Turgor regulation in the salt-tolerant alga Chara longifolia

Authors

  • N. A. Stento,

    1. Department of Biological Sciences, State University of New York at Buffalo, Cooke Hall 109, Buffalo NY 14260–1300, USA
    Search for more papers by this author
    • *Present address: 316 Maple Street, Vestal, NY13850, USA.

  • N. Gerber Ryba,

    1. Department of Biological Sciences, State University of New York at Buffalo, Cooke Hall 109, Buffalo NY 14260–1300, USA
    Search for more papers by this author
  • E. A. Kiegle,

    1. Department of Biological Sciences, State University of New York at Buffalo, Cooke Hall 109, Buffalo NY 14260–1300, USA
    Search for more papers by this author
    • Present address: Ceres, Inc., 3007 Malibu Canyon Road, Malibu, CA 90265, USA.

  • M. A. Bisson

    1. Department of Biological Sciences, State University of New York at Buffalo, Cooke Hall 109, Buffalo NY 14260–1300, USA
    Search for more papers by this author

Mary A. Bisson. E-mail: bisson@acsu.buffalo.edu

ABSTRACT

Chara longifolia is a salt-tolerant Charophyte which regulates its turgor inresponse to osmotic stress. Membrane depolarization, in creased membrane conductance, and cessation of cytoplasmic streaming (due to increase in cytoplasmic Ca2+) precede regulation in response to hypotonic stress. Measurements of these three parameters are presented here with simultaneous turgor measurements. Variability in the occurrence, rate and extent of turgor regulation in individual cells was correlated with magnitude of the stress. Hypertonic stress showed the same slow time course as was found previously, requiring several days for complete regulation. Fifty μM nifedipine, a Ca2+ channel blocker, inhibited turgor regulation. In the presence of 5 μM nifedipine, turgor regulation was delayed. An increase in conductance preceded regulation, but membrane depolarization was less and no detectable change in cytoplasmic streaming was observed, requiring modifications to a previously presented model for turgor regulation. There was no significant difference in 45Ca2+ influx under control and stress conditions. However, the control flux was insensitive to nifedipine, whereas under stress the flux is inhibited 54% by nifedipine. We suggest that osmotic stress results in a rapid increase in a nifedipine-sensitive Ca2+ entry mechanism, followed very quickly by a decrease in the control entry mechanism.

INTRODUCTION

The Characean algae have long been used as experimental model systems for a number of plant functions, including membrane transport. Most members of this group are freshwater, and do not regulate their turgor under normal circumstances ( Gutknecht, Hastings & Bisson 1978; Sanders 1981; Bisson & Bartholomew 1984). However, algae growing in saline habitats usually regulate their turgor ( Gutknecht et al. 1978 ; Bisson & Kirst 1995), and salt-tolerant Charophytes are no exception. Chara longifolia [earlier referred to as Chara buckellii ( Hoffmann & Bisson 1986) following Wood & Imahori (1965) but renamed following Proctor (1980)] is capable of growing in both saline and freshwater media ( Hoffmann & Bisson 1986, 1990). It regulates its turgor by increasing or decreasing internal osmotic pressure, mainly KCl in the vacuole, in response to alterations in external osmotic pressure ( Hoffmann & Bisson 1986, 1990). Correlation of the time course of turgor regulation with changes in membrane potential and conductance enabled us to generate a model for the mechanism of turgor regulation ( Bisson, Kiegle & Kiyosawa 1992; Bisson et al. 1995 ). In this model, which is similar to one generated for another salt-tolerant Charophyte, Lamprothamnium ( Okazaki & Tazawa 1990), we suggested that stress-induced depolarization of the membrane results in opening of K+ and Ca2+ channels. As long as the membrane is more positive than the Nernst potential for K+ (EK), K+ can exit the cell passively, but its ability to do so is limited by the movement of a counterion to balance charge. Opening Ca2+ channels allows Ca2+ to enter the cytoplasm from the external medium which, together with possible calcium-induced Ca2+ release from internal stores, results in an increase in cytoplasmic Ca2+. The opening of Ca2+-sensitive Cl channels permits Cl to leave the cell, resulting in a net loss of KCl and hence a regulatory turgor loss.

To test this hypothesis, we need to manipulate some of the physiological responses, and look at the consequences for turgor regulation. In our previous experiments, however, electrophysiological parameters and turgor were measured in separate cells ( Hoffmann & Bisson 1990; Bisson et al. 1992 , 1995). When calcium channel blockers were added, results were variable, both in electrophysiological response and in turgor regulatory response, making it difficult to correlate the two. We therefore needed a way to monitor turgor and electrophysiological parameters simultaneously over long periods of time. We utilized the pressure probe technique ( Steudle & Zimmermann 1974) together with standard electrophysiological techniques to do this. We show here individual variation in turgor and turgor regulation, as well as the dependence of turgor regulation on the magnitude of the turgor stress. We also suggest the existence of two parallel mechanisms of turgor regulation. In addition to the one described above, which is dependent on membrane depolarization and increased cytoplasmic Ca2+, a separate mechanism can regulate turgor without the transient Ca2+ influx and with the cell remaining more negative than EK.

MATERIALS AND METHODS

Algae

Chara longifolia is a salt-tolerant alga collected from the saline Waldsea Lake in Saskatchewan, Canada ( Hoffmann & Bisson 1986). Culture conditions were as described previously ( Hoffmann & Bisson 1986). Chara longifolia was cultured in garden soil, using a medium designed to mimic the water of Waldsea Lake (Waldsea water) ( Hoffmann & Bisson 1986), although at a somewhat reduced strength [osmotic pressure 0·54 MPa (225 mOsmol kg−1) instead of 0·9 MPa (375 mOsmol kg−1)]. Cells acclimated at reduced osmotic strength were used because cells cultured at higher osmotic strength accumulated vacuolar inclusions which were able to clog the probe. The major ions in this solution are at the following concentrations: K+, 4·6 m M; SO42−, 74·5 m M; Mg2+, 62 m M; Cl, 62 m M; Ca2+, 4·5 m M; Na+, 73·5 m M. Experiments were performed in solutions with the same major ion composition, buffered to pH 8·5 with 5 m M 2-[N-cyclohexylamino]ethane-sulfonic acid (CHES).

Turgor probe

The turgor probe and amplifier were initially obtained from Professor E. Steudle (University of Bayreuth, Germany) and the probe was subsequently modified to accommodate our cells. The internal volume of the apparatus and micropipette was approximately 35·5 μL, whereas the average volume of C. longifolia was 3·5 μL and of C. corallina was 10 μL. We used Dow Corning 200® silicone oil which had a viscosity of 1·5 centistoke and a coefficient of compressibility (κ) of 10−3 MPa−1. This fulfilled under all conditions the requirement that Vapp/Vcell < < (κoil×ɛ)−1 ( Steudle 1993). The pressure transducer which we used was CD-140-200D from Kulite (Leonia, NJ, USA). Data were accumulated on a chart recorder or on a Data Electronics data-logger (DT50) (Data Electronics U.S.A. Inc., Irvine, CA, USA).

For simultaneous measurement of turgor pressure and electrophysiological parameters, we used a chain of two adjacent internodes, separated by the nodal cells. These internodes were isolated from an intact plant, trimmed of branchlets, and incubated in experimental medium for at least 24 h prior to impaling. One cell was impaled with the pressure probe axially through the node and was monitored for cytoplasmic streaming. The second cell was used for electrophysiological studies using two intracellular microelectrodes ( Fig. 1). Membrane conductance was measured by injecting a small hyperpolarizing current and monitoring the resultant change in membrane potential. In the short Chara longifolia cells used, cable corrections were not necessary ( Bisson et al. 1995 ).

Figure 1.

Apparatus for simultaneous measurement of turgor pressure and electrophysiological properties in two adjacent internodes.

To deliver a hypotonic shock, the medium was exchanged for another that was diluted to the desired osmotic pressure with 5 m M CHES/NaOH, pH 8·5. Percentage regulation was calculated by measuring the initial increase in turgor and the subsequent decrease in turgor from the maximum. Percentage regulation was calculated as (decrease from the maximum increase) × 100. When osmotic shock was delivered with nifedipine, cells were first incubated at the desired concentration of nifedipine in the control medium for 15 min before changing to the hypotonic solution.

45Calcium fluxes

45Calcium fluxes were measured as reported previously ( Bisson et al. 1995 ). Briefly, isolated internodes were exposed to a solution containing 75–100 μCi of 45Ca for 3, 20, 40 or 60 s. Four cells were harvested at each time point and rinsed in an ice-cold wash solution containing La3+ to block efflux. Internodal cell contents were separated from the cell wall by removing the nodes, inserting a syringe and flushing with water. Flux was calculated from the slope of the curve of cpm m−2 versus time over the linear extent of the graph. Most experiments were linear for the complete 60 s, but in some cases the flux began to decrease at 60 s, and in these cases only the first three points were used. For nifedipine experiments, the cells were exposed to nifedipine for 15 min prior to the exposure to isotope. For stress experiments, cells were exposed to the hypotonic solution for 5 s prior to the exposure to isotope.

RESULTS

Correlation of responses of adjacent cells

Ideally, we would have liked to measure turgor regulation and electrophysiological parameters in the same cell. This was not possible, however, because regulation was relatively slow (0·5–2 h), and cells impaled with three probes did not often survive for that length of time. We chose instead to use a chain of two adjacent internodes. Charophyte cells are well connected by plasmodesmata which facilitate large ion movements between the cells ( Spanswick & Costerton 1967; Bostrom & Walker 1975; Cook et al. 1997 ), which would bring the membrane potentials of the two cells towards the same value. We tested whether the two internodes did have the same potential by simultaneously impaling two adjacent internodes with microelectrodes. In all seven cases, as long as the impalement was good (i.e. the electrode was not excluded from the cytoplasm, as indicated by a slow depolarization which could be reversed by further advancement of the microelectrode [see Fig. 2 arrow]) and cells remained healthy (as indicated by steady negative membrane potential and rapid cytoplasmic streaming) the difference in membrane potential between the two cells was always less that 20 mV; at steady state, the difference was less than 10 mV. This was true whether or not one cell was impaled with the pressure probe. In particular, in response to osmotic stress, if one cell depolarized, the other did simultaneously ( Fig. 2a). In the three experiments in which hypotonic shock was administered, two cell pairs showed the depolarization characteristic of turgor regulation and one did not (compare Fig. 2a, b). When the transient cessation of streaming occurred, it occurred in both cells simultaneously. The membrane parameters of the adjacent cell are thus a good indicators of those of the cell impaled with the turgor probe.

Figure 2.

Comparison of membrane potentials in two adjacent cells. In both cases, the trace from one cell (cell a) is shown with a solid line, and the other (cell b) with a dotted line. Where only the solid line is shown, the two lines are superimposed. (a) Cell showing a strong depolarization on osmotic stress. (b) Cell which did not strongly depolarization on osmotic stress.

Hypotonic stress: turgor regulation

Control turgor ranged from 0·4 to 0·65 MPa, and was a function of size, with smaller cells having a lower turgor ( Fig. 3). In response to a 0·24 MPa (100 mOsmol kg−1) hypotonic stress, not all cells regulated their turgor within 2 h ( Fig. 4). Cells which regulated appeared to have a higher initial turgor, but difference was not significant (P = 0·0725) ( Table 1). The clearest electrophysiological difference between the two sets of cells was that those which regulated showed a rapid depolarization, where those which did not showed only a slow depolarization and remained negative to EK [–85 mV, assuming a cytoplasmic K+ concentration of 140 m M, less than cells cultured in 375 mOsmol kg−1 Waldsea water, but slightly more than those cultured in 375 mOsmol kg−1 artificial seawater, which has a higher Na+ and lower Mg2+ concentration ( Hoffmann & Bisson 1986)]. The time course of regulation of a typical regulating cell is shown in Fig. 5. Cells which regulated always showed a decrease in streaming rate. The 5 min average ( Table 1) shows only a slight decrease, because of differences in time of decrease in streaming. In a separate experiment, we showed that in every osmotically stressed cell in which streaming slowed, it stopped completely at some point between 2 and 5 min.

Figure 3.

Turgor pressure as a function of size in C. longifolia.

Figure 4.

Percentage turgor recovery in 11 C. longifolia cells subjected to a 0·24-MPa hypotonic shock.

Table 1.  Comparison of turgor pressure, membrane electrophysiological characteristics, and streaming rate in cells with differing abilities to accomplish turgor regulation within 2 h following a hypo-osmotic stress of 100 mOsmol/kg. Values are means ± SD
 Initial5 min120 min
Regulating cells
Turgor pressure (MPa)0·57 ± 0·060·77 ± 0·030·66 ± 0·05
% recovery  52%± 16%
n = 6   
Membrane potential (mV)− 161 ± 13− 107 ± 14− 127 ± 19
n = 6   
Conductance (S m−2) 1·99 ± 1·23·36 ± 1·62·59 ± 1·2
n = 5   
Streaming rate (μm s−1) 66 ± 4·151 ± 1065 ± 1·5
n = 5   
Non-regulating cells
Turgor pressure (MPa)0·49 ± 0·050·72 ± 0·060·71 ± 0·06
% recovery  2%± 2%
n = 5   
Membrane potential (mV)− 157 ± 9·9− 148 ± 22− 122 ± 6·5
n = 5   
Conductance (S m−2) 2·31 ± 1·23·78 ± 2·33·11 ± 0·7
n = 4   
Streaming rate (μm s−1) 69 ± 2·264 ± 6·269 ± 4·2
n = 4   
Figure 5.

Time course of turgor pressure (a), membrane potential (b) and membrane conductance (c) following hypotonic stress (delivered at time = 0) in a typical cell pair. Streaming showed a decrease between 2 and 5 min following the hypotonic stress. A cessation was not observed, but likely occurred somewhere in this time (see text).

Regulation was slower and less complete than reported earlier ( Hoffmann & Bisson 1990; Bisson et al. 1995 ). To test whether this was due to interference with regulation due to the impalement, we performed parallel experiments using cells which were not continuously impaled, but whose turgor was sampled by impalement with the pressure probe at different times following a hypotonic shock. Streaming was monitored in all cells immediately after the shock. About one-third of the cells failed to show a decrease in streaming rate; these were considered to be non-regulators and their turgor was not monitored. This is similar to the degree of failure to regulate in impaled cells. Those cells which did show a decrease in streaming rate in the first 6 min after the shock were impaled with the turgor probe at intervals following the shock. After the impalement, cells were discarded and not further monitored. These cells showed a similar time course of regulation to that of impaled cells ( Fig. 6). Possible reasons for the difference from previous results are given in the Discussion.

Figure 6.

Time course of turgor regulation in cells not continuously impaled. Average (± SE) of turgor in cells subjected to hypotonic stress, impaled only at the time of measurement. n = 9 for each time point.

Previously we had shown that turgor regulation failed in solutions containing low (10 μM) Ca2+. We had earlier attempted to look at the effects of the calcium channel blocker verapamil, but both the turgor response and the electrophysiological responses were variable. We could not correlate the effects on turgor regulation and membrane potential or conductance because these turgor and electrophysiological parameters were measured in different cells in those experiments. Using the two-cell chain, we found that 50 μM nifedipine consistently blocked turgor regulation. In these cells, the depolarization (from − 149 ± 16 mV to − 139 ± 16 mV) was not significant at the 5% level (P = 0·078), and no increase in conductance or decrease in streaming was measured. The effect was irreversible, since removing the nifedipine from the solution two hours after the hypotonic stress did not result in regulation in the subsequent hour, the longest that typical cells could maintain a negative potential and vigorous streaming under these conditions. A concentration of 25 μM nifedipine also completely inhibited regulation. At a lower concentration of nifedipine, 5 μM, cells were able to regulate, but later ( Fig. 7a). Under control conditions, cells which regulated showed an initial decrease in turgor after between 5 and 15 min. Of the six cells studied with 5 μM nifedipine, one showed no regulation at all by 120 min, two showed initial regulation by 10 min, and the remaining three began to regulate between 30 and 60 min. In no case was a decrease in cytoplasmic streaming measured. For those cells which did regulate their turgor, the membrane depolarized ( Fig. 7b), but more slowly and to a lower extent, remaining, on average, negative to EK (−85 mV). Depolarization preceded regulation in every case although in one case, depolarization was very small, and membrane potential remained negative to −150 mV for 90 min following the shock. The increase in conductance also occurred later than in control cells ( Fig. 7c). In four of the five cells which regulated, this increase occurred between 5 and 55 min before the initiation of turgor regulation. In the fifth case, the conductance increased and turgor decreased between 30 and 45 min; because conductance measurements were made infrequently in this time period, we cannot say whether or not the conductance increase preceded turgor regulation. In the cell which did not regulate, conductance showed an initial sharp decrease, along with a hyperpolarization ( Fig. 7). This is the only cell which showed this particular response. Later (60 min) an increase in conductance was seen, along with membrane depolarization, but since cytoplasmic streaming was slowing at this time, from 63 to 54 μm s−1, this was probably due to deterioration in cell condition.

Figure 7.

Time course of turgor regulation (a) and changes in membrane potential (b) and conductance (c) following hypotonic stress in the presence of 5 μM nifedipine. In all cases, the open circles with the errors bars represents the mean ± standard error of all the cells which regulated (n = 5). On the same graph is shown the data (squares) from an individual cell, s235, which did not regulate its turgor, and was not included in the averages.

Hypotonic stress: 45Ca2+ fluxes

Calcium influxes are shown in Table 2. Control Ca2+ influx is not sensitive to nifedipine. After a brief shock (5 s pre-exposure plus 3–60 s exposure to labelled solution) the flux was not significantly different, and longer exposure to stress (30 s pre-exposure) also did not reveal an increased flux. However, during osmotic shock the flux was now sensitive to nifedipine (53% inhibition, P = 0·043). The only explanation for this is that a nifedipine-sensitive flux is induced by osmotic stress, whereas the nifedipine-insensitive flux is depressed. The possible physiological significance is discussed below.

Table 2.  Influx of 45Ca2+, nmol m−2 s−1. Values are mean ± SD (n)
 Control+ 50 μM nifedipine
  1. aP = 0·043.

Control (no stress)17·2 ± 5·4 (3)17·1 ± 2·4 (3)
Hypotonic stress (5 s)17·3 ± 3·8 (5)8·02 ± 1·4 (4) a

Hypertonic stress: turgor regulation

Regulation in response to hypertonic stress showed little or no sign of regulation over the time course during which cells could be continuously impaled (up to 2 h). We therefore subjected the cells which had not been impaled to 0·36 MPa hypertonic stress, and impaled them at intervals afterwards. The slow development of reduced turgor is an artefact of measuring different cells, and was not generally seen in continuous impalements. There was a relatively rapid regulation at 2–8 h, and slower regulation thereafter ( Fig. 8). There was no obvious interruption in streaming during this time except that associated with handling, presumably due to action potentials.

Figure 8.

Time course of turgor regulation in response to 0·36 MPa hypertonic stress.

DISCUSSION

Previous studies on turgor regulation in Charophytes

Turgor regulation in response to a hypotonic stress has been studied in C. longifolia previously ( Bisson et al. 1992 , 1995), and is similar to that found for the Charophyte Lamprothamnium ( Bisson & Kirst 1980a, 1980b; Okazaki & Tazawa 1990). In those studies as well as this one, regulation under normal conditions appeared to require three components: a strong depolarization to a value positive to the Nernst potential for potassium, EK; a transient cessation in streaming, which has been correlated with an increase in cytosolic Ca2+ ( Okazaki & Tazawa 1990); and a transient increase in conductance. At least part of the increase in cytosolic Ca2+ is due to influx across the plasma membrane, since reducing external Ca2+ prevents it ( Okazaki & Tazawa 1990; Bisson et al. 1995 ). To explain these results, we suggested the model shown in Fig. 9 ( Bisson et al. 1995 ; Bisson & Kirst 1995). In this model, membrane depolarization initiates opening of K+ and Ca2+ channels, and the subsequent increase in cytosolic Ca2+ opens Cl channels. As long as the membrane potential is positive to EK but negative to ECl, there will be a driving force for exit of both K+ and Cl, resulting in a net electroneutral loss of KCl and decrease internal osmotic pressure, and hence in turgor.

Figure 9.

Model for turgor regulation. See text for details.

This feedback mechanism has been criticized by Beilby & Shepherd (1996). They used inhibitors of K+ and Ca2+ channels to separate the increases in conductance due to K+ influx and the Ca2+-dependent Cl influx. They suggested that the increased conductance to K+ and Cl occurred at different times. To enable the large net fluxes of K+ and Cl which are required to alter turgor, there must therefore be other counterions moving at these different times. They also showed that these increases remained transient even though turgor regulation was impaired under these conditions, suggesting that feedback was not required to control the fluxes. One problem with these experiments is that there is no assurance that the fluxes of one ion remain the same when the other is inhibited. Instead, it seems likely that there are mutual influences on these ions, especially as the response of the membrane potential is altered. Another problem is that turgor was not measured in the same cells, and one cannot say with confidence what the turgor actually was, including when and whether it was regulated. Nevertheless, an alternative hypothesis is that in response to an increased turgor, there is an early transient loss of Cl (balanced by an unknown counterion) and a later transient loss of K+ (balanced by another unknown counterion) which result in an approximate restoration of turgor.

Discrepancies in the time course of turgor regulation

In the earliest experiments, turgor was estimated by measuring internal osmotic pressure and subtracting from it the external osmotic pressure ( Hoffmann & Bisson 1990). This is a destructive technique, so that batch comparisons had to be made between different individual cells. It also required a rather large (10 μL) sap sample, so that large cells were utilized. In these experiments, regulation required several days for completion. A second set of experiments utilized the non-destructive turgor balance technique, which allowed us to measure the time course of the response in a single cell ( Bisson et al. 1995 ; cf. Okazaki & Tazawa 1986). This technique relies on differences in compressibility at different turgors, and requires a slight compression in the cell. In these experiments, all control cells regulated turgor. There was an immediate rapid regulation after hypotonic stress, and regulation was 90% complete in 30–60 min. These experiments were done on smaller cells than the first experiments, and we suggest that size of the cells could affect the time course of turgor regulation.

In the experiments reported here, with cells continuously impaled with the pressure probe, 30–40% of the cells did not regulate their turgor. Beilby, Cherry & Shepherd (1999) showed that Lamprothamnium cells varied in their ability to regulate turgor, and this was correlated with the developmental stage of the cells. In our experiments, however, we were using a uniform size of internodal cells (1·5–2·5 cm long), although there is not a simple correlation between size and developmental status of the internodes, as they may vary in size due to light intensity, degree of prior harvesting, etc. However, no significant correlation between size and rate of regulation was noticed. When cells did regulate, regulation did not start immediately but gradually developed, beginning to decrease in about 5 min, and took longer, with about 50% regulation after 120 min

Size may have some influence on the rate of regulation. The largest cells utilized in the earliest studies were the slowest to regulate. But similarly sized cells were used with the turgor balance and the pressure probe, and these showed consistently different rates of regulation. Impalement with the probe may have damaged the cell and interfered with the regulatory mechanism. To test for this, we did a batch experiment, in which cells which had not been impaled were shocked, then the cell was impaled by the pressure probe at various times after the shock to assess the progress of regulation. In this case we saw a time course of regulation similar to that seen with the impaled cell ( Fig. 6). This indicated that the impalement was not affecting the time course.

The differences in the time course of regulation are instead most likely due to the differences in the size of the osmotic stress. There are two differences between these hypotonic stress experiments and those performed earlier. The first is that we used cells acclimated to a less saline medium, 0·54 MPa instead of 0·9 MPa (225 mOsmol kg−1 instead of 375 mOsmol kg−1). We did this because cells acclimated to more saline medium had more inclusions in their vacuole, which clogged the turgor probe and made long-term measurements difficult. The second difference is that the osmotic stress was less, 0·24 MPa instead of 0·36 MPa (100 mOsmol kg−1 instead of 150 mOsmol kg−1). This was because the higher turgor obtained with the larger stresses were more likely to break the seal between the probe and the membrane. These two methodological differences can explain the differences in regulation described above, if there is a minimum turgor which initiates and terminates the regulatory response. Thus, if the threshold turgor for initiation of regulation is, for instance, 0·70 MPa, cells with an initial turgor of 0·45 MPa subject to an osmotic stress of 0·24 MPa will not achieve the minimum turgor required to initiate the regulatory response. In addition, if the threshold turgor for termination of the regulatory turgor decrease response is 0·65 MPa, a cell which begins with a turgor of 0·55 MPa may only regulate back to 0·65 MPa, resulting in less than 100% regulation. The regulatory process itself may not be uniform, but more rapid responses may be initiated in response to a higher turgor.

Role of calcium in turgor regulation

The model for turgor regulation presented in Fig. 9 and described above is confirmed by these experiments in many but not all cases. Of particular interest is Table 1, which compares the electrophysiological responses in cells which spontaneously varied in their turgor response. These data confirm that depolarization of the membrane and increase in conductance, which are preceded by a decrease in streaming rate, always occur when rapid turgor regulation occurs, and therefore appear to be necessary for rapid turgor regulation. To further test this theory, we attempted to interrupt the regulatory process by preventing the entrance of Ca2+ following hypotonic stress. We utilized the Ca2+-channel blocker nifedipine, which has been shown in the Charophyte Lamprothamnium to inhibit the turgor regulatory response ( Okazaki & Tazawa 1986). At 50 and 25 μM, nifedipine had the anticipated effect, i.e. it abolished all components of the turgor regulatory response. Unlike previous experiments with reduced Ca2+ ( Okazaki & Tazawa 1990; Bisson et al. 1995 ), even the initial depolarization was inhibited. This may be due to a requirement for a smaller influx of Ca2+, that was not inhibited in previous experiments but is inhibited here. The possibility that nifedipine has some independent effect that prevents depolarization, such as stimulation of the electrogenic pump or inhibition of K+ channels remains to be tested, although no such effects of nifedipine have been shown.

At 5 μM, nifedipine had an unexpected effect. Five out of six cells tested did initiate turgor regulation, but it occurred more slowly than in the control cells. In the regulating cells, the membrane depolarized less than in the controls, and a cessation of streaming was never seen. After a long time, streaming was examined only over long (30 min) intervals, and therefore a brief cessation may have been missed. However, even in those cells which responded rapidly, when streaming and electrophysiological parameters were monitored at 5 min intervals, no slowing or cessation of streaming was observed. Conductance did increase, again at later times, but still preceding initiation of turgor regulation.

These experiments suggest that there is a second, slower response, which can regulate turgor when the earlier response failed. The occurrence of two different regulatory responses had been suggested earlier ( Bisson et al. 1992 ). A second, small increase in conductance is often seen under control conditions ( Hoffmann & Bisson 1990; Bisson et al. 1995 ), but its significance was unclear. We had also earlier reported that a different Ca2+-channel blocker, verapamil, sometimes blocked the early conductance increase, but sustained larger later increase in conductance ( Bisson et al. 1992 ). In those experiments, however, turgor and electrophysiological responses were being measured in different cells, so we could not assess the relationship between the increase in conductance and the initiation of the turgor response, which we show here in the presence of 5 μM nifedipine ( Fig. 7).

Regulation in the presence of 5 μM nifedipine requires us to add a second loop to the model shown in Fig. 9, shown with dashed arrow. Regulation occurs in the presence of a Ca2+ channel blocker without an apparent cessation of streaming, and while the cell remains negative to EK. This suggests a parallel signal transduction pathway which may run simultaneously with the rapid, primary response which normally dominates the process. We suggests that this pathway generates a small second increase in conductance even under control conditions. However, when the primary pathway is inhibited, the second, late pathway dominates. This second pathway does not entail a decrease in streaming, which suggests that it does not involve an increase in cytosolic calcium, although we cannot exclude the possibility that there is a Ca2+ increase by a nifedipine-insensitive pathway (or a pathway sensitive to 25μM but not 5μM nifedipine) which is too small, too brief, or too localized to affect streaming rates. An alternative explanation, based on the model of Beilby & Shepherd (1996) is that the early increase in conductance, due to the Ca2+-dependent Cl conductance, is inhibited, but the later increase, due to an increased K+ conductance, is not. Although this is consistent with the activity of nifedipine as an inhibitor of the hypotonic stress-induced Ca2+ influx, it does not explain the fact that turgor decreases during this time, since if K+ conductance increases when the cell is negative to EK, K+ should enter rather than leave the cell, and turgor should increase. This is a problem for both models. There are two possible explanations: that K+ is lost from the cell by an active transport mechanism, or that turgor is lost by some mechanism other than by K+ loss. These two possibilities remain to be tested.

We tested whether nifedipine inhibited 45Ca2+ influx. Ca2+ influx ( Table 2) is similar to that measured earlier in C. longifolia ( Bisson et al. 1995 ), in which control fluxes ranged from 10 to 20 nmol m−2 s−1, and stressed fluxes from 10 to 73 nmol m−2 s−1. We show here that nifedipine did inhibit the stress-induced Ca2+ influx, but not the control flux. This is consistent with the effect of nifedipine on the turgor regulatory process being due to an inhibition of Ca2+ influx. Although the nature of the Ca2+ influx was altered by osmotic stress, becoming nifedipine sensitive, the total influx did not increase under these conditions. This was also seen in some cases in earlier studies ( Bisson et al. 1995 ). In order for these two contradictory facts to be true, the control, nifedipine-insensitive Ca2+ flux must be inhibited to the same extent that the stress-induced flux is enhanced. How can this be reconciled with the hypothesis that an increase in cytosolic Ca2+, due at least in part to an increase in Ca2+ influx, is responsible for initiating the turgor regulatory response? The enhanced influx which is necessary to raise cytosolic Ca2+ levels must be too small or too brief to be seen with our methods, which require up to 60 s, with the increase in nifedipine-sensitive flux quickly balanced by a decrease in nifedipine-insensitive control flux.

Conclusions

These studies suggest that our earlier model for turgor regulation must be expanded by addition of a second regulatory pathway. This pathway is initiated more slowly, and may be independent of Ca2+. It may occur simultaneously with the first pathway, which would explain the second, small increase in conductance seen in normal turgor regulation. When the first pathway is inhibited, however, it becomes the main mechanism of turgor regulation, and the slow conductance increase is much larger. It must involve a different mechanism for alteration of internal osmotic pressure, which does not rely on passive K+ loss. Elucidation of this mechanism awaits further experiments.

ACKNOWLEDGMENTS

This grant was supported by USDA No. 96351003164. In addition, grant No. INT 9600020 from NSF enabled M.A.B. and E.A.K. to travel to the laboratory of A. D. Tomos to study the turgor probe technique. Travel to the laboratories of D. Cosgrove and J. Boyer was also useful in familiarizing M.A.B. with the probe. G. L. Zhu initially set up the turgor probe in our lab. We also thank G. Nottingham for his skill, patience, and help in machining the probe.

Ancillary