The Contractile Vacuole Fluid Discharge Rate is Determined by the Vacuole Size Immediately Before the Start of Discharge in Paramecium Multimicronucleatum

It is well known that the contractile vacuole of the freshwater ciliate Paramecium changes its size and pulsation frequency to control the fluid output of the cell in response to changes in the external osmotic conditions. The vacuole is larger and the pulsation frequency higher when the external solution is more dilute (Kitching, 1967; Patterson, 1980; Zeuthen, 1992). We recently reported (Naitoh et al. 1997) that fluid discharge from contractile vacuoles of P. multimicronucleatum in vivo was caused by the cytosolic pressure and that the rate of discharge was higher when the vacuole size at the start of discharge (systole) was larger.

According to the Law of Hagen and Poiseuille, the rate of fluid discharge from the contractile vacuole is determined by the pressure difference between the inside of the vacuole and the external solution, and by the width and length of the cylindrical pore of the vacuole. It can be anticipated that the cell will swell more when it is bathed in a dilute solution, as more water will enter the cell under these conditions. The higher rate of discharge from larger vacuoles in dilute solution might, therefore, be attributable to an increased cytosolic pressure due to swelling of the cell. Alternatively, the higher rate of discharge could also be attributable to the change in the pore shape, i.e. widening and/or shortening of the pore. To determine which of these factors, the pressure and/or the pore shape, affects the rate of fluid discharge, we examined the quantitative relationship between the rate of discharge and the diameter of the vacuole at the start of discharge under various conditions. We conclude that the pore shape of the vacuole, particularly its diameter, is a major factor in determining the rate of fluid discharge from the vacuole. We will also discuss the involvement of tension at the surface membrane of the vacuole, which is proportional to the membrane area (Naitoh et al. 1997), in determining the pore shape. Some of the results described in this paper have been presented verbally elsewhere (Allen et al. 1996).

Cells of P. multimicronucleatum (syngen 2) (Allen et al. 1988) were grown in an axenic culture medium (Fok and Allen, 1979) and were harvested at the mid-logarithmic growth phase. These cells were washed with a standard saline solution containing (final concentration in mmol l^{− 1} ): 0.5 NaCl, 2.0 CaCl₂, 1.0 MgCl₂, 1.0 KCl and 1.0 Tris–HCl buffer (pH 7.4). Cells were then transferred into each experimental solution and equilibrated for more than 40 min prior to experimentation. The solutions employed were a standard saline solution, a saline solution diluted with water to 1/10 or 1/100 of the original concentration and an 80 mmol l^{− 1} sorbitol-containing saline solution.

Equilibrated cells were slightly compressed between a glass slide and a coverslip by introducing 20 µ m diameter polystyrene beads into the medium. Profile views of contractile vacuoles were video-recorded (AG-6300, Panasonic Indust. Co., Secaucus, NJ, USA) at 30 ms per frame using Nomarski microscope optics (Leitz × 60 objective, Leica Mikrosk u. System, GmbH, Wetzlar, Germany) and a video camera (CCD-72, Dage MIT Inc., Michigan City, IN, USA). Video images were fed into a computer (Power Macintosh 7100/80, Apple Computer Inc., Cupertino, CA, USA) using a frame grabber (LG-3, Scion Corp., Frederick, MD, USA) for analysis of the time course of the change in vacuole diameter during systole (Inoué, 1986). The analysis was performed using the public domain NIH Image Program (developed at the US National Institutes of Health and available on the Internet at http://rsb.info.nih.gov./nih-image/).

As shown in Fig. 1, the contractile vacuole was essentially spherical in the early and middle stages of systole. Later, it took on the shape of an ‘Erlenmeyer flask’ as a result of the presence of microtubules extending from the pore region and passing over the region of the vacuole next to the pore (Hausmann and Allen, 1977). For this reason, we measured the vacuole diameter only in the early and middle phases of systole to determine the rate of discharge.

For electron microscopy, a cell in a small droplet of saline solution was fixed instantaneously by squirting a fixative containing 2 % glutaraldehyde in 0.05 mol l^{− 1} cacodylate buffer against the cell through a fine-bore pipette placed close to the cell. The precise moment of fixation was monitored through a video camera, and the phase of discharge was later determined by observing the resulting video frames. Conventional techniques (Allen and Fok, 1988) were employed for obtaining the subsequent electron micrographs of the same cells as recorded by video microscopy. All the experiments were performed at a room temperature of 24–26 °C.

The internal diameters of contractile vacuoles at the start of systole (their initial diameters) and their rates of fluid discharge were determined in three groups of cells treated with three different solutions; standard saline solution, 1/10-diluted saline solution and 80 mmol l^{− 1} sorbitol-containing saline solution. As shown in Fig. 2, the rate of discharge was significantly higher in 1/10-diluted saline solution and lower in 80 mmol l^{− 1} sorbitol-containing saline solution than in standard saline solution (ANOVA, P<0.01). The initial diameters were significantly larger in 1/10-diluted saline solution and smaller in 80 mmol l^{− 1} sorbitol-containing solution than in standard saline solution (ANOVA, P<0.01). It is clear that the rate of discharge was higher in the cell group where the contractile vacuoles had larger initial diameters.

To examine the relationship between the rate of discharge and the initial vacuole diameter, the values for the rates of discharge for all contractile vacuoles obtained from the three different groups of cells treated with three different solutions, i.e. standard saline solution, 1/10-diluted saline solution and 80 mmol l^{− 1} sorbitol-containing solution, were plotted against their respective initial diameters (Fig. 3). All the points in the figure were distributed close to the same curved line.

During the course of these experiments, we also found that the initial vacuole diameter at which the vacuole enters systole changed after a cell had been compressed in a thin (20 µ m) space. The time course for the change in the initial diameter and that for the rate of discharge after compression were determined in the same cell bathed in standard saline solution. A representative result is shown in Fig. 4. The initial diameter (open circles in Fig. 4), 13.5 µ m at the beginning of compression, decreased to a minimum value of 11.3 µ m approximately 6 min after the start of compression. The diameter then gradually increased to an approximately steady value of 12.2 µ m at 40 min after the start of compression. The rate of discharge (triangles in Fig. 4) changed in parallel with the change in the initial diameter, i.e. the rate was 5.3× 10³ µ m³ s^{− 1} at the beginning of compression and decreased cells with contractile vacuoles that had different initial diameters. The cells were then individually fixed at a specific phase of the contractile vacuole cycle. Each vacuole was then serially sectioned, and electron micrographs were used to determine the diameter and the length (depth) of the pore. Fig. 5A shows a representative micrograph of a contractile vacuole in late systole obtained from a cell treated with standard saline solution, and Fig. 5B shows a micrograph of 40 nm membrane tubules that are bound to the bands of microtubule ribbons that arise from the vacuole pore. Table 1 shows morphological data obtained for 15 contractile vacuole pores obtained from 10 different cells. The mean value for the pore diameter was 1.5±0.3 µ m (N=15) and for the pore length was 2.0±0.2 µ m (N=9; mean ± S.D.).to a minimum value of 3.8× 10³ µ m³ s^{− 1} approximately 6 min after the start of compression. The rate then increased to a steady value of approximately 4.3× 10³ µ m³ s^{− 1} at 40 min after the start of compression.

To determine whether the vacuole pore morphology changes with an increase or decrease in the discharge rate, the cells were first treated with four different solutions (standard saline solution, 1/10-diluted saline solution, 1/100-diluted saline solution and 80 mmol l^{− 1} sorbitol-containing saline solution) in order to obtain

An approximate cytosolic pressure when the cells were in standard saline solution was calculated according to the Law of Hagen and Poiseuille using a vacuole discharge rate of 4.4× 10³ µ m³ s^{− 1} (Fig. 2), a vacuole pore radius of 0.75 µ m and a vacuole pore length of 2.0 µ m (Table 1) to be 73 N m^{− 2} . This pressure is far smaller than the osmotic pressure difference (approximately 5× 10⁵ N m^{− 2} ) would be between a standard saline solution (approximately 15 mosmol l^{− 1} ) and the cytosol (approximately 200 mosmol l^{− 1} ). However, this pressure is larger than the cytosolic pressure of some marine invertebrate eggs (approximately 5 N m^{− 2)} (Cole, 1932; Yoneda, 1964). In marine invertebrate eggs, the osmotic pressure difference between sea water and the cytosol is negligible and, therefore, the cytosolic pressure is caused solely by tension at the cell surface (surface boundary tension and/or membrane elastic tension) (Harvey, 1954). In contrast, cells of P.multimicronucleatum are hyperosmotic to their environment (approximately 200 mosmol l^{− 1} for the cytosol versus 40 mosmol l^{− 1} for the culture medium). The external water, therefore, enters the cell continuously through the water-permeable cell membrane, and much of the excess water must be expelled through the contractile vacuole system (Patterson, 1980). The cytoplasm of P. multimicronucleatum, therefore, must exert a slight positive pressure on the cell membrane, together with its associated cytoskeletal elements, due to the swelling of the cell that offsets the osmotic pressure. Linder and Staehelin (1979) estimated the pressure difference between the lumen of the contractile vacuole and the external solution (in the flagellar pocket) in the flagellate Leptomonas collosoma, on the basis of the rate of fluid discharge and the pore size, to be 6.7× 10⁴ N m^{− 2} . This value is higher than the value calculated for P. multimicronucleatum.

The rate of fluid output through the contractile vacuole system of Paramecium is higher when the cell is in more dilute solution (Ishida et al. 1993; Kitching, 1967; Patterson, 1980; Zeuthen, 1992). It is, therefore, highly probable that the degree of osmotic swelling of the cell that determines the cytosolic pressure is kept approximately constant irrespective of changes in the external osmotic pressure, at least in the range we have employed (see also Rifkin, 1973, for Tetrahymena pyriformis). However, we initially assumed that compression of the cell should increase the cytosolic pressure. Therefore, it seemed reasonable to predict that the rate of discharge of a vacuole would be increased by cellular compression. Contrary to this prediction, however, the rate of discharge of a vacuole decreased after compression (Fig. 4). Previously we had found (Naitoh et al. 1997) that disruption of the decorated spongiome, the locus of fluid segregation in P. multimicronucleatum (Ishida et al. 1993), did not change the rate of discharge from a filled vacuole, although the cell tended to swell due to the overall decrease in fluid output. These findings suggest that osmotic swelling of a cell or a mechanical compression of the cell, at least to the extent used in the present experiments, does not produce an increase in the cytosolic pressure large enough to bring about a significant increase in the rate of contractile vacuole discharge. In other words, the cytosolic pressure remains almost constant under the experimental conditions employed. The change in the pore shape may therefore be responsible for the change in the fluid discharge rate when the cells are subjected to different external conditions, although the exact cause remains to be determined.

The rate of discharge was 8.8× 10³ ±4.5× 10³ µ m³ s^{− 1} (N=56) from vacuoles in cells equilibrated in 1/10-diluted saline solution. This value is approximately five times larger than that for the rate from vacuoles in cells equilibrated in 80 mmol l^{− 1} sorbitol-containing saline solution (1.8× 10³ ±0.7× 10³ µ m³ s^{− 1}, N=33) (Fig. 2). If R were to remain unchanged during the switch from a dilute to a more concentrated solution, then L in 80 mmol l^{− 1} sorbitol-containing saline solution would be five times longer than that in 1/10-diluted saline solution. Conversely, if L were to remain unchanged, then R in 1/10-diluted saline solution would be 1.5 times wider than that in 80 mmol l^{− 1} sorbitol-containing saline solution.

Our examination of serial-sectioned vacuole pores by electron microscopy (Table 1) showed no obvious differences in L and R between those vacuoles in 1/10-diluted saline solution (L, 2.1 µ m; R, 0.87 µ m) and those in 80 mmol l^{− 1} sorbitol-containing saline solution (L, 2.2 µ m; R, 0.75 µ m), although the sample size was too small to examine the measured values statistically. It is apparent that the shape of the vacuole can be somewhat distorted when the vacuole is subjected to electron microscopical procedures. It is, therefore, conceivable that a 1.5-fold increase in R in the 1/10-diluted saline solution would not be easily recognizable in the electron micrographs. In contrast, a fivefold increase in L should be detectable even though the shape of the vacuole becomes fairly distorted after fixation and sectioning for electron microscopy. Since we never saw such a large elongation of the vacuole pore, we concluded that a change in the pore radius is the major cause for the change in the rate of discharge. This, however, does not rule out some minor change in the pore length. It can be concluded that the vacuole pore is wider (and somewhat shorter) when the overall vacuole size is larger at the beginning of systole. In this connection, it should be noted that a widening and a shortening of the contractile vacuole pore were reported during the diastolic phase of vacuole filling in T. pyriformis. This was demonstrated electron microscopically many years ago by Cameron and Burton (1969).

To determine the relationship between vacuole size at the start of systole and vacuole pore size, as represented by the rate of discharge, all the data presented in Figs 3 and 4 were combined and replotted on a logarithmic scale (Fig. 6). A straight line with the slope of 2 was drawn according to the least-squares method so as to fit all the points in the figure (χ² =1.777, N=156). It is clear from this figure that the different groups of cells, irrespective of treatment, were all distributed along this line. The slope of 2 implies that the membrane area is proportional to the rate of discharge. This suggests that the vacuole membrane area affects the vacuole pore shape in the same way when the external conditions are changed.

In our previous paper (Naitoh et al. 1997), we proposed a hypothesis that the internal pressure of a contractile vacuole in a ruptured cell is caused by the tension of the vacuole membrane. The tension would be proportional to the planar vacuole membrane area. This tension would be generated as the 40 nm (diameter) tubules became flattened into an essentially planar vacuole membrane. Membrane tubulation had previously been reported to occur when the vacuoles of normal (unruptured) cells entered systole (Allen and Fok, 1988). It is, therefore, conceivable that membrane tubulation-related tension is responsible for determining the pore shape.

The contractile vacuole membrane is anchored to the microtubule ribbons which originate from around the vacuole pore (Fig. 5) (Hausmann and Allen, 1977). Stretching of the pore structure caused by a build-up of membrane tension as more tubules become flattened to form the expanding contractile vacuole could conceivably bring about a shortening and/or a widening of the pore.

As shown in Fig. 1C, the rate of fluid discharge was constant during the early and middle phases of systole (Naitoh et al. 1997). This suggests the presence of a ratchet mechanism in the pore structure which keeps the pore shape constant during systole. Thus, the shape of the pore is set by the membrane tension immediately before the start of each discharge. It should also be noted that the rate of vacuole discharge, in the last stage of systole, varies from this earlier constant rate. This is clearly shown in Fig. 1C, where values deviated upwards from the straight line at all points in the late phase of systole. It is clear that the rate of discharge decreases in the late phase of systole, suggesting that the vacuole pore size decreases, most probably as a result of narrowing of the vacuole pore.

At present, we do not know what cytoskeletal structure corresponds to this hypothetical ratchet mechanism or how this structure could function as a ratchet. However, it is highly probable that the mechanism resides in the helically arranged microtubules surrounding the pore membrane (Hausmann and Allen, 1977; McKanna, 1973). As early as 1967, McIntosh and Porter proposed a hypothetical mechanism for generating a force capable of bringing about a sustained constriction of the sperm head by a highly ordered helical array of microtubules.

According to biophysical studies on the fusion of phospholipid bilayers and vesicles (Chernomordik et al. 1995), an increase in bilayer tension was a key factor for promoting fusion. In their patch-clamp studies on the degranulation of mast cells, Monck and Fernandez (1992) and Monck et al. (1990) found that high membrane tension in the secretory granule was the critical stress for bringing about its exocytotic fusion. It is, therefore, probable that stretching of membrane tubules to cause tension is involved in the contractile vacuole pore opening, in which fusion of the vesicle membrane with the plasma membrane covering the pore occurs at the start of each systole.

This work was supported by NSF Grant MCB 95 05910.