Electrical properties and fusion dynamics of in vitro membrane vesicles derived from separate parts of the contractile vacuole complex of Paramecium multimicronucleatum
Pacific Biomedical Research Center, Snyder Hall 306, University of Hawaii at Manoa, 2538 The Mall, Honolulu, HI 96822, USA
Author for correspondence (e-mail:
naitoh{at}pbrc.hawaii.edu)
Accepted 24 August 2005
![]() |
Summary |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The membrane vesicles shrank when the external osmolarity was increased, and swelled when the osmolarity was decreased, implying that the contractile vacuole complex membrane is water permeable. The water permeability of the membrane was 420x107 µm s1 Pa1. The vesicles containing radial arm membrane swelled after initially shrinking when exposed to higher external osmolarity, implying that the V-ATPases energize osmolyte transport mechanisms that remain functional in the vesicle membrane. The vesicles showed an abrupt (<30 ms), slight, slackening after rounding to the maximum extent. Similar slackening was also observed in the contractile vacuoles in situ before the opening of the contractile vacuole pore. A slight membrane slackening seems to be an indispensable requirement for the contractile vacuole membrane to fuse with the plasma membrane at the pore. The contractile vacuole complex-derived membrane vesicle is a useful tool for understanding not only the biological significance of the contractile vacuole complex but also the molecular mechanisms of V-ATPase activity.
Key words: contractile vacuole complex, membrane vesicle, membrane potential, membrane resistance, membrane dynamics, membrane tension, patch clamp, water permeability, V-ATPase, Paramecium multimicronucleatum
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Excess cytosolic water is translocated into the lumen of the CVC in
association with transport of cytosolic osmolytes, most likely K+
and Cl (Stock et al.,
2002a,b
).
The transport is assumed to be energized by the electrogenic V-ATPase activity
in the decorated spongiome membrane (Allen
and Fok, 1988
; Ishida et al.,
1993
,
1996
;
Fok et al., 1995
;
Merzendorfer et al., 1997
;
Tominaga et al.,
1998a
,b
;
Wieczorek et al., 1999
;
Grønlien et al., 2002
;
Iwamoto et al., 2004
). The
fluid flows into the CV through collecting canals, causing the CV to swell
(the fluid-filling phase). After a certain length of time the CV begins to
round and the radial arms detach from the CV (the rounding phase). The
ampullae begin to swell immediately after detachment of the radial arms, since
the cytosolic water is continuously translocated into the lumen of the radial
arm. At the end of the rounding phase the CV membrane fuses with the plasma
membrane at the CV pore and the pore opens. The CV fluid is pushed out of the
CV through the open pore by cytosolic pressure (Naitoh et al.,
1997a
,b
),
and the CV collapses and becomes microscopically invisible (the
fluid-discharging phase). Immediately after fluid discharge the pore closes as
the CV membrane is separated from the plasma membrane at the pore. The radial
arms then reattach to the CV, so that the ampullae collapse while the CV
swells. The CV continues to swell as fluid flows into it from the collecting
canals (next fluid-filling phase).
Tominaga et al. (1999)
found that the vesicles derived from the fragmented membrane of the CVC in
mechanically ruptured cells continue to show rounding and slackening cycles
but the vesicle cycles occur at different intervals, i.e. they are out of
phase one with the other. This finding implies that the CVC membrane itself
possesses a mechanism(s) by which its tension is periodically altered. Using a
microcantilever placed on the surface of a CVC membrane vesicle, Tani et al.
(2002
) directly measured the
tension of the vesicle membrane and found that the tension increased to
5
mN m1 as the vesicle rounded, and then returned to
0.1
mN m1 as the vesicle slackened.
Based on these findings we proposed a hypothesis that periodic changes in
the CVC membrane tension govern the timing of the fluid discharge cycles of
the CVC. That is, an increase in the tension leads to severing of the radial
arms from the CV and to rounding of the CV, followed by opening of the pore
that leads to CV fluid discharge. Conversely, a decrease in the tension causes
closure of the pore and reattachment of the radial arms to the CV at the start
of the fluid-filling phase (Allen and
Naitoh, 2002; Tani et al.,
2002
).
Tani et al. (2000)
demonstrated that fusion of two CVC membrane-derived vesicles with different
periods of roundingslackening occurred when both vesicles were in their
slackening phases and the period of roundingslackening cycles after
fusion was closest to that of the fusing vesicle that had the shorter period.
They also demonstrated that the CVC membrane-derived vesicle exhibited an
extra rounding in response to suction of a small portion of the membrane into
a micropipette when the vesicle was in the slackening phase. These findings
support the idea that the mechanism(s) in the CVC membrane that controls the
membrane tension is mechano-sensitive.
The primary objectives of the present study were to examine the presence of
specific ion channels in the CVC membrane, i.e. (i) mechano-sensitive ion
channels that are postulated to be involved in the control of tension
development in the membrane and (ii) voltage-sensitive ion channels that are
postulated to be involved in the control of the fluid segregation mechanisms
energized by the electrogenic V-ATPase. Conventional patch-clamp techniques
were employed for the present study
(Ogden, 1994;
Rudy and Iverson, 1992
).
Besides data on channels in the CVC membrane, we found some interesting
CVC-derived membrane dynamics, which are also described in this paper.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Experimental solutions
Artificial cytosol
The cytosolic osmolarity was found to be 160 mOsmol
l1 (Stock et al.,
2001
) and the cytosolic KCl concentration was
40 mmol
l1 (Stock et al.,
2002b
) in Paramecium cells growing in a standardized
medium of
80 mOsmol l1. Therefore, a mixture (mmol
l1 final concentration) of 40 KCl, 10 MgCl2, 1
EGTA, 50 sorbitol and 10 Hepes (pH 7.0) was used as an artificial cytosol for
the reference electrode.
Artificial CV fluid
The cytosolic KCl concentration was 80 mmol l1
(Stock et al., 2002b
). The
osmolarity of the CV fluid was assumed to be equal to the cytosolic osmolarity
(Tominaga et al., 1998b
).
Therefore a mixture of 80 mmol l1 KCl and 10 mmol
l1 Hepes (pH 7.0; Stock
et al., 2002a
) was used for the patch electrode as an artificial
CV fluid.
Osmolarity-changing solution
A solution of 3 mol l1 KCl, 2 mol l1
sorbitol or 0.5 mol l1 sorbitol was used to increase and
distilled water to decrease the external osmolarity around the vesicle.
Assembly of the experimental apparatus
All experiments were performed on the stage of an inverted microscope
(Leitz DMIRB, Leica Mikroskopie und Systeme GmbH, Wetzlar, Germany) placed on
a vibration-free table (LW 3030B-OPT, Newport Corp., Irvine, CA, USA) at a
regulated room temperature of approximately 25°C. Five micromanipulators
(one PCF 5000, Burleigh Instruments, Inc., Fishers Victor NY, USA and four
MHW-3, Narishige Group, East Meadow, NY, USA) were used for cell incision and
manipulation of the extruded CVC-derived membrane vesicles. A VCR (Panasonic
ACC 6300, Matsushita Electric Inc., Osaka, Japan) and a CCD camera (CCD-72,
DAGE MIT, Michigan City, IN, USA) were used for continuous recording of the
images of incised cells and their CVC membrane vesicles during
experimentation, so that the origin of the membrane vesicle being studied
could be distinguished. A patch-clamp amplifier (EPC-8, Hekaelektronik,
Lambrecht, Germany) and computer interface (ITC-16, Instrutech Corp., Great
Neck, NY, USA) were employed. A computer (Macintosh Power Mac G4, Apple
Computer Inc., Cupertino, CA, USA) was used to control experiments and for
acquisition and subsequent analysis of the data.
Experimental procedures
Extrusion of the CVC-derived membrane vesicles from the cell
Paramecium cells were introduced into a mineral oil droplet on a
coverslip together with a minute amount of axenic culture medium. Excess
culture medium surrounding the cell was removed using a suction micropipette
(approximately 3 µm i.d.) until the cells were compressed and flattened by
the mineral oilartificial cytosol boundary tension. The plasma membrane
adjacent to the CV was incised using a microneedle. The CVC was extruded
together with the cytosol through the incision by cytosolic pressure (Naitoh
et al.,
1997a,b
).
Sometimes the microneedle was needed to push the CVC out of the cell through
the incision. The extruded CVC was always transformed into several
membrane-bound vesicles (Tominaga et al.,
1998b
).
Patch-clamp experiments
The tip of a reference electrode (approximately 3 µm i.d.) filled with
artificial cytosol was placed in the extruded cytoplasm. The tip of a
patch-clamp pipette filled with artificial CVC fluid was first placed in the
mineral oil surrounding the extruded cytoplasm containing the CVC membrane
vesicles. A slight positive hydrostatic pressure was applied to the inside of
the patch pipette, so that the artificial CV fluid was gently pushed out of
the pipette when its tip was moved into the extruded cytosol from the
surrounding mineral oil. The tip was then gently moved toward a membrane
vesicle. When the tip touched the surface of the membrane vesicle, the overall
electrode resistance slightly increased. A slight negative pressure was then
applied to the inside of the pipette, so that the membrane of the vesicle
became firmly attached to the opening of the pipette. In favorable
preparations the overall electric resistance increased to the giga-seal level
in several seconds. Voltage-clamp experiments in the on-vesicle patch-clamp
mode were then performed. To perform experiments in the whole-vesicle
patch-clamp mode, a minute negative pressure was further applied to the
pipette until the overall electric resistance of the patch electrode suddenly
decreased to a value comparable to the input resistance of the vesicle.
Changing the osmolarity around the CVC membrane vesicles
The tip (5 µm i.d.) of a micropipette filled with the
osmolarity-changing solution was placed close to the CVC membrane vesicle and
the solution was ejected over several seconds by gently raising the
hydrostatic pressure inside the pipette.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The IV relationship was linear in a voltage range between 80 and +80 mV in all vesicles examined (>20). Most of the vesicles showed the membrane potential (the potential at zero current) to be positive with reference to the excised cytosol (540 mV). Vesicles formed from only the CV membrane showed little membrane potential (less than 5 mV).
Representative results obtained from two different cells are shown in Fig. 1. As shown in Fig. 1Aia, three small, radial arm-derived vesicles (identified by blue dots) and a larger CV-derived vesicle (red dot) were seen immediately after incision of the cell. These `blue' vesicles were then topographically separated from the `red' vesicle by placing a microneedle between them to prevent their fusion, as shown in Fig. 1Aib.
|
As shown in Fig. 1Bie, a large `blue' vesicle formed by fusion of a CV with some radial arm vesicles lies near to two smaller `blue' vesicles that formed from the membrane along two radial arms, respectively, immediately after cell incision. These three `blue' vesicles later fused into a large vesicle (a `blue' vesicle in Fig. 1Bif). A `red' vesicle in Fig. 1Bie is the other CV of the cell and was in the rounding phase, becoming detached from the radial arms at the moment of cell incision. The CV-derived vesicle is shown in Fig. 1Bif as a `red' vesicle.
The IV relationship for the `blue' vesicle in
Fig. 1Bif is shown as
a blue line labeled f in Fig.
1Bii. The input resistance and the membrane potential estimated
from the line were 70 M
and
20 mV, respectively. This `blue'
vesicle later fused with the CV-derived vesicle
(Fig. 1Big) into a
larger vesicle (labeled half blue, half red in
Fig. 1Bih). The
IV relationship for the fused vesicle is shown as a half-blue,
half-red line labelled h in Fig.
1Bii. The input resistance and the membrane potential for the
vesicle estimated from the line were
58 M
and
14 mV,
respectively.
The electrical characteristics of a membrane patch from the membrane vesicle examined in the on-vesicle patch-clamp mode
Membrane currents generated in a membrane patch on a CVC-derived membrane
vesicle in response to voltage steps varied in the range 80 to +80 mV
were examined in the on-vesicle patch-clamp mode. As shown in
Fig. 2, no voltage-gated
current was observed in the membrane patch. In all vesicles tested no membrane
patches were found that showed voltage-gated unit currents.
|
|
In Fig. 3Ai, e and f correspond to the vesicle's maximum rounding and the start of slackening, respectively. Although the relationship between changes in Rm and Vm accompanied by roundingslackening of the vesicle is rather ambiguous, Rm tended to increase as the vesicle rounded (Fig. 3Aiiae), while Vm decreased when the vesicle showed maximum rounding (Fig. 3Aiiid,e). Inversely, Rm decreased (Fig. 3Aiieg), while Vm increased (Fig. 3Aiiieg) as the vesicle slackened.
Fig. 3Bia corresponds to the slackening state of this vesicle. The vesicle started to round (Fig. 3Biag) and finally ruptured (Fig. 3Bih). Rm continued to increase (Fig. 3Biibg), while Vm continued to decrease (Fig. 3Biiibg) as the degree of rounding increased with time until the vesicle ruptured. It should be noted that Vm became slightly negative immediately before the vesicle's rupture.
Change in vesicle volume after changing the surrounding osmolarity
To examine whether the CVC membrane is permeable to water, the osmolarity
of the excised cytosol surrounding the membrane vesicles was changed by
applying a more concentrated solution of solutes (3 mol l1
KCl or 1 mol l1 or 0.5 mol l1 sorbitol) or
distilled water to the cytosol through a micropipette placed close to the
vesicles, at the same time that images of the vesicles were continuously
video-recorded for subsequent estimation of vesicle volume.
Decreasing the osmolarity
When the surrounding cytosol was diluted by distilled water, all vesicles
swelled to an extent that depended on their initial size and the degree of
cytosol dilution. When the cytosol was extensively diluted, vesicles ruptured
after they had swelled to their maximum size. A representative result is shown
in Fig. 4, where
Fig. 4Ai shows a consecutive
series of five frames of one vesicle and
Fig. 4Aii the time course of
swelling of the vesicle after cytosol dilution. Vesicle volume increased to a
stationary level of 2.5 times larger than the initial level
130 s
after the start of dilution.
|
Increasing the osmolarity
When a more concentrated solution of solutes was added to the surrounding
cytosol, all the vesicles decreased in size, depending on their initial sizes
and the final osmolarity of the solution. When the final osmolarity was very
high, the vesicles decreased in size until they became almost invisible. A
representative result is shown in Fig.
4, where Fig. 4Bi
shows a series of five consecutive frames of the vesicle and
Fig. 4Bii shows the time course
of shrinkage of the vesicle after increasing the external osmolarity. The
volume of the vesicle decreased to a stationary level of 1/10 of the
initial level
20 s after the increase in the osmolarity.
Reswelling of the membrane vesicles held in an increased external osmolarity
The membrane vesicle was found to be able to swell again after it had
decreased in volume in an increased osmolarity. Some vesicles even ruptured
after they had reached their respective maximum sizes. A representative result
is shown in Fig. 5, where
Fig. 5Aaf shows
a consecutive series of frames of a vesicle in which the volume increases;
Fig. 5B shows the time course
of increase in volume of the vesicle after it had first reached its minimum
level in an increased external osmolarity. Vesicle volume increased almost
linearly with time to its maximum level in approximately 40 min, then
gradually decreased to a stationary level of a little lower than the maximum
level (Fig.
5gj). Examination of the origin of the
membrane from their video-recorded images revealed that the vesicles deriving
membrane from only CV showed no reswelling when subjected to an increased
osmolarity.
|
Fusion of vesicles follows osmotic shrinkage
The CVC-derived membrane vesicles ended up more or less rounded when they
were kept immersed in the excised cytosol for more than 1 h. They did not fuse
with each other after they had rounded. When a more concentrated solution of
solutes was added to the extruded cytosol, the rounded vesicles decreased in
volume osmotically and some of them were then able to fuse with their
neighboring vesicles. A representative series of consecutive pictures of the
vesicles that showed fusion after increasing the surrounding osmolarity is
shown in Fig. 6.
|
An abrupt slight slackening was seen immediately after the degree of rounding reached its maximum in both in vitro CVC-derived vesicles and in the in situ CVs
The CVC-derived membrane vesicles were found to show an abrupt slight
slackening immediately after maximum rounding, which was then followed by a
more gradual slackening. The abrupt slackening was detected as a slight
increase in the area of the image of the rounded vesicle after removing the
surrounding cytosol and thereby strongly compressing the vesicle under the
mineral oilcytosol boundary tension. A representative result is shown
in Fig. 7, where
Fig. 7Ai shows pictures of the
vesicle and Fig. 7Aii the time
course of change in the area of the vesicle's image.
Fig. 7Aib,c correspond
to the vesicle's maximum rounding and abrupt slackening, respectively. These
pictures were obtained from two consecutive frames in a tape-recorded series
of images, so that the abrupt slackening took place within the time required
to produce a single frame (<30 ms). The slackening was visible to the naked
eye as a faint flicker during a replay of the tape-recorded images, although
the difference in the still pictures between b and c is
inconspicuous. The abrupt increase in the area is clearly shown in the inset
graph of Fig. 7Aii, where the
expanded time course is shown as a red line. Red circles on the red line in
the inset labeled b, c and d, respectively, correspond to
the time when each corresponding picture was taken. It is clear from the inset
that the abrupt increase took place between b and c, as
indicated by an arrowhead labeled S.
|
The abrupt increase was followed by a further increase in the area (red circles labeled d and e that correspond to Fig. 7Bid,e, respectively. The area then abruptly decreased to almost 0 (from e to j in both the pictures and the time course). This decrease in the area corresponds to the phase of discharge of the CV fluid through the CV pore. The pore opening, therefore, took place between e and f, as indicated by an arrowhead labeled P.
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In fact, most of the CVC-derived membrane vesicles will contain membranes
from both the smooth spongiome and the decorated spongiome, since both
membranes are found along most of length of the radial arms and both will be
incorporated in each vesicle derived from the radial arms that forms in the
excised cytoplasm (Tani et al.,
2000,
2001
). We found that the
membrane potential in these vesicles varied within a range 130 mV. The
amplitude of the membrane potential is assumed to be proportional to the
number of the V-ATPases relative to the overall vesicle membrane area
(Grønlien et al., 2002
).
In fact, a membrane potential of
20 mV for a vesicle derived from both
the CV and radial arm membrane (a `blue' vesicle in
Fig. 1Bif) decreased
to
14 mV after the vesicle fused with another vesicle that was derived
only from the CV, which means it did not contain V-ATPases (`red' vesicle in
Fig.
1Bieg), to form a larger vesicle (`half
blue-half red' vesicle in Fig.
1Bih). The membrane potential of the vesicle corresponds
to the point of intersection of the IV plot for the
vesicle (a blue line for the `blue' vesicle and a half blue-half red line for
the `half blue-half red' vesicle in Fig.
1Bii) with the voltage axis. The IV relationships
also reveal that the input resistance (which corresponds to the slope of the
IV plot) of
70 M
for the vesicle before fusion
decreased to
48 M
after fusion. This decrease in the input
resistance corresponds to an increase in the membrane area produced by the
fusion. An electric current generated by the V-ATPases was shunted by an
increased membrane area so that the membrane potential decreased.
Voltage-activated ion channels are not present in the CVC membrane
The IV relationships for the vesicles in the whole-vesicle
patch-clamp mode were always linear in a voltage range from 80 to +80
mV (Fig. 1Aii,Aiii,Bii). This
implies that voltage-sensitive ion channels, gated by a membrane potential
change within this range, are not present in the CVC membrane.
Voltage-activated single-channel currents were not observed in the current
traces for membrane patches in the on-vesicle patch-clamp mode
(Fig. 2A). This strongly
supports the idea that no voltage-activated ion channels are present in the
CVC membrane. By employing a two-electrode current-clamp technique, Tominaga
et al. (1998a) demonstrated
that the IV relationship for the CV in the rounding phase (a
time when the CV is detached from the radial arm membrane) was straight in a
voltage range from 60 to +60 mV and that it was also straight for the
CV in the fluid-filling phase (a time when the CV is connected to the rest of
the CVC) in a voltage range from 20 to +20 mV, and concluded that very
few voltage-activated ion channels are present in any part of the CVC. The
present study confirms this over a wider range of membrane voltage. By
contrast, Yoshida et al.
(1997
) demonstrated a voltage-
and K+-dependent K+ channel in an artificial lipid
bilayer membrane into which a membrane fraction enriched in the CV membrane of
Dictyostelium discoideum was incorporated. They suggest that the
channels may play a role in the cytosolic water transport into the CV lumen of
this cell.
Non-voltage-sensitive ion channels are present in the CVC membrane
The specific membrane resistances of the membrane vesicles were estimated
(Table 1) with six different
vesicles (two derived from the CV, two derived from the radial arm and two
derived from a mixture of both the CV and the radial arm). The membrane areas
of the vesicles were estimated based on the assumption that they were
spherical when their IV relationships were determined. The
mean value for the specific resistance was 5.2±1.3 103
cm2. This value is consistent with the value of
5.7
103
cm2 for the in situ CVC obtained by
Tominaga et al. (1998a
). This
specific resistance value is within a range consistent with those for
conventional cells, including Paramecium
(Eckert and Naitoh, 1970
;
Fain, 1999
), and indicates the
presence of some non-voltage-sensitive ion channels in the membrane.
|
Residual positive membrane potential found in the vesicles derived from
only the CV (Table 1) is
assumed to be caused by the presence of these ion channels. By employing
ion-selective microcapillary electrodes, Stock et al.
(2002a,b
)
found that K+ and Cl activities in the CV fluid
were higher than those in the cytosol. It is, therefore, highly probable that
Cl is responsible for generation of the residual membrane
potential. The residual membrane potential was also observed in the in
situ CV after detachment of the radial arms that occurs immediately
before fluid discharge (Tominaga et al.,
1998a
,b
).
To identify the ion specificity of these channels, extensive examinations of
the effects of various ion species and of differing concentrations on the
vesicle membrane potential are needed.
Rounding and slackening of the vesicle modify the specific membrane resistance and the membrane potential of the vesicle
As is shown in Fig. 3, the
input resistance increased, while the membrane potential shifted in a negative
direction when the membrane vesicle rounded. Inversely, the input resistance
decreased, while the membrane potential shifted towards a positive direction
when the membrane vesicle slackened. Based on membrane capacitance
measurements of the in situ CV, Tominaga et al.
(1998b) found that the
membrane area of the CV did not change during the CV's rounding. Therefore, an
increase in the vesicle's input resistance corresponds to an increase in the
specific membrane resistance of the vesicle and vice versa.
As was previously mentioned, Cl activity is higher in the
CV fluid than the cytosol (Stock et al.,
2002b). It is, therefore, probable that the shift of the membrane
potential towards a negative direction is caused by a decrease in the
Cl conductance of the membrane as the vesicle rounds. By
using a microcantilever placed on the surface of an isolated CVC membrane
vesicle, Tani et al. (2001
)
directly measured an increase in the membrane tension (a force generated by a
vesicle) to
5 mN m1 (which corresponds to
36 nN)
during rounding from its lowest value of
0.1 mN m1
(which corresponds to
1.8 nN) during vesicle membrane slackening. The
membrane tension of
5 mN m1 that causes a decrease in
the ion (presumably Cl) conductance of the vesicle membrane
is much higher than the membrane tension that activates mechano-sensitive ion
channels, i.e.
1 mN m1
(Gustin, 1992
;
Sokabe et al., 1991
;
Morris, 2001
). It can,
therefore, be concluded that there are no conventional mechano-sensitive ion
channels in the CVC membrane. The mechanical distortion of the channel
molecules by a high membrane tension would be expected to cause a decrease in
the leakage conductance of the CVC membrane. According to Sheetz and Dai
(1996
), lytic tension for
lipid bilayers approximates 210 mN m1, a value
comparable to the highest tension generated by the CVC membrane vesicle. In
fact, the in vitro CVC membrane vesicle is sometimes lysed after
rounding (Fig. 3Bih);
it can be assumed that the membrane tension of the vesicle has exceeded the
threshold for membrane lysis.
It was also noted that Vm sometimes became negative
with reference to the cytosol as the vesicle rounded
(Fig. 3Biii). This implies that
some cations are involved in generating the membrane potential of the vesicle.
K+ is the most likely candidate for the negative potential
(Stock et al., 2002b).
K+ leakage conductance is assumed to be less affected by the
membrane tension, so that K+ potential comes to predominate over
Cl potential as the membrane tension increases.
Based on our electrophysiological examinations of the CVC membrane vesicles, we conclude that neither voltage-nor mechano-sensitive ion channels are involved in the mechanisms governing the transport of osmolytes energized by the V-ATPases and those controling the membrane dynamics associated with cyclic fluid discharge in the CVC.
The CVC membrane is water permeable
The in vitro CVC membrane vesicle swelled when the surrounding
excised cytosol was diluted by distilled water
(Fig. 4Ai,Aii), and shrank when
the osmolarity of the cytosol was increased by adding 3 mol
l1 KCl (Fig.
4Bi,Bii) or 2 mol l1 sorbitol (data not shown)
to the cytosol. The degree of swelling was larger when the cytosol was more
diluted. The vesicles ruptured after they had attained their respective
maximum sizes when the extruded cytosol was extremely diluted. At the other
extreme, the degree of shrinkage was larger when the osmolytes in the excised
cytosol became more concentrated. The minimum size of the vesicles in a
solution of increased osmolarity differed from one vesicle to the other. These
results strongly support the idea that the CVC membrane is water permeable.
The water permeability coefficient of the vesicle membrane estimated from the
data shown in Fig. 4
approximated 420x107 µm s1
Pa1. This value suggests the presence of water channels
(such as aquaporins) in the CVC membrane.
Tominaga et al.
(1998a,b
)
proposed a hypothesis for fluid segregation in the CVC that V-ATPases in the
decorated spongiome energize a hypothetical transport mechanism that conveys
osmolytes from the cytosol into the CVC lumen. Water molecules are
consequently osmotically transported from the cytosol into the CVC lumen
through water channels that distribute throughout the CVC membrane. The
osmolarity of the fluid in the CVC lumen is, therefore, virtually equal to
that of the cytosol. Our present finding of water permeability of the membrane
vesicle strongly supports this hypothesis.
Recently Nishihara et al.
(2004) estimated the water
permeability coefficient of the CV membrane in a freshwater amoeba Amoeba
proteus, based on the rate of shrinkage of the isolated CV in a
hyperosmotic solution of 3.8x107 µm
s1 Pa1. This value is strikingly similar
to our value for the water permeability coefficient of the CVC of
Paramecium. Based on this high value for the water permeability, they
suggested the presence of water channels in the CV membrane of the amoeba. A
recent study of the CVC in Trypanosoma cruzi demonstrated the
presence of a functional aquaporin water channel in the CVC of this parasitic
protozoan (Montalvetti et al.,
2004
).
Fluid segregation takes place in the in vitro membrane vesicles
We found that the volume of the CVC-derived membrane vesicles increased
almost linearly with time after the vesicles had decreased to their minimum
size when subjected to an increased osmolarity
(Fig. 5B). This implies that
the hypothetical V-ATPase-energized osmolyte transport mechanisms, which are
present in the vesicle membrane, remain active after the vesicles have been
subjected to an increased osmolarity, and may explain why some vesicles
ruptured after attaining their respective maximum sizes (data not shown).
The rate of increase in vesicle volume was calculated from the slope of the
linear approximation of the time course of change in vesicle volume (a solid
straight line in Fig. 5B) and
was found to be 46.5 fl s1. This value is comparable to
the rate of fluid segregation of a single CVC of an 84 mOsmol
l1-adapted cell in its adaptation solution, i.e.
45 fl
s1 (Stock et al.,
2001
).
Most of the in vitro membrane vesicles were found to swell in the extruded cytosol (data not shown). This swelling ceased concomitantly with cessation of ciliary beating, which requires ATP, on the fragmented plasma membranes in the cytosol. This implies that the swelling needs ATP. Our preliminary observation showed that application of ATP to the excised cytosol reactivated the vesicle volume increase concomitantly with reactivation of ciliary beating, implying that the V-ATPases incorporated into the membrane vesicle are responsible for the water transport from the cytosol into the vesicle lumen, as is also the case in the in situ CVC.
Some notable membrane dynamics in the in vitro CVC membrane vesicles
Osmotic shrinkage of the vesicles permits their fusion
Tani et al. (2000,
2001
) found that the in
vitro CVC membrane vesicles ceased to show roundingslackening
cycles and became more-or-less rounded (the rigor state of the vesicle) when
they were exposed to a lowered ATP concentration. They also found that two
membrane vesicles showing roundingslackening cycles with different
cycle periods fused only when they both showed slackening at the same time.
Fig. 6 clearly demonstrates
that the in vitro CVC membrane vesicles in the rigor state (vesicles
labeled 1, 2 and 3 in Fig. 6,
0) fused with each other when they had slackened in response to an increased
external osmolarity. These observations support the idea that the slackening
of the membrane vesicle, i.e. a decrease in the membrane tension, is
indispensable for membrane fusion, and ATP is not needed for fusion per
se. The homotypic fusion process in the CVC membrane vesicles seems to be
a purely physical phenomenon, as is the case for fusion between artificial
lipid bilayer membrane vesicles.
Rounding of the membrane vesicle was always followed by a slight and abrupt slackening
Frame-by-frame analysis of the video-recorded images of a compressed
membrane vesicle revealed that a slight increase in the area occupied by the
vesicle's image took place immediately after its area had reached a minimum in
a period of time corresponding to a single frame of the tape-recording
(30 ms; Fig. 7Aii). The
increase was visible as a faint flicker to the naked eye in the replayed
tape-recorded images of the vesicle. The minimum area corresponds to the
vesicle at its maximum rounding, or the maximum membrane tension, and a
subsequent slight increase in the image area corresponds to a slackening of
the vesicle, or a slight decrease in the membrane tension.
By using a microcantilever placed on the surface of a membrane vesicle,
Tani et al. (2001) directly
measured cyclic changes in the membrane tension of the vesicle that
accompanied its rounding-slackening cycles. Although they did not mention it,
they clearly demonstrated this slight and abrupt decrease in the membrane
tension from its maximum value of
5 mN m1 for the
rounding phase to
4.7 mN m1 within
0.4 s. This
change was significant but so small that it was not discussed.
An extremely compressed membrane vesicle can be regarded as a right
circular cylinder in shape. The relationship between the membrane tension and
the area of the image of the vesicle compressed by a definite force can be
formulated as:
![]() |
An abrupt slackening after rounding was also observed in the in situ CV
As is shown in Fig. 7Bii,
the area of the image of an extremely compressed in situ CV slightly
increased in a period of time corresponding to a single frame of the video
recording immediately after it had reached its minimum value, much like the
in vitro CVC membrane vesicle. This implies that a slight and abrupt
decrease in the membrane tension took place after the membrane tension had
reached its maximum value (maximum rounding).
It is notable that the area increased more during the next two frames (for
60 ms) after its abrupt increase (Fig.
7Bid,e, Biid,e), then decreased abruptly
(Fig 7Bif,
Biif). This abrupt decrease corresponds to the start of
fluid discharge from the CV, so that the CV area subsequently decreases to 0.
The start of fluid discharge corresponds to fusion of the CV membrane with the
plasma membrane at the cell's CV pore. We conclude that fusion takes place
when the CV membrane tension has begun to decrease from its highest value that
was reached at maximum rounding.
We had previously concluded (Tominaga et al.,
1998b, 2000, 2001) that fusion
of the CV membrane with the plasma membrane at the pore takes place when the
CV membrane tension reached its maximum value. This hypothesis can now be
corrected based on this new information to say that the CVC membrane fuses
with other CVC membrane vesicles or with the pore membrane when the membrane
tension has started to decrease. However, rounding accompanied by an increase
in membrane tension of the CV to its highest value is indispensable for
positioning the CV membrane topographically in the right place for its
subsequent fusion with the pore membrane, a process aided by the cytoskeletal
microtubular structures that bind to the CV and that surround the CV pore.
![]() |
Acknowledgments |
---|
![]() |
Footnotes |
---|
Present address: Laboratory for Brain-Operative Devices, Brain Science
Institute (BSI), The Institute of Physical and Chemical Research (Riken), 2-1
Hirosawa Wako, Saitama 351-01, Japan
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Allen, R. D. and Fok, A. K. (1988). Membrane dynamics of the contractile vacuole complex of Paramecium. J. Protozool. 35,63 -71.
Allen, R. D. and Naitoh, Y. (2002). Osmoregulation and contractile vacuoles of protozoa. Int. Rev. Cytol. 215,352 -394.
Allen, R. D., Ueno, M. S., Pollard, L. W. and Fok, A. K. (1990). Monoclonal antibody study of the decorated spongiome of contractile vacuole complex of Paramecium. J. Cell Sci. 96,469 -475.[Abstract]
Eckert, R. and Naitoh, Y. (1970). Passive
electrical properties of Paramecium and problems of ciliary
coordination. J. Gen. Physiol.
55,467
-483.
Fain, G. L. (1999). Molecular and Cellular Physiology of Neurons. Cambridge, MA, London: Harvard University Press.
Fok, A. K. and Allen, R. D. (1979). Axenic Paramecium caudatum. I. Mass culture and structure. J. Protozool. 26,463 -470.[Medline]
Fok, A. K., Aihara, M. S., Ishida, M., Nolta, K. V., Steck, T.
L. and Allen, R. D. (1995). The pegs on the decorated tubules
of the contractile vacuole complex of Paramecium are proton pumps.
J. Cell Sci. 108,3163
-3170.
Grønlien, H. K., Stock, C., Aihara, M. S., Allen, R. D. and Naitoh, Y. (2002). Relationship between the membrane potential of the contractile vacuole complex and its osmoregulatory activity in Paramecium multimicronucleatum. J. Exp. Biol. 205,3261 -3270.[Medline]
Gustin, M. C. (1992). Mechano-sensitive ion channels in yeast. Mechanism of activation and adaptation. Adv. Compar. Environ. Physiol. 10,19 -38.
Hausmann, K. and Allen, R. D. (1977). Membranes and microtubules of the excretory apparatus of Paramecium caudatum. Eur. J. Cell Biol. 15,303 -320.
Ishida, M., Aihara, M. S., Allen, R. D. and Fok, A. K.
(1993). Osmoregulation in Paramecium: the locus of fluid
segregation in the contractile vacuole complex. J. Cell
Sci. 106,693
-702.
Ishida, M., Fok, A. K., Aihara, M. S. and Allen, R. D.
(1996). Hyperosmotic stress leads to reversible dissociation of
the proton pump-bearing tubules from the contractile vacuole complex in
Paramecium. J. Cell Sci.
109,229
-237.
Iwamoto, M., Allen, R. D. and Naitoh, Y. (2004). Hypo-osmotic or Ca2+-rich external conditions trigger extra contractile vacuole complex generation in Paramecium multimicronucleatum. J. Exp. Biol. 206,467 -473.
McKanna, J. A. (1973). Fine structure of the contractile vacuole pore in Paramecium. J. Protozool. 20,631 -638.[Medline]
Merzendorfer, H., Graf, R., Huss, M., Harvey, W. R. and
Wieczorek, H. (1997). Regulation of proton-translocating
V-ATPases. J. Exp. Biol.
200,225
-235.
Montalvetti, A., Rohloff, P. and Docampo, R.
(2004). A functional aquaporin co-localizes with the vacuolar
proton pyrophosphatase to acidocalcisomes and the contractile vacuole complex
of Trypanosoma cruzi. J. Biol. Chem.
279,38673
-38682.
Morris, C. E. (2001). Mechano-sensitive ion channels in eukaryotic cells. In Cell Physiology Sourcebook, A Molecular Approach, 3rd edn (ed. N. Sperelakis), pp.745 -760. San Diego, San Francisco, New York, Boston, London, Sydney, Tokyo: Academic Press.
Naitoh, Y., Tominaga, T. and Allen, R. D.
(1997a). The contractile vacuole fluid discharge rate is
determined by the vacuole size immediately before the start of discharge in
Paramecium multimicronucleatum. J. Exp. Biol.
200,1737
-1744.
Naitoh, Y., Tominaga, T., Ishida, M., Fok, A. K., Aihara, M. S.
and Allen, R. D. (1997b). How does the contractile vacuole of
Paramecium multimicronucleatum expel fluid? Modelling the expulsion
mechanism. J. Exp. Biol.
200,713
-721.
Nishihara, E., Shinmen, T. and Sonobe, S. (2004). Functional characterization of contractile vacuole isolated from Amoeba proteus. Cell Struct. Funct. 29,85 -90.[CrossRef][Medline]
Ogden, D. (ed.) (1994). Microelectrode Techniques, The Plymouth Workshop Handbook, 2nd edn. Cambridge: Company of Biologists.
Rudy, B. and Iverson, L. E. (ed.) (1992). Methods in Enzymology, vol.207 , Ion Channels. San Diego, New York, Boston, London, Sydney, Tokyo: Academic Press.
Sheetz, M. P. and Dai, J. (1996). Modulation of membrane dynamics and cell motility by membrane tension. Trends Cell Biol. 4,85 -89.[CrossRef]
Sokabe, M., Sachs, F. and Jing, Z. (1991). Quantitative video microscopy of patch clamped membranes stress, strain, capacitance, and stretch channel activation. Biophys. J. 59,722 -728.[Abstract]
Stock, C., Allen, R. D. and Naitoh, Y. (2001).
How external osmolarity affects the activity of the contractile vacuole
complex, the cytosolic osmolarity and the water permeability of the plasma
membrane in Paramecium multimicronucleatum. J. Exp.
Biol. 204,291
-304.
Stock, C., Grønlien, H. K. and Allen, R. D. (2002a). The ionic composition of the contractile vacuole fluid of Paramecium mirrors ion transport across the plasma membrane. Eur. J. Cell Biol. 91,505 -515.[CrossRef]
Stock, C., Grønlien, H. K., Allen, R. D. and Naitoh,
Y. (2002b). Osmoregulation in Paramecium: in situ
ion gradients permit water to cascade through the cytosol to the contractile
vacuole. J. Cell Sci.
115,2339
-2348.
Tani, T., Allen, R. D. and Naitoh, Y. (2000).
Periodic tension development in the membrane of the in vitro
contracile vacuole of Paramecium multimicronucleatum: modification by
bisection, fusion and suction. J. Exp. Biol.
203,239
-251.
Tani, T., Allen, R. D. and Naitoh, Y. (2001).
Cellular membrane that undergo cyclic changes in tension: Direct measurement
of force generated by an in vitro contractile vacuole of
Parmecium multimicronucleatum. J. Cell Sci.
114,785
-795.
Tani, T., Tominaga, T., Allen, R. D. and Naitoh, Y. (2002). Development of periodic tension in the contractile vacuole complex membrane of Paramecium governs its membrane dynamics. Cell Biol. Int. 26,853 -860.[CrossRef][Medline]
Tominaga, T., Allen, R. D. and Naitoh, Y.
(1998a). Electrophysiology of the in situ contractile
vacuole complex of Paramecium reveals its membrane dynamics and
electrogenic site during osmoregulatory activity. J. Exp.
Biol. 201,451
-460.
Tominaga, T., Allen, R. D. and Naitoh, Y.
(1998b). Cyclic changes in the tension of the contractile vacuole
complex membrane control its exocytotic cycle. J. Exp.
Biol. 201,2647
-2658.
Tominaga, T., Naitoh, Y. and Allen, R. D.
(1999). A key function of non-planar membranes and their
associated microtubular ribbons in contractile vacuole membrane dynamics is
revealed by electrophysiologically controlled fixation of Paramecium.
J. Cell Sci. 112,3733
-3745.
Wieczorek, H., Brown, D., Grinstein, S., Ehrenfeld, J. and Harvey, W. R. (1999). Animal plasma membrane energization by proton-motive V-ATPases. BioEssays 21,637 -648.[CrossRef][Medline]
Yoshida, K., Ide, T., Inouye, K., Mizuno, K., Taguchi, T. and Kasai, M. (1997). A voltage- and K+-dependent K+ channel from a membrane fraction enriched in contractile vacuole of Dictyostelium discoideum. Biochim. Biophys. Acta 1325,178 -188.[Medline]