Regulation of intracellular pH (pHi) following acidosis induced by NH4+/NH3 exposures was re-investigated in a crayfish stretch receptor neurone using H+- and Na+-selective microelectrodes. All experiments were performed in nominally HCO3/CO2 -free salines. From studies in Na+-free saline and from electrochemical calculations, we concluded that pHi regulation was dependent on extracellular Na+ concentration ([Na+]o ). The half-maximal rate of pHi recovery had an apparent Michaelis–Menten constant of 21mmol l 1 [Na+]o. The use of this experimental approach and an improved technique enabled us to observe pHi regulation even in Cl -free saline, in contrast to earlier findings. In addition, amiloride (2mmol l 1) inhibited the maximum rate of pHi recovery by about 80%, SITS (1mmol l 1) by about 20%.

The results strongly suggest the operation of two separate pHi-regulating mechanisms, a Na+-dependent HCO3 /Cl antiporter (probably the Na+/H+/HCO3 /Cl antiporter described earlier) and a Na+/H+ antiporter. Both mechanisms have been described in crayfish ganglion cells and muscle fibres, but the individual contribution to pHi regulation varies considerably in these preparations. Functional aspects of the pHi-regulating mechanisms in relation to ionic changes during the moulting cycle are discussed.

Several preparations from various crayfish species have been investigated to find the ion transporters responsible for regulating intracellular pH (pHi). In crayfish, Na+, HCO3 and Cl are the ions involved in pHi regulation, via a Na+/H+ and/or a Na+-dependent HCO3 /Cl exchange mechanism. Although these mechanisms operate separately in crayfish ganglion cells (Moody, 1981) and muscle cells (Galler and Moser, 1986), the pHi-regulating mechanism in the sensory neurone of the stretch receptor (Moser, 1985) seems to consist of a single combined mechanism, involving Na+/H+/HCO3 /Cl exchange or its equivalent.

During our recent investigations (Moser et al. 1989; Fresser et al. 1991; Mair, 1992) we noticed that pHi was fully regulated in nominally HCO3 -free salines. Even under such experimental conditions the operation of the HCO3 /Cl antiporter cannot be excluded since HCO3 may be formed from CO2 by the carbonic anhydrase located at the outside of this cell membrane (Kaila et al. 1992b). Complete pHi recovery and the small alkaline shift of the resting pHi value with time during experiments indicate that, in the stretch receptor neurone, pHi regulation was hardly impaired by the nominal absence of HCO3. In addition, experiments in Cl -free saline (Mair, 1992), supported by the results from Voipio et al. (1991), provide further evidence for a minor contribution of the HCO3 /Cl antiporter to pHi regulation. All these results contradict earlier findings and this caused us to re-investigate pHi regulation in the sensory neurone using a different approach and an improved technique.

In this work we demonstrate the presence of two independent pHi-regulating mechanisms, a Na+/H+ antiporter and a Na+-dependent HCO3 /Cl (probably a Na+/H+/HCO3 /Cl ) antiporter, both known in many other animals (Roos and Boron, 1981; Thomas, 1984; Chesler, 1990; Schwiening and Thomas, 1992). Some aspects of both antiporters were studied in further detail.

Preliminary results have been presented at a scientific meeting (Moser et al. 1989).

Experiments and calibration of ion-selective microelectrodes were performed at 16°C. Astacus astacus L., bought from a hatchery (Keller, Augsburg, Germany), were dissected to obtain the slowly adapting sensory neurone of the stretch receptor, which was mounted in an experimental chamber, connected to a flow-through system, as described earlier (Brown et al. 1978; Moser, 1985; Fresser et al. 1991; Mair, 1992). The flow exchanged bath volume within about 6s.

Chemicals and solutions

Chemicals of highest purity were obtained from Merck (Darmstadt, Germany), Sigma (Deisenhofen, Germany) and Fluka (Buchs, Switzerland). The amiloride analogue EIPA (ethylisopropylamiloride) was a present from Ciba-Geigy (Basel, Switzerland). The pH of all solutions was measured with an Orion 8162 (Ross-type) glass electrode. Normal Astacus saline (NAS) was modified from that devised by van Harreveld (1936) and had the following composition (mmol l− 1): 207 NaCl; 5.4 KCl; 2.4 MgCl2 ; 13.5 CaCl2 ; 10 Hepes, adjusted to pH7.4 with NaOH. In NH4+-NAS, NH4+ was substituted for 20mmol l− 1 Na+.

N-Methyl-D-glucamine (NMDG) was used as an equimolar substitute for Na+, when reduced-Na+ (e.g. 30Na+-NAS), or Na+-free (0-Na+-NAS) salines were applied. In all NMDG-containing solutions, HCl was used to adjust pH to 7.4.

Cl -free salines (0-Cl-NAS) had the following composition (mmol l− 1): 207 sodium isethionate; 5.4 potassium gluconate; 9.6 D-gluconic acid hemimagnesium salt; 54 D-gluconic acid hemicalcium salt; 10 Hepes, adjusted to pH7.4 with NaOH. For 0-Cl -NH4 -NAS, 20mmol l-1 sodium isethionate was substituted by NH4 NO3.

Measurements with Ca2+ macroelectrodes showed that calcium gluconate concentration had to be increased fourfold to obtain the free Ca2+ concentration found in NAS, owing to the substantial chelating action of the Cl substitutes used. A similar action was assumed for Mg2+, although the amount could have been overestimated. In view of our recent (N. Mair, H. Moser and F. Fresser, unpublished observations) results with varying amounts of Ca2+, an overestimation of divalent cation concentration seems less critical than an underestimation.

SITS (1mmol l− 1), a blocker of the HCO3 /Cl antiport, was added to NAS, and exposed to ultrasound just before being used. Amiloride and EIPA, blockers of the Na+/H+ antiport, were dissolved in distilled water by ultrasound and kept light-protected and cooled as a 20 or 10mmol l− 1 stock solution. Before use, the amounts for the final amiloride or EIPA concentration were taken and diluted to half the final volume with distilled water. This was made up to the final volume with double-concentrated NAS to give the correct concentration after mixing.

Electrodes

Glass for both single-(GC 150) and double-barrelled (TGC 150) microelectrodes was obtained from Clark (Reading, UK) and pulled vertically (model 700C, David Kopf Instruments, USA). Simultaneous intracellular recordings were performed with a double-barrelled microelectrode, monitoring membrane potential (Em ; and serving as an internal reference) and Na+ concentration ([Na+]; thus assuming an equal activity coefficient in the neurone and the saline), and a single-barrelled microelectrode, monitoring H+ concentration ([H+]) or pH. Signals from the Em barrel (filled with a mixture of K2 SO4 and KCl; Deisz and Lux, 1982; Moser, 1985) and from an additional bath microelectrode (filled with 3mol l− 1 KCl) were measured with an electrometer (WPI 750) against a grounded calomel electrode (Amagruss, Halmstadt-Bargen, Germany) connected to the sink of the bath through a 3mol l− 1 KCl/agar bridge. The signal of the bath electrode was subtracted from Em, which is only important during exposures in Cl -free salines, when the (external) reference potential changes.

Ion-selective microelectrodes (ISMs) were made as described by Fresser et al. (1991) and Grafe et al. (1985), using Fluka no. 71176 and no. 95293 (TDDA) as Na+ and H+ membrane cocktails, respectively. In a few cases, Fluka no. 95297 (H+ cocktail) was used, but this proved to interfere with organic blocker substances.

Na+-and H+-ISMs were calibrated in parallel, using five calibrating solutions (S1–S5) of constant ionic strength (215mmol l− 1). In the sequence S1–S5 the concentrations (mmol l− 1) for NaCl (+NaOH) were: 205 (+5), 50, 21, 10, 5; and for KCl (+KOH): 5, 157 (+8), 189 (+5), 191.5 (+13.5), 205 (+5). The buffers used were 10mmol l-1: Hepes, Hepes, Hepes, Pipes and Hepes, respectively, which, in combination with the above-mentioned base, resulted in alkaline salines which were adjusted to pH7.4, 8.0, 7.4, 6.4 and 7.4, respectively, by the addition of HCl.

Pre- and post-calibration curves of Na+-ISMs (N=50) were not significantly different and showed slopes of 54mV/pNa in the 210–21mmol l− 1 Na+ range, and 40mV/pNa in the 50–5mmol l− 1 Na+ range. A difference of 16.1mV between NAS and S1 saline indicated a selectivity value of +1.8 for Ca2+, which was higher than values in the literature (Ammann, 1986). The mean post-calibration potential value of (Fluka no. 95293) H+-ISMs was 30.9mV between pH7.4 and 8.0, and 56.1mV between pH7.4 and 6.4, these being smaller by 1.8mV and by 5.7mV, respectively, than during pre-calibration. Pre- and post-calibrations with Fluka no. 95297 H+-ISMs differed by only about 1mV in the ranges given above.

Electrical arrangement, recording and display

Electrical arrangement and recording were as reported in Fresser et al. (1991). In addition, original pen recordings were digitized, using SummaSketch II hardware, and processed with Sigma-Scan V 3.90 software, converting mV to mmol l− 1 or pH. To present the data, the graphics program Sigma Plot 4.1 (Jandel Scientific) was used. Most of the fine steps in the illustrations are due to limitations of resolution.

Measurement of the rate of acid extrusion

Maximum pHi recovery rate

In single cells and in most of our results this rate remained constant for about 1.5–4min, during which time pHi recovered by 0.3–0.5 units. For example, the same maximum recovery rate could be obtained if the measurement was taken at a pHi of 6.4 or 6.7. The main criterion for evaluation was that the ranges of maximum pHi recovery rate to be compared should overlap. Under these conditions, the error due to pHi-dependent effects on intracellular buffering capacity and on allosteric modulation of the pHi-regulating transporters should be small.

Normalized maximum rate of pHi recovery

The maximum rate of pHi recovery from the control is taken as 100%. This type of evaluation had to be used when there was no overlap in the pHi range of maximum recovery (e.g. ‘low-Na+’-experiments). This measure is much more subject to experimental error, because of the reasons given above, which may now become important.

The values given are means ± standard deviation (S.D.) from a certain number (N) of experiments, unless stated otherwise.

Controls

All experiments were performed in nominally HCO3 /CO2 -free solutions. Cells were acidified according to the NH4+/NH3 rebound technique (Boron and De Weer, 1976), and the terminology introduced by these authors will be used here.

Controls were usually performed prior to a test. A control run started with exposure to normal Astacus saline (NAS) until the variables of interest (pHi, [Na+]i, Em ) had stabilized to a reasonable extent. The cells were then exposed to 20mmoll-1 NH4+-NAS for 3min, followed by a washout in NAS. A test run consisted of a 3-min NH4+-NAS exposure with a subsequent 6-min exposure to the test saline (e.g. amiloride-NAS) and a final washout in NAS. Modifications are noted in the text.

Measurements in 231 control exposures gave a mean pHi of 7.36±0.10. During NH4+-NAS exposures the initial alkalization reached a pHi maximum of 7.70±0.10 at a rate of 0.91±0.28pHunitsmin− 1. The maximum rate of slow acidification was 0.24±0.05pHunitsmin− 1, resulting in a final pHi of 7.27±0.10 at the end of the NH4+-NAS exposure. During washout in NAS, the cells acidified at a maximum rate of 0.96±0.21pHunitsmin− 1, resulting in a maximum acidification to a pHi of 6.50±0.31, from which pHi finally recovered at a maximum rate of 0.19±0.06pHunitsmin− 1.

Intracellular Na+ concentration ([Na+]i ) was measured in 178 control experiments. At rest, [Na+]i was 18.8±9.4mmol l− 1. During fast acidification, [Na+]i increased at a maximum rate of 8.2±3.1mmol l-1 min-1, to reach a peak value of 31.1±11.0mmol l− 1 during the early period of pHi recovery. The maximum rate of [Na+]i decrease during pHi recovery was 2.9±1.3mmol l-1 min-1.

At rest the mean membrane potential (Em ) from 231 control experiments was –65.7±5.0mV. At the end of the NH4+-NAS exposure the cells had depolarized to –55.3±6.1mV.

Effect of 0-Na+-NAS on pHi recovery

In the control exposure shown in Fig. 1, pHi recovered at a maximum rate of 0.23pHunitsmin-1 and reached its original value after about 20min. In a subsequent experiment, exposure to 0-Na+-NAS, immediately after NH4+-NAS, resulted in a block of pHi recovery. Twenty-three similar experiments were performed in 16 different cells, the exposures to 0-Na+-NAS lasting 6 or 12min. Five different types of pHi response were found, as shown in Fig. 2 and these were studied in greater detail. By calculating the (individual) electrochemical equilibrium pHi (pHi = pHe + 0.0174Em, where pHe is extracellular pH), we were able to decide whether the responses were below or above equilibrium. The majority of cells (N=7) showed a type 1 response, i.e. a slow increase in pHi even in 0-Na+-NAS. This pHi change generally slowed down with time, revealing a passive movement of H+, until electrochemical equilibrium was achieved. In other cells (N=4) we found type 2 and type 3 responses where a stable pHi value was finally obtained. In these cells the measured pHi was close to, or identical with, the calculated electrochemical equilibrium pHi. A type 4 response, also seen in Fig. 1, was found in three additional cells. The measured pHi values were clearly above their electrochemical equilibrium so that the H+ gradient remained inwardly directed. It was characteristic of the type 4 response that acidification in 0-Na+-NAS proceeded with time, but became slower. Only 2 of 23 exposures exhibited a type 5 response, in which some pHi regulation may have taken place in 0-Na+-NAS.

Electrochemical considerations of H+ gradients with respect to the calculated H+ equilibrium are sufficient to explain 21 (out of 23) pHi responses observed during 0-Na+-NAS exposures. The results strongly suggest a dominant role of Na+ in the pHi-regulating mechanism in the sensory neurone. In addition, the type 1 response shown in Fig. 2 suggests either a permeability of the plasma membrane to H+ or OH or indicates that some of the pHi-regulating mechanism is driven by the H+ gradient.

Approximately 30% more acidification occurred in 0-Na+-NAS than in controls (Fig. 1). Releasing the block on pHi regulation by addition of NAS resulted in an accelerated rate of pHi recovery compared with controls.

The slower pHi recovery rate in controls could arise from the fact that they still contain intracellular NH4+ ([NH4+]i ), which exerts a prolonged acidifying action. However, we exclude this argument since, at the moment of maximum acidification, all [NH4+]i has probably left the cell as NH3 (see Fig. 5 in Fresser et al. 1991), which is in line with the Boron and de Weer (1976) model. Instead, the acceleration of the pHi recovery rate on exposure to 0-Na+-NAS might be due partly to a steeper (outwardly directed) H+ gradient resulting from the stronger acidification and partly to a steeper (inwardly directed) Na+ gradient caused by a considerable loss of [Na+]i in 0-Na+-NAS. Such factors would indicate that a combined Na+ plus H+ gradient could act as the driving force for the operation of the (subsequently identified) Na+/H+ exchanger. Further results (Moser et al. 1989) support this view, since pHi recovered in cells heavily loaded with Na+, in which the transmembrane ratio [Na+]o /[Na+]i was close to 1, rendering the Na+ gradient negligible. From an electrochemical point of view, the H+ diffusion gradient could have served to drive H+ transport, mimicking some kind of pHi regulation in these cells.

Effect of 0-Cl-NAS on pHi recovery

There are two major differences in experimental method between this and previous work: (1) while Moser (1985) used CO2 /HCO3, we have applied the NH4+/NH3 rebound technique (using NH4 NO3 instead of NH4 Cl) for intracellular acidification in nominally CO2 /HCO3-free salines, thus obtaining larger intracellular acidification; (2) we have compensated for the chelating action of the Cl substitutes. As we will show, this approach brought new insight into the mechanisms of pHi regulation.

In order to decrease intracellular [Cl ], the cells were incubated in 0-Cl -NAS for a minimum of 6min, but usually for longer. After a 10-min incubation, intracellular [Cl ] should be very low (Deisz and Lux, 1982; Voipio et al. 1991) and be unable to play a role in pHi regulation, if this involves a HCO3 /Cl exchange mechanism.

Incubation in 0-Cl -NAS had minor effects on the steady-state variables, generally causing short transient changes in pHi and [Na+]i, as can be seen in Figs 3 and 4. These might indicate a short reversal in the operation of a Cl-dependent pHi-regulating mechanism or changes in cell volume. Compared with controls, the maximum intracellular acidification after NH4+-0-Cl -NAS exposure in 0-Cl -NAS was markedly reduced (by about 50%; N=15). This phenomenon will be the subject of a separate paper and is not relevant to the pHi regulation process. The main conclusion from Fig. 3 is that pHi recovered even in the absence of Cl, supporting the findings of Voipio et al. (1991). Fig. 4, showing [Na+]i measurements, suggests that Na+ was directly involved in pHi regulation even in 0-Cl-NAS. The operation of a Na+/H+ exchanger, already assumed by Moser et al. (1989), could explain both results.

Similar experiments, performed in salines with uncompensated levels of divalent cations, help to explain results obtained previously (Moser, 1985). Cells placed in ‘uncompensated’ 0-Cl-NAS showed a several-fold increase in [Na+]i within 3min, which considerably reduced the transmembrane Na+ gradient. This, together with a minor acid load caused by CO2 /HCO3 -exposure, tends to reduce the activity of the Na+/H+ antiporter and, thus, electrochemical equilibrium conditions with no net flux of these ions could have been achieved.

Effects of SITS on pHi regulation

SITS, a specific blocker of both Na+-dependent and Na+-independent Cl/HCO3 exchange (Cabantchik et al. 1978; Roos and Boron, 1981), was used at a concentration of 1 mmol l-1. When SITS was applied immediately after NH4+-NAS for 6min, the maximum rate of pHi recovery was decreased by 21±9% (N=3) compared with controls without SITS. This result can be explained only by assuming the operation of two separate pHi-regulating mechanisms, both of which depend on extracellular Na+, as shown before. We suggest that these two mechanisms are (1) a Na+-dependent Cl/HCO3 exchange, which is responsible for approximately 20% of total pHi regulation (in nominally HCO3 -free medium; see Discussion), and (2) a Na+/H+ exchange, dominating pHi regulation under such conditions. The experiments below were performed in order to obtain further information on the Na+/H+ exchange mechanism.

Effects of amiloride or EIPA on pHi regulation

Amiloride, a specific blocker of the Na+/H+ antiport (Gaillard and Rodeau, 1978; Grinstein et al. 1989), exerts its blocking action by competing with Na+ (Roos and Boron, 1981). The effects of various amiloride concentrations on pHi, [Na+]i and Em in a single cell are illustrated in Fig. 5. Control and amiloride exposure were alternated but, for clarity, only one control trace is shown.

The dose-dependent effect of amiloride and the amiloride analogue EIPA (Kleyman and Cragoe, 1988) on inhibition of the maximum rate of pHi regulation is shown in Fig. 6. Amiloride had an instant effect, even at low concentrations. In the range 10− 6 to 10− 4 mol l− 1 both drugs showed a similar potency, with a calculated half-maximal inhibition at 3×10− 5 mol l− 1 amiloride. At the same concentrations, there was no significant difference (at the 0.05% level) between the mean values for amiloride and EIPA (Student’s t-test).

Compared with amiloride, EIPA showed pronounced side-effects: within 6min, EIPA (10− 4 mol l− 1) depolarized the membrane potential by about 30mV, an effect never seen even with higher concentrations of amiloride. During washout in NAS, complete recovery of pHi after EIPA exposures took much longer and was accompanied by a slower recovery of [Na+]i.

Compared with controls, amiloride inhibited the maximum rate of pHi regulation by 15.0±3.6% (N=3) at 10− 6 mol l− 1, the lowest concentration used. Maximum inhibition of about 80±5% (N=7) was obtained at an amiloride concentration of 2mmol l− 1. This value did not differ significantly from that obtained at 1mmol l− 1 (0.05%, unpaired Student’s t-test).

We conclude that, in nominally HCO3 -free saline, about 80% of pHi regulation is due to Na+/H+ antiport and about 20% to Na+-dependent Cl /HCO3 antiport, this value being close to that determined in SITS. Summarizing, the mean values of inhibition of the maximum rate of pHi regulation obtained either with SITS or with amiloride (2mmol l− 1) account for complete blockage of pHi regulation.

We tried to demonstrate total inhibition of pHi regulation by blocking both antiporters simultaneously. However, any attempt to combine the blocker of the Na+-dependent Cl /HCO3 antiport (SITS) with the inhibitor of the Na+/H+ antiport (amiloride) resulted in precipitation of amiloride, and it appears that this experiment cannot be performed.

The effects of various concentrations of amiloride on [Na+]i and on maximum pHi regulation were not as consistent as those found for the maximum rate of pHi regulation. We argue that amiloride, and more clearly EIPA, interacted in a concentration-dependent manner not only with the Na+/H+ exchanger but also with Na+/K+-ATPase. Compared with controls, slower rates of [Na+]i recovery were seen during and after amiloride exposures in previously acid-loaded cells (Fig. 5). Similar amiloride interactions have been reported by Grinstein et al. (1989) for a different preparation (for EIPA interactions, see Kleyman and Cragoe, 1988).

Dependence of the operation of the Na+/H+ and the Na+/H+/HCO3/Cl antiporter on extracellular [Na+]

As reported above, pHi regulation was completely inhibited in 0-Na+-NAS. As soon as 5 mmol l-1 Na+ was added (5Na+-NAS) acid extrusion started at a normalized maximum rate of pHi recovery of 24±3% (N=5). The normalized maximum pHi recovery rate in 10Na+-NAS was 31±2% (N=6), in 30Na+-NAS it was 68±4% (N=9), in 50Na+-NAS it was 75±7% (N=6) and in NAS (equivalent to 210mmol l− 1 Na+) it was 100%. A plot of these values (Fig. 7) indicates a hyperbolic saturation curve, allowing application of the Michaelis–Menten equation. The calculated apparent Michaelis Menten constant (Km ) was 21.1±0.7mmol l− 1 (extracellular) Na+.

Steady-state values of pHi, [Na+]i and Em

A comparison of the pHi, [Na+]i and Em values, obtained at physiological rest, with previous data (Moser, 1985) shows minor differences. The 0.15 pH units more alkaline pHi found here seems to be caused by several factors, such as a different mode of calibration of the pH electrodes, which accounts for a difference of 0.06 pH units. Additional effects could arise from altered measurements of Em (K2 SO4 /KCl electrode versus 3 mol l− 1 KCl electrode), from seasonal effects (H. Moser, unpublished observation) and from chemical differences in the water in which the animals were kept for several months. We also noted a slow alkaline shift of pHi with time of experimentation. The deviation in mean pHi values cannot be explained by the effects of temperature on pHi as they are inversely related to the observations (Aickin and Thomas, 1977; Rodeau, 1984).

Such arguments could also serve to explain the 4mmol l− 1 higher [Na+]i value previously observed, but the most likely explanation is that we did not wait long enough until we started the tests. When the cells were penetrated by the double-barrelled electrode we observed a substantial increase in [Na+]i from which the cells slowly recovered. Comparing ‘early’ with ‘late’ controls in a single cell, we often found that [Na+]i had decreased in later controls, a result similar to that of Edman et al. (1983, 1986).

The configuration of the double-barrelled electrode may have caused an extra activation of the stretch-activated channels (Erxleben, 1989) or the electrode seal may have been less than perfect, resulting in a 1mV lower mean value for Em.

Ion transport involved in pHi regulation in the sensory neurone

In addition to NH4+-NAS exposures, we carefully studied pHi changes in 0-Na+-NAS (Figs 1 and 2) and concluded that pHi regulation was Na+-dependent. From further experiments, we concluded that two separate exchange mechanisms existed in the sensory neurone, a Na+-dependent HCO3 /Cl antiporter and a Na+/H+ antiporter.

In addition to the observations of Moser (1985), two observations support the existence of a Na+-dependent HCO3 /Cl antiporter: (1) high concentrations of amiloride (2mmol l− 1; Figs 5 and 6), a specific blocker of the Na+/H+ antiporter, never inhibited pHi regulation completely, suggesting that an additional pHi-regulating mechanism exists (with Na+ also being involved); (2) SITS, which blocks the HCO3 /Cl antiporter, reduced the maximum rate of pHi regulation by about 20%. This reduction in the rate of recovery was approximately equal to the amount needed to cause complete inhibition of pHi regulation in amiloride experiments.

The fact that all salines were nominally HCO3 -free cannot be taken as an argument against the existence of an HCO3 /Cl antiporter, as two sources of HCO3 are always present: the CO2 of the air and metabolically produced CO2. CO2 can easily diffuse through the plasma membrane and will then be converted to HCO3 in a reaction mediated by carbonic anhydrase, known to be located at the outside of plasma membranes even in excitable cells (Chen and Chesler, 1992; Kaila et al. 1990, 1992a).

From the results presented here, there is no clear evidence to indicate whether Na+ acted indirectly as a modifier of HCO3 /Cl antiport or whether it was directly transported by a Na+/H+/HCO3 /Cl exchanger or an equivalent. As found by several authors (Aickin et al. 1982; Deisz and Lux, 1982; Kaila et al. 1992b), ECl is more negative than Em, resulting in an inwardly directed driving force for Cl. If a Na+-modulated 1:1 HCO3 /Cl antiport exists, it should export HCO3, causing an intracellular acid load. However, an inwardly directed driving force for HCO3 is unlikely, since metabolic CO2 production together with both intra-and extracellular carbonic anhydrase (Chen and Chesler, 1992; Kaila et al. 1992a) will keep CO2 at about the same level on both sides of the plasma membrane. Thus, electrochemical considerations favour the presence of Na+/H+/HCO3 /Cl antiport, as already indicated by Moser (1985).

The existence of Na+/H+ antiport can be deduced (1) from the effect of external Na+ on pHi regulation and the direct involvement of Na+, as sensed with Na+-ISM (Figs 1 and 7), (2) from experiments with selective blockers of Na+/H+ antiport, amiloride and EIPA (Figs 5 and 6), which cause a maximum inhibition of pHi regulation of about 80% at a concentration of 1 or 2 mmoll− 1, and (3) from the observation that pHi was well regulated even when Na+/H+/HCO3 /Cl antiport was blocked by SITS or by prolonged exposure to 0-Cl-NAS in nominally HCO3-free saline (Fig. 3). During such 0-Cl -NAS exposures, the cells should have lost most of their internal Cl (Deisz and Lux, 1982; Voipio et al. 1991), thus retarding the operation of the HCO3 /Cl exchanger or even stopping it. In 0-Cl -NAS, we observed considerable changes in [Na+]i (Fig. 4), suggesting direct involvement of Na+ in an additional pHi-regulating mechanism, Na+/H+ antiport.

At physiological pH, the Na+/H+ antiport is half-maximally inhibited at an amiloride concentration of 3×10− 5 mol l− 1. In addition, previous results (Moser et al. 1989) indicated that it was the combined Na+/H+ gradient that fuelled Na+/H+ antiport (Kinsella and Aronson, 1980; Grinstein and Rothstein, 1986) and that the Na+/H+ antiport was largely, or totally, unaffected by the transmembrane potential.

At an extracellular pH of 7.4, the normalized maximum rate of pHi recovery depends on [Na+]o following a simple Michaelis–Menten relationship, suggesting a single binding site (Fig. 7). We assume that the Km value is mainly linked to the Na+/H+ antiport, which carries out about 80% of pHi regulation in our experiments. We found an apparent Michaelis–Menten constant (Km ) of 21mmol l-1 [Na+]o, indicating that the operation of the Na+/H+ antiporter is unlikely to be limited by physiological [Na+]o changes, since [Na+]o normally exceeds this value by approximately 10-fold. Even when [Na+]o is lowered to about half of its normal value, the action of the Na+/H+ antiporter is hardly impaired (Fig. 7) and the pHi value will only change a little in the acid direction.

The mechanisms of pHi regulation in crayfish preparations

All three preparations of crayfish, ganglion cell (Moody, 1981), muscle cell (Galler and Moser, 1986) and the sensory neurone of the stretch receptor, have two separate pHi-regulating mechanisms in common, a Na+-dependent HCO3 /Cl antiport and a Na+/H+ antiport. Interestingly, the contribution of each mechanism to pHi regulation seems to be different. Muscle fibres had the slowest rate of pHi recovery, the Na+-dependent HCO3 /Cl antiport contributing 75% and the Na+/H+ antiport 25% to pHi regulation (Galler and Moser, 1986). The sensory cell showed the fastest pHi recovery, mainly due to Na+/H+ antiport (by about 80%) and, to a lesser extent (20%), to Na+-dependent HCO3 /Cl antiport. Between these two extremes, the crayfish neurone (Moody, 1981) exhibited an intermediate rate of pHi recovery, with both mechanisms contributing substantially to pHi regulation under normal conditions. All these observations are in line with the observation that amiloride hardly affects pHi regulation in muscle fibres (Galler, 1985) but has a considerable influence on the sensory cell. The opposite was true for exposure to SITS, 0-Cl -saline or nominally HCO3 -free saline, which moderately inhibited pHi recovery in the sensory neurone but had a considerable effect in the muscle fibres.

Moulting and pHi regulation

During the moulting cycle of crustaceans, cuticular CaCO3 mineralization and demineralization is accompanied by changes in [Ca2+], [HCO3] and [H+] (e.g. Cameron and Wood, 1985; Greenaway, 1985; Roer and Dillaman, 1984; Wheatly and Ignaszewski, 1990). The increase in body size in freshwater crayfish is accomplished by uptake of environmental water. Both processes affect the concentration of ions involved in pHi regulation. The effects of concentration changes on the mechanisms of pHi regulation are discussed below.

The dependence of the normalized maximum pHi recovery rate on [Na+]o (Fig. 7) gives Km values of 21mmol l− 1 for the crayfish sensory cell and 39mmol l− 1 for the neurone (Gaillard and Rodeau, 1987). Although direct comparisons cannot be drawn, because the experimental situations were somewhat different, we assume that pHi remains regulated to a reasonable extent (and may be fully regulated) during the moulting cycle, thus keeping the nervous system functioning, mainly by the operation of the Na+/H+ antiport.

In crayfish muscle, the functional role of the Na+/H+ antiport, contributing approximately 25% to pHi regulation (Galler and Moser, 1986), is less clear. Assuming a Km value in the range 3–50mmol l− 1 [Na+]o, as found in many preparations from vertebrates and invertebrates (Aronson, 1985; Moolenaar, 1986), the Na+/H+ antiport could provide some kind of ‘safety mechanism’ for pHi regulation with variable importance during the moulting cycle. At moulting, ionic conditions could have slowed or stopped the operation of the Na+/H+/HCO3/Cl antiport, but the Na+/H+ antiport, hardly affected by ionic alterations, would remain functional.

We wish to thank Dr Juha Voipio for critically reading the manuscript and Mag Gabriele Buemberger for linguistic improvements. The work was partly sponsored by the FWF (project 6177) and the University of Innsbruck.

Aickin
,
C. C.
,
Deisz
,
R. A.
and
Lux
,
H. D.
(
1982
).
Ammonium action on post-synaptic inhibition in crayfish neurones: implications for the mechanism of chloride extrusion
.
J. Physiol., Lond.
329
,
319
339
.
Aickin
,
C. C.
and
Thomas
,
R. C.
(
1977
).
Micro-electrode measurement of the intracellular pH and buffering power of mouse soleus muscle fibres
.
J. Physiol., Lond.
267
,
791
810
.
Ammann
,
D.
(
1986
).
Ion-Selective Micro-Electrodes. Berlin, Heidelberg: Springer-Verlag
.
Aronson
,
P. S.
(
1985
).
Kinetic properties of the plasma membrane Na+–H+ exchanger
.
A. Rev. Physiol.
47
,
545
560
.
Boron
,
W. F.
and
De Weer
,
P.
(
1976
).
Intracellular pH transients in squid giant axons caused by CO2, NH3 and metabolic inhibitors
.
J. gen. Physiol.
67
,
91
112
.
Brown
,
H. M.
,
Ottoson
,
D.
and
Rydqvist
,
B.
(
1978
).
Crayfish stretch receptor: an investigation with voltage-clamp and ion-sensitive electrodes
.
J. Physiol., Lond.
284
,
155
179
.
Cabantchik
,
I. Z.
,
Knauf
,
P. A.
and
Rothstein
,
A.
(
1978
).
The anion transport system of the red blood cell. The role of membrane protein evaluated by the use of ‘probes’
.
Biochim. biophys. Acta
515
,
239
302
.
Cameron
,
J. N.
and
Wood
,
C. M.
(
1985
).
Apparent H+ excretion and CO2 dynamics accompanying carapace mineralization in the blue crab (Callinectes sapidus) following moulting
.
J. exp. Biol.
114
,
181
196
.
Chen
,
J. C. T.
and
Chesler
,
M.
(
1992
).
Modulation of extracellular pH by glutamate and GABA in rat hippocampal slices
.
J. Neurophysiol
.
67
,
29
36
.
Chesler
,
M.
(
1990
).
The regulation and modulation of pH in the nervous system
.
Prog. Neurobiol.
34
,
401
427
.
Deisz
,
R. A.
and
Lux
,
H. D.
(
1982
).
The role of intracellular chloride in hyperpolarizing post-synaptic inhibition of crayfish stretch receptor neurones
.
J. Physiol., Lond.
326
,
123
138
.
Edman
,
A.
,
Gestrelius
,
S.
and
Grampp
,
W.
(
1983
).
Intracellular ion control in lobster stretch receptor neurone
.
Acta physiol. scand
.
118
,
241
252
.
Edman
,
A.
,
Gestrelius
,
S.
and
Grampp
,
W.
(
1986
).
Transmembrane ion balance in slowly and rapidly adapting lobster stretch receptor neurones
.
J. Physiol., Lond.
377
,
171
191
.
Erxleben
,
C.
(
1989
).
Stretch-activated current through single ion channels in the abdominal stretch receptor organ of the crayfish
.
J. gen. Physiol.
94
,
1071
1083
.
Fresser
,
F.
,
Moser
,
H.
and
Mair
,
N.
(
1991
).
Intra- and extracellular use and evaluation of ammonium-selective microelectrodes
.
J. exp. Biol.
157
,
227
241
.
Gaillard
,
S.
and
Rodeau
,
J. L.
(
1987
).
Na+/H+ exchange in crayfish neurons: dependence on extracellular sodium and pH
.
J. comp. Physiol. B
157
,
435
444
.
Galler
,
S.
(
1985
).
Intracelluläre pH-Regulation in Muskelfasern des Flusskrebses (Astacus fluviatilis). Diplomarbeit am Institut für Zoologie der naturwissenschaftlichen Fakultät der Universität Innsbruck, Austria
.
Galler
,
S.
and
Moser
,
H.
(
1986
).
The ionic mechanism of intracellular pH regulation in crayfish muscle fibres
.
J. Physiol., Lond.
374
,
137
151
.
Grafe
,
P.
,
Ballanyi
,
K.
and
Ten Bruggencate
,
G.
(
1985
).
Changes of intracellular free ion concentrations, evoked by carbachol or GABA, in rat sympathetic neurons
. In
Ion Measurements in Physiology and Medicine
(ed.
M.
Kessler
,
E. K.
Harrison
and J. Höper
), pp.
184
188
. Berlin, Heidelberg: Springer Verlag.
Greenaway
,
P.
(
1985
).
Calcium balance and moulting in Crustacea
.
Biol. Rev.
60
,
425
454
.
Grinstein
,
S.
and
Rothstein
,
A.
(
1986
).
Mechanisms of regulation of the Na+/H+ exchanger
.
J. Membr. Biol
.
90
,
1
12
.
Grinstein
,
S.
,
Rotin
,
D.
and
Mason
,
M. J.
(
1989
).
Na+/H+ exchange and growth factor-induced cytosolic pH changes. Role in cellular proliferation
.
Biochim. biophys. Acta
988
,
73
97
.
Kaila
,
K.
,
Paalasmaa
,
P.
,
Taira
,
T.
and
Voipio
,
J.
(
1992a
).
pH transients due to monosynaptic activiation of GABAA receptors in rat hippocampal slices
.
NeuroReport
3
,
105
108
.
Kaila
,
K.
,
Rydqvist
,
B.
,
Pasternak
,
M.
and
Voipio
,
J.
(
1992b
).
Inward current caused by sodium-dependent uptake of GABA in the crayfish stretch receptor neurone
.
J. Physiol., Lond
.
453
,
627
645
.
Kaila
,
K.
,
Saarikoski
,
J.
and
Voipio
,
J.
(
1990
).
Mechanism of action of GABA on intracellular pH and on surface pH in crayfish muscle-fibers
.
J. Physiol., Lond
.
427
,
241
260
.
Kinsella
,
J. L.
and
Aronson
,
P. S.
(
1980
).
Properties of the Na+–H+ exchanger in renal microvillus membrane vesicles
.
Am. J. Physiol.
238
,
F461
F469
.
Kleyman
,
T. R.
and
Cragoe
,
E. J.
(
1988
).
Amiloride and its analogs as tools in the study of ion-transport
.
J. Membr. Biol.
105
,
1
22
.
Mair
,
N.
(
1992
).
pHi-Regulation beim sensorischen Neuron des Muskelrezeptororgans (MRO) von Astacus astacus. Diplomarbeit am Institut für Zoologie der naturwissenschaftlichen Fakultät der Universität Innsbruck, Austria
.
Moody
,
W. J.
(
1981
).
The ionic mechanism of intracellular pH regulation in crayfish neurones
.
J. Physiol., Lond.
316
,
293
308
.
Moolenaar
,
W. H.
(
1986
).
Regulation of cytoplasmic pH by Na+/H+ exchange
.
Trends physiol. Sci
.
11
,
141
143
.
Moser
,
H.
(
1985
).
Intracellular pH regulation in the sensory neurone of the stretch receptor of the crayfish (Astacus fluviatilis)
.
J. Physiol., Lond.
362
,
23
38
.
Moser
,
H.
,
Mair
,
N.
,
Fresser
,
F.
and
Rydqvist
,
B.
(
1989
).
Effects of ions and drugs on ionic regulation in a sensory neuron of crayfish
.
Acta physiol. scand.
136
, Suppl.
582
,
77
.
Rodeau
,
J. L.
(
1984
).
Effect of temperature on intracellular pH in crayfish neurons and muscle fibers
.
Am. J. Physiol.
26
,
C45
C49
.
Roer
,
R. D.
and
Dillaman
,
R.
(
1984
).
The structure and calcification of the crustacean cuticle
.
Am. Zool
.
24
,
893
909
.
Roos
,
A.
and
Boron
,
W. F.
(
1981
).
Intracellular pH
.
Physiol. Rev
.
61
,
296
434
.
Schwiening
,
C. F.
and
Thomas
,
R. C.
(
1992
).
Mechanism of pHi regulation by locust neurones in isolated ganglia: a microelectrode study
.
J. Physiol., Lond
.
447
,
693
709
.
Thomas
,
R. C.
(
1984
).
Experimental displacement of intracellular pH and the mechanism of its subsequent recovery
.
J. Physiol., Lond.
354
,
3P
22P
.
Van Harreveld
,
A.
(
1936
).
A physiological solution for freshwater crustaceans
.
Proc. Soc. exp. Biol. Med.
34
,
428
432
.
Voipio
,
J.
,
Pasternack
,
M.
,
Rydqvist
,
B.
and
Kaila
,
K.
(
1991
).
Effect of gamma-aminobutyric acid on intracellular pH in the crayfish stretch-receptor neurone
.
J. exp. Biol.
156
,
349
361
.
Wheatly
,
M. G.
and
Ignaszewski
,
L. A.
(
1990
).
Electrolyte and gas exchange during the moulting cycle of a freshwater crayfish
.
J. exp. Biol.
151
,
469
483
.