factors influencing [Ca²⁺]₁, in salivary gland cells of the pond snail Planorbis corneus. Previous studies have investigated the effects of acetylcholine (ACh) upon the cells (Barber, 1985) and upon the ion concentrations in the extracellular fluid bathing the glands (Barber & ten Bruggencate, 1985b). In the present study, [Ca²⁺]_i was found to be raised by brief exposure to ACh, followed by recovery to the resting level. The source of the additional [Ca²⁺]_i, and the means by which it is removed were investigated. Some of the results given below have been presented to a meeting of the German Physiological Society (Barber & ten Bruggencate, 1985a).

Specimens of Planorbis comeus were purchased from commercial suppliers and kept in the laboratory at room temperature (20–28°C) in aquaria. They were fed ad libitum on lettuce and remained active and to all appearances healthy under these conditions for many months. Pairs of salivary glands were isolated from snails as described previously (Barber, 1985). Only snails with a shell diameter greater than 20 mm were used in the present experiments because intracellular recording with Ca²⁺-sensitive electrodes was found to be easier in glands from larger snails.

Isolated glands were transferred to a recording chamber (volume 0·2 ml) which was perfused continuously at a rate of 7·5 ml min⁻¹ with snail saline at room temperature. Normal physiological saline contained (in mmol 1⁻¹) NaCl, 39; KC1, l⁻3; CaC1₂,4·5; MgCl₂ l·5; and NaHCO₃, 7·0 and was bubbled with a CO₂/O₂gas mixture until it had a pH between 7·5 and 7·2. ‘Ca²⁺-free’ saline was made by substituting MgCl₂ for CaCl₂ and adding 1 mmol 1⁻¹ EGTA. Low-Na⁺ saline, in which LiCl was substituted for NaCl, was also used. Acetylcholine bromide (Sigma) was dissolved in these salines and perfused through the chamber for periods of 10 s.

Ca²⁺-selective microelectrodes were made in a manner similar to that of Grafe, Rimpel, Reddy & ten Bruggencate (1982). Briefly, micropipettes were pulled from theta capillary tubing so as to have a tip diameter of 0·5 μm or less. The insides of the tips of the ion-selective barrels were then silanized with hexamethyl disilazane (Sigma) and filled with a small column of Fluka 21048 (ETH 1001) ion-exchange ligand. The shank and barrel of the ion-selective side of the electrode were then filled with 200 mmol 1⁻¹ CaCl₂ and the reference barrel was filled with 1 mol 1⁻¹ potassium acetate. Both barrels of the electrode were connected via chlorided silver wires to the input stage of a high impedance amplifier. The membrane potential (E_M) of impaled cells was measured with the reference barrel. This potential (V_ref) was also subtracted simultaneously from the potential from the ion-selective barrel to give the pure Ca²⁺ signal (V_ca). An agar bridge served as the bath ground electrode.

Electrodes were calibrated upon being withdrawn from the cell in Ca²⁺-buffered solutions made as described by Tsien & Rink (1980). They had a mean slope of 29mV/decade between 10⁻³ and 10⁻⁶moll⁻¹ Ca²⁺, 15mV/decade between 10⁻⁶ and 10⁻⁷moll⁻¹ Ca²⁺, and 6mV/decade between 10⁻⁷ and 10⁻⁸moll⁻¹ Ca²⁺, (N = 13). Ca²⁺-electrodes were tested for their sensitivity to Na⁺ and Li⁺ by adding 3 mmol 1⁻¹ NaCl or LiCl, respectively, to 10⁻⁷moll⁻¹ Ca²⁺ buffer (measurements with Na⁺-selective microelectrodes have indicated that the intracellular Na⁺ concentration in Planorbis salivary gland cells rises by a mean of 3 mmol 1⁻¹ after stimulation with ACh; A. Barber, in preparation). This produced a change in V_Ca of around +0·5 mV in the case of Na⁺, and +1 mV in the case of Li⁺. In the following, [Ca²⁺]_i is expressed in terms of concentration rather than activity (see Thomas, 1982, for a discussion of this point).

The E_M and [Ca²⁺]_i recorded from salivary gland cells with Ca²⁺-selective microelectrodes stabilized slowly over the course of the 10–20 min following impalement, as shown in Fig. 1. The mean [Ca²⁺]_i measured after this ‘sealing in’ period was 1· × 10⁻⁷ mol 1⁻¹ (S.D. ± 0· 84 ×10⁻⁷, N = 9), while the E_M of these cells was −67·1 mV (s.D. ± 4·1) (see Table 1). Since the extracellular fluid contains 4·5 mmol 1⁻¹ Ca²⁺, such a level of [Ca²⁺]₁ indicates that the equilibrium potential for Ca²⁺ in these cells should lie around +132 mV, as calculated from the Nemst equation. Ca²⁺ is therefore subjected to an inward driving force of around +200 mV. Measurements of EM were also made with the large number of electrodes whose ion-selective barrels were found to be either insensitive, too noisy, or too slow to be useful for measuring [Ca²⁺]_i. The mean value of EM made with these electrodes was −68-·0mV (s.D. ± 7·2, N = 80).

Planorbis salivary gland cells are known to respond to the brief application of ACh to the bathing medium by producing an electrical response in which a depolarization is followed by a hyperpolarization of the contraluminal membrane (Barber, 1985; Barber & ten Bruggencate, 1985a,b). In the present study it was found that the depolarizing phase of the response to ACh (mean amplitude 49·7mV, S.D. ± 6·3,N=12) was associated with an increase in [Ca²⁺]_i by 2·9× 10⁻⁷ mol 1⁻¹ (S.D. ± l·0x 10⁻⁷, N= 12), after which the original [Ca²⁺]_i was restored slowly during the subsequent hyperpolarization (Figs 2, 3, 4).

The ACh-induced increase in [Ca²⁺]_i was abolished (Fig. 2), or much reduced in size, when glands were bathed in ‘Ca²⁺-free’ saline (N =4). The saline used here probably contains less than 10⁻⁷moll⁻¹ free Ca²⁺ (Hubbard, Jones & Landau, 1968). This result clearly demonstrates that the application of ACh leads to an influx of Ca²⁺ from the extracellular fluid, rather than causing the release of Ca²⁺ from intracellular stores. The record shown in Fig. 2 also indicates that the Cab electrodes are in fact making accurate measurements of changes in [Ca²⁺]₁, and are not being influenced to any significant extent by influxes of interfering ions, such as Na⁺ (see Discussion).

Further experiments indicated that Ca²⁺ entry stimulated by ACh did not occur through voltage-dependent Ca²⁺ channels, such as have been described in the gland cells of other gastropods (Kater, 1977; Hadley, Murphy & Kater, 1980; Goldring, Kater & Kater, 1983) and the leech (Marshall & Lent, 1984). Two experiments were performed in which glands were first stimulated with ACh in normal saline, and then in saline which contained 4 mmol 1⁻¹ Co²⁺ in order to block voltage-dependent Ca²⁺ channels (Edwards, 1982). Spontaneous action potential activity was blocked rapidly and reversibly by Co²⁺ (not shown), but the ACh-induced increase in [Ca²⁺], was not influenced (Fig. 3).

The means by which [Ca²⁺]_i is brought back to its original level following stimulation with ACh remains unknown, though this process was found to be inhibited (N = 3) when Na⁺ in the extracellular fluid was replaced to a large extent (85%) by Li⁺ (Fig. 4). In this respect the salivary glands of Planorbis resemble snail neurones (Aldenhoff & Lux, 1985) in which Li⁺ has been shown to interfere with the regulation of [Ca²⁺]_i. Li⁺-saline also made the changes in [Ca²⁺]_i following the application of ACh larger (Fig. 4). Whatever the mechanisms of these effects (see Discussion), both the baseline [Ca²⁺]_i and the size of the ACh-induced change in [Ca²⁺]i returned to their previous level after a 20 min wash in normal saline. Another effect of Li⁺-saline was to reversibly reduce the hyperpolarizing phase of the response to ACh (Fig. 4). Such an action of Li⁺ was to be expected because this hyperpolarization is produced by the activity of an electrogenic Na⁺/K⁺ pump (Barber, 1985; Barber & ten Bruggencate, 1985a,b), and such pumps are known to be inhibited when extracellular Na⁺ is replaced by Li⁺ (e.g. Padjen & Smith, 1983).

A fundamental difficulty involved in measuring the activities of intracellular ions with ion-selective microelectrodes is the problem of leak artifacts caused by the penetration of the electrode (Thomas, 1982; Tsien, 1983). This is more important for measurements of [Ca²⁺]_i than other ions because the enormous inward driving force on Ca²⁺ means that even the smallest leak in the cell membrane will raise [Ca²⁺]_i significantly near the tip of the electrode (cf. Rose & Loewenstein, 1975). Furthermore, gland cells are known to be especially susceptible to damage caused by microelectrodes because of their small size (Petersen, 1980).

The extent to which [Ca²⁺]_i was modified as a result of such injury in the present experiments is possible to assess only indirectly. Planorbis acinar cells are certainly small (mean diameter 16·6μm, S.D. ± 3·8, N= 35; Barber, 1985), but the value of EM measured in this study with double-barrelled microelectrodes (−67·1 mV, S.D. ± 4·1, N = 9) was only slightly less than the −72mV (s.D. ±7, N= 231) recorded from cells in this preparation with conventional single-barrelled microelectrodes (Barber, 1985). This, together with the fact that stable recordings of E_M and [Ca²⁺], from apparently healthy cells were possible for periods as long as 3 h, suggests that the Ca²⁺-selective electrodes do not cause excessive damage to these cells. It may be that the electrical coupling found between neighbouring acinar cells in this preparation (Barber, 1985) allows the impaled cell to tolerate the insertion of the electrode more easily.

Another complication associated with using Ca²⁺-selective electrodes concerns cross-sensitivity to other ions, most notably Na⁺ and K⁺. This problem of interference becomes most serious under conditions where low levels of Ca²⁺ have to be measured against a relatively high background concentration of Na⁺ and K⁺, as inside living cells. However, the estimates of basal [Ca²⁺]_i given in the present study are probably not influenced significantly by Na⁺ and K⁺ because the intracellular free concentrations of Na⁺ and K⁺ are very low in these cells (mean of 2-3 and 42mmol 1⁻¹, respectively; A. Barber, in preparation).

In view of the reservations mentioned above, it is reassuring that the baseline level of [Ca²⁺]_i measured in the present study (l·24×10⁻⁷moll⁻¹, S.D. ± 0·84) is comparable to that found in a number of other kinds of gland cells (Berridge, 1980; O’Doherty & Stark, 1982; Knight & Kesteven, 1983; Ochs et al. 1983; O’Doherty, Stark, Crane & Brugge, 1983; Streb et al. 1984; Snowdowne & Borle, 1984) using several different techniques. The value of [Ca²⁺]_i in Planorbis salivary gland cells also resembles that measured recently with Ca²⁺-selective electrodes in several types of neurones and muscle cells (Tsien & Rink, 1983; Gorman, Levy, Nasi & Tillotson, 1984; Weingart & Hess, 1984). Moreover, the calculated equilibrium potential for Ca²⁺ ( + 132 mV) is similar to the +148 mV estimated indirectly to be the mean reversal potential for Ca²⁺ in neurones of Helix aspersa (Meech & Standen, 1975).

There is good reason to believe that the increases in [Ca²⁺]_i produced by the application of ACh also represent genuine responses to Ca²⁺. The observation that ACh-induced changes in [Ca²⁺]₁ are abolished (Fig. 2) or much reduced in size in Ca²⁺-free saline shows that the apparent increase in [Ca²⁺]_i is not produced by an influx of interfering ions such as Na⁺. It is also unlikely that the ACh-induced rise in [Ca²⁺]_i is caused by a net loss of water from the cytoplasm of gland cells as a result of secretion. The application of ACh also leads to an increase in the intracellular concentration of Na⁺, whereas the level of intracellular K⁺ falls, and that of Cl⁻ does not change (A. Barber, in preparation). This result would not be expected if ACh led to a general concentration of the cytoplasm of the gland cells.

The present experiments have demonstrated that the normal increase in [Ca²⁺]_i induced during stimulation with ACh is dependent on the presence of Ca²⁺ in the fluid bathing the glands. This observation is therefore good evidence that the application of ACh leads to an influx of Ca²⁺ from the extracellular space. Alternative mechanisms by which [Ca²⁺]_i could be raised, for example, by the release of Ca²⁺ from intracellular stores as a result of an influx of Na⁺ (e.g. Lowe, Richardson, Taylor & Donatsch, 1976; Watson, Farnham, Friedman & Farnham, 1981; Di Virgilio & Gomperts, 1983; Nedergaard, 1984), or the build-up of a ‘second messenger’ such as inositol-1,4,5-trisphosphate (e.g. Dawson & Irvine, 1984; Streb et al. 1984), appear to play no part in this process in Planorbis salivary gland cells. However, the participation of some form of Ca²⁺-dependent release of Ca²⁺ from an intracellular store cannot at present be ruled out.

Previous measurements of extracellular Ca²⁺ ([Ca²⁺]_e) in the salivary glands of Planorbis made with ion-selective electrodes (Barber & ten Bruggencate, 1985b) failed to reveal any reduction in [Ca²⁺]_e following the application of ACh. This is probably because extracellular Ca²⁺-selective electrodes are only able to respond to changes in [Ca²⁺]_e of more than about ± 0·1 mmol 1⁻¹ (because of the relatively high background [Ca²⁺]_e), and the changes in Ca²⁺ reported in the present study are several orders of magnitude lower than this.

Although the action of ACh on [Ca²⁺]_i is tiny in absolute terms, this increase is highly significant as a physiological effect because it represents more than a threefold increase in the resting level of [Ca²⁺]_i. Such a rise in [Ca²⁺]_i following stimulation with agonist is comparable to that described in cells of the pancreas (O’Doherty & Stark, 1982; Ochs et al. 1983), parotid gland (O’Doherty et al. 1983) and adrenal medulla (Knight & Kesteven, 1983), though larger changes in [Ca²⁺]_i have been recorded from salivary gland cells of the blowfly (Berridge, 1980) and cells of the mammalian pituitary (Snowdowne & Borle, 1984). In all but one of these examples (Ochs et al. 1983), agonist-induced elevations in [Ca²⁺]_i have been found to depend on the presence of [Ca²⁺]_e.

The evidence suggests that significant amounts of Ca²⁺ do not enter Planorbis salivary gland cells through voltage-dependent Ca²⁺ channels, because the ACh-induced increase in [Ca²⁺]_; was unaffected in saline containing 4 mmol 1⁻¹ Co²⁺ (Fig. 3). It may be that Ca²⁺ permeates through non-selective cation channels depened by ACh, as has been described in other systems (e.g. Tokimasa & North, 1984). In contrast, Ca²⁺ probably enters the endocrine GH₃ pituitary (Snowdowne & Borle, 1984) and adrenal medullary (Knight & Kesteven, 1983) cells through voltage-dependent channels. Increases in [Ca²⁺]_i can also be elicited in these cells by saline containing a high concentration of K⁺, and, in the case of the adrenal medulla, the agonist-induced elevation in [Ca²⁺]_i was found to be blocked by Co²⁺. Although there is no evidence that Ca²⁺ enters mammalian exocrine gland cells through voltage-dependent channels, the means by which it does so remain obscure. Petersen & Maruyama (1983) have proposed recently that stimulation of pancreatic acinar cells with ACh or cholecystokinin releases Ca²⁺ from an intracellular store (probably the plasma membrane) which then leads to the opening of a non-selective cation channel and an influx of Ca²⁺.

While a number of workers have successfully measured increases in [Ca²⁺]_i in gland cells as a result of stimulation with agonists, the factors influencing the recovery of [Ca²⁺]_i to its resting level after stimulation have received considerably less attention. The observation that low-Na⁺ saline inhibits the recovery of [Ca²⁺]_i following the application of ACh (Fig. 4) is interesting because this may indicate the presence of a Na⁺/Ca²⁺ exchange system in the contraluminal membrane of Planorbis salivary cells. Extracellular Na⁺ is important for extruding Ca²⁺ from cells in other preparations (e.g. Satin, 1984), and Li⁺ does not appear to be able to substitute for Na⁺ in this process (Baker, 1978). Ca²⁺ which enters gland cells from the bathing fluid may also be extruded across the luminal membrane as a component of saliva (Berridge & Lipke, 1979). Interestingly, extracellular Li⁺ has been reported to block the transport of Ca²⁺ in other preparations (Barkai & Williams, 1984).

Extracellular Li⁺ is also known to enter cells very rapidly (Thomas, Simon & Oehme, 1975; Grafe et al. 1982), where it can inhibit the uptake of Ca²⁺ into non-mitochondrial intracellular stores (Blaustein, Ratzlaff & Kendrick, 1978). Under normal conditions the uptake of Ca²⁺ into intracellular stores may be inhibited by Na⁺ (e.g. Blaustein et al. 1978; Suchard, Lattanzio, Rubin & Pressman, 1982) which permeates into Planorbis salivary gland cells through channels opened by ACh, and is removed subsequently by the activity of an electrogenic Na⁺/K⁺ pump (Barber, 1985; Barber & ten Bruggencate, 1985a,b; A. Barber, in preparation). When Na⁺ in the bathing medium is replaced by Li⁺, the efflux of Li⁺ is very low (Ehrlich & Russell, 1984), which should lead to an accumulation of intracellular Li⁺ and, therefore, a prolonged elevation of [Ca²⁺]_i However, it was not possible to distinguish between these three possible mechanisms (extrusion, secretion or uptake) in the present experiments.

Li⁺-saline also leads to a large increase in the size of the ACh-induced rise in [Ca²⁺]_i (Fig. 4). This may be because Li⁺, by inhibiting the mechanism for the extrusion, secretion or uptake of Ca²⁺, removes the brake which normally prevents [Ca²⁺]_i from rising to such high levels. The replacement of extracellular Na⁺ with Li⁺ has also been described as prolonging the mean open time of ACh channels at the vertebrate neuromuscular junction (Van Helden, Hamill & Gage, 1977) and on Aplysia neurones (Marchais & Marty, 1979), which may mean that more Ca²⁺ is able to enter Planorbis gland cells following stimulation in Li⁺-saline than in normal saline. Another consideration is that Ca²⁺-selective electrodes of the kind used in this study are more susceptible to interference from Li⁺ than Na⁺ (see Materials and Methods), so the Li⁺ that enters the gland cells through the ACh channels (Edwards, 1982) may be contributing to the Vca signal.

Financial support for this work was provided by the University of Munich. I thank Ms G. Schneider for technical assistance and Mrs C. Müller for typing the manuscript. I am also grateful to Professor G. ten Bruggencate, Dr P. Grafe and Dr N. T. Slater for their critical comments on an earlier draft of the manuscript.