- Email: [email protected]
New method to measure water permeability in emptied-out Xenopus oocytes controlling conditions on both sides of the membrane
J. Biochem. Biophys. Methods 63 (2005) 187 – 200 www.elsevier.com/locate/jbbm
New method to measure water permeability in emptied-out Xenopus oocytes controlling conditions on both sides of the membrane Marcelo Ozu, Ricardo Dorr, Mario Parisi* Laboratorio de Biomembranas, Departamento de Fisiologı´a, Facultad de Medicina, Universidad de Buenos Aires, Paraguay 2155 piso 7, (C1121ABG) Buenos Aires, Repu´blica Argentina Received 7 January 2005; received in revised form 19 April 2005; accepted 27 April 2005
Abstract Membrane water permeability is habitually calculated from volume changes in Xenopus laevis oocytes during external osmotic challenges. Nevertheless, this approach is limited by the uncertainty on the oocyte internal composition. To circumvent this limitation a new experimental set up is introduced where the cell membrane of an emptied-out oocyte was mounted as a diaphragm between two chambers. In its final configuration the oocyte membrane was part of a closed compartment and net water movements induced swelling or shrinking of it. Volume changes were followed by videomicroscopy and digitally recorded. In this manner, water movements could be continuously monitored while controlling chemical composition and hydrostatic pressure on both sides of the membrane. Using this novel experimental approach an increasing hydrostatic pressure gradient was applied to both mature (stage VI) and immature (stage IV) oocytes. The relative maximal volume change tolerated before disruption was similar in both cases (1.26 F 0.07 and 1.27 F 0.03 respectively) and similar to those previously reported under maximal osmotic stress. Nevertheless the osmotic permeability coefficient ( P OSM) in mature oocytes ((1.72 F 0.58) 10 3 cm s 1; n = 6) was significantly lower than in immature oocytes ((5.18 F 0.59) 10 3 cm s 1, n = 5; p b 0.005). D 2005 Elsevier B.V. All rights reserved. Keywords: Xenopus oocyte; Water permeability; Perfusion
* Corresponding author. E-mail address: [email protected] (M. Parisi). 0165-022X/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jbbm.2005.04.007
188
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
1. Introduction Volumetric techniques have been widely employed to measure net water transfers in isolated cells, either indirectly using light scattering techniques [1] or directly measuring cell volume changes [2]. Between the latter, Xenopus laevis oocyte, a system recurrently utilized to express heterologous proteins [3,4], was used to measure water permeability. This is the case when a pore-forming antibiotic as amphotericin B was inserted in the oocyte membrane [5] or before and after expression of aquaporins (AQP) in its membrane [6]. Osmotic permeability variation during an external osmotic challenge is calculated from volume changes in oocytes, and it is typically monitored using video-microscopy. This methodology has been widely employed since it was described [7] and it has been useful to identify water channels in living beings, including plants, mammals, other amphibians and their oocytes [8]. However, the previous described approaches using Xenopus oocytes are limited because of the incertitude on the oocyte internal composition, which does not allow the possibility of permeability measurements controlling the media composition on both sides of the membrane. Furthermore, in general it is assumed that essentially all the water in cells has the same ideal motional and colligative properties as water does in bulk liquid state [9], but oocyte cytoplasmic content is far from being an ideal water reservoir. Some efforts were made to control the compounds of intracellular and extracellular solutions in living oocytes, as in the method described by Dascal et al. [10] for internal perfusion of X. laevis oocytes. Also Taglialatela et al. [11] developed a technique for voltage-clamping Xenopus oocytes in which part of the membrane is isolated by a vaseline gap and the cytoplasmic fluid is exchanged by cutting or permeating the remaining membrane. However, no method was developed to our present knowledge to measure water permeability with the full control of compounds on both sides of the oocyte membrane. Now we are introducing a new set up where the cytoplasmic membrane of an emptiedout oocyte was mounted as a diaphragm between two independent chambers, and where water permeability could be continuously monitored controlling chemical compounds and hydrostatic pressure on both sides of the cell membrane. Our goal is to approximate the experimental conditions employed when measuring water fluxes across artificial lipid membranes doped with water permeable structures [12]. The here presented results, made in mature and immature oocytes, validate the new experimental approach.
2. Materials and methods 2.1. Immature and mature oocytes isolation Adult female X. laevis were kept in tanks containing filtered water and were fed twice a week. For the surgery the frog was anesthetized by hypothermia (40–45 min) induced by placing the animal in ice. A 1 cm incision was made in the abdominal wall, and a lobe of ovary containing oocytes was excised. The piece was rinsed several times with OR-2 solution (for solutions compositions see below) until the solution was clear. The oocytes were separated in small groups (10–20 oocytes per group) with fine forceps and washed several times with OR-2 solution. The oocytes were incubated with 20 ml of OR-2 solution plus 2
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
189
mg/ml of Colagenase A for 1 h at 18 8C with gentle shaking. Cells were washed 5 times with OR-2 solution plus Bovine Serum Albumin (BSA) (0.1 g / 100 ml) and then incubated with 50 ml of a high potassium solution allowing gentle shaking for 1 h. Every 15 min the oocytes were passed through a plastic pipette to help their separation. After that, the oocytes were washed 5 times with Barth’s solution plus BSA (0.1 g / 100 ml). Finally, the oocytes were placed in Barth’s containing gentamicine (1 Ag/ml) overnight at 18 8C. The selected cells for measurements were between 0.8 and 1.3 mm in diameter. According to the classification established by Dumont [13] the oocytes were catalogued as belonging to stages IV and VI. 2.2. Experimental chamber description The original proposed technique implied to glue an oocyte to a support to be placed in a modified Ussing chamber. The experimental set up where the Xenopus oocyte was attached to perform measurements had four acrylic pieces as shown in Fig. 1. The oocyte was glued to the side A1 of the piece A, which had a hole from side to side according to the profile indicated in detail in Fig. 2a. A fine layer of silicone grease was carefully extended on the external face of the hole on surface A1 without obtruding the internal channel that connects both sides of A (Fig. 2b). This silicone layer allowed to accommodate the whole oocyte and to stabilize it on the acrylic piece (Fig. 2c and d). In addition, the silicone layer was used to limit and isolate the surface of the oocyte membrane that will be in contact with the cyanoacrylate glue utilized to fix the oocyte to the acrylic support. The cyanoacrylate adhesive was applied from the inside of the channel (Fig. 2e). The exposed surface of the oocyte was continuously maintained under Barth’s solution, allowing 1 h to complete the solidification of the cyanoacrylate. Subsequently, with the help of a sharp metal needle that was introduced from side A2 (Fig. 2f), the oocyte membrane was broken allowing the exposure and the elimination of the cytoplasm through the channel which was previously filled with Barth’s solution. The exit of the cytoplasm from the core of the oocyte was improved with a micro-perfusion system which
Fig. 1. Experimental chamber parts. Piece A is the acrylic piece where the oocyte is glued. Piece B contains the core of the sealed chamber as a result of being hermetically attached to side A2 of piece A. Piece C represents the extracellular medium of the oocyte. Piece D has a glass window to maintain the focus during video-microscopy acquisition. All parts are in scale. (See next figures for further details).
190
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
Fig. 2. Methodology to glue and empty-out an oocyte. The oocyte was glued to piece A. An open duct allowed breaking a side of the cell membrane to empty-out the oocyte. X: length = 10 mm; Y: thickness = 2 mm; Z: external diameter of the pore = 2 mm; U: internal diameter of the pore: 0.8 mm; V: oocyte bed diameter = 1.5 mm. All parts are in scale. (See text for details).
allowed simultaneously injecting and removing liquid volume to/from the inside of the channel (Fig. 2g) until the partial or full emptying-out of the cell was completed (Fig. 2h). In a subsequent step, piece B of the experimental chamber was hermetically attached to the side A2 of the piece A where the emptied-out oocyte was glued (Fig. 3a). This configured a bsealed chamberQ where the internal pool (0.2 ml) represented the bintracellular mediumQ. The system was then coupled to piece C which pool represented the bextracellular mediumQ (Fig. 3b). This part was opened to the atmosphere and because of its large volume (3.4 ml) it could be considered as an infinite reservoir. A Hamilton 0.10 ml syringe (Microliter #710, Hamilton Company, USA), manipulated with the help of a translation stage, was used to inject or remove Barth’s solution to/from the sealed chamber to initially give to the oocyte a similar diameter to the one observed before manipulation. Even though the pressure in the intracellular side of the chamber was controlled via the Hamilton syringe, the presence of a drain plug was necessary to dissipate undesired hydrostatic pressure, which could damage membrane structure when the chamber was being closed. Piece D was positioned on the bundle formed by pieces A, B and C (Fig. 3c). The chimney allowed the control of the depth of the solution column to manage hydrostatic pressure. As the images were taken from the top, the glass window of piece C contacting the experimental solution ensured a correct optical focus of the oocyte membrane. A top side view of the bundled chamber (without piece D) is presented on Fig. 3d.
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
191
Fig. 3. Assembling and closing of the experimental chamber. The steps to close the experimental chamber (a, b and c) and a top side view (d) of the bundled chamber without piece D are shown. Both the drain duct and the duct where a Hamilton syringe is connected allow controlling the hydrostatic pressure inside the chamber. PA: piece A; PB: piece B; PC: piece C; PD: piece D. All parts are in scale.
2.3. Video-microscopy and image analysis An MC-350 CCD color video camera (O’Rite Technology Co., Ltd) was used to obtain images from a zoom stereo-microscope (SZ40, Olympus Co., Japan). The camera was connected through an USB port to a PC computer. A RGB24 24 bit (True color) capture format without compression was elected to obtain a 640 (horizontal) 480 (vertical) till 15
192
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
frames per second video AVI file. Single images (BMP format) were extracted from the video file using VirtualDub video capture/processing utility for 32-bit Microsoft Windows platforms (http://www.virtualdub.org). Canny Edge Detector (by Nikos Papamarkos, Democritus University of Thrace, Greece, http://ipml.ee.duth.gr/~papamark/), Scion Image (Scion Corporation, http://www.scioncorp.com/) and ImageJ (by Wayne Rasband, Research Services Branch, National Institute of Mental Health, Bethesda, Maryland, USA, http://rsb.info.nih.gov/ij/) image processing and analysis software for Windows were used to determine volume changes from oocytes membrane displacements. To quantify the observed changes the perimeter of the studied oocyte was identified using the specific edge detection software on images acquired at different times. Subsequently image math was done to overlay the images obtained in the different experimental conditions. Area differences were digitally quantified and data were used to calculate volume changes assuming a spherical morphology of the oocyte. 2.4. Estimation of the osmotic permeability coefficient The osmotic permeability coefficient ( P OSM) was estimated from POSM ¼ JW =VW dAdDOsm where J W is the volume of water (in Al) transferred in the unit of time (s) under an initial osmotic gradient DOsm (mol cm 3), V W is the partial molar volume of water (18 ml/mol) and A the observed area of the oocyte (mm2). J W values were obtained from the slope of the volume vs. time function.
3. Reagents Colagenase A and BSA were purchased from Sigma, gentamicine from GIBCO, silicone grease KLS-G4 from Silicon Argentina SRL and cyanoacrylate glue from Akapol S.A.C.I.F.I.A, Argentina.
4. Statistical analysis Data were presented as mean F SEM. P b 0.05 was considered statistically significant (Student’s t-test).
5. Solutions Barth’s solution (in mM): NaCl 88, KCl 1, NaHCO3 2.4, HEPES 10, Ca(NO3)2d 4H2O 0.33, CaCl2d 2H2O 0.41, MgSO4d 7H2O 0.82, pH 7.4. High K+ solution: 100 mM K2HPO4d 3H2O, BSA (0.1 g / 100 ml), pH 6.5. OR-2 solution (in mM): NaCl 82.5, KCl 2, HEPES 5, MgCl2d 6H2O 1, pH 7.5.
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
193
Fig. 4. A partially emptied-out oocyte glued to the experimental chamber. The channel connecting to the internal pool and the inner of a partially emptied-out stage IV oocyte are shown in the picture acquired by videomicroscopy (Bar = 0.1 mm).
Fig. 5. Effect of changing the hydrostatic pressure on the oocyte membrane position. An increase in the bintracellularQ hydrostatic pressure causes the displacement of the membrane towards the chamber representing the extracellular medium. The edges of two acquired images were superimposed to detect membrane displacements. Insert: original images superimposed (Bar = 0.1 mm).
194
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
6. Results Fig. 4 shows a picture of an emptied-out oocyte attached to the acrylic support showing the channel connectivity to the internal pool. Because the oocyte membrane was attached to a sealed and rigid compartment, any net fluid movement across it indicates a proportional swelling or shrinkage of the emptied-out oocyte. In the described set up membrane displacements were observed when a hydrostatic pressure was increased in the bintracellular sideQ of the membrane or when a non-isotonic media was placed in the bextracellularQ compartment. 6.1. Hydrostatic pressure gradients and the mechanical behavior of the oocyte membrane As stated in Methods a slight pressure was applied to the internal side of the emptiedout oocyte to reproduce its initial volume. The lack of change in volume as a function of time was a proof of the absence of leaking in the studied membrane. The micro syringe attached to the system allowed injecting known volumes of solution in the bintracellular chamberQ at very low rates. The rise of the hydrostatic pressure was visualized as a change in the position of the plasma membrane (Fig. 5). Emptied-out
Fig. 6. Effect of rising hydrostatic pressure until the rupture of the membrane. An excessive pressure increase in the bintracellularQ side of partially emptied-out oocytes causes the sudden release of cytoplasmic material from around the seal and the consequent decrease in cell volume. a. The volume of a mature emptied-out oocyte at the beginning of a hydrostatic experiment; b. The maximal volume the oocyte can reach before the rupture; c. The arrow indicates the point of rupture located around the ring of the seal; it can be observed that in this case the oocyte volume has decreased; d. The cytoplasmic material is released by the oocyte from the site of the rupture.
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
195
oocytes were able to increase their internal volume before they exploded. The relative maximal volume change tolerated before rupture when an increasing hydrostatic pressure gradient was applied was similar in mature (1.26 F 0.07) and immature oocytes (1.27 F 0.03). Oocytes usually broke close to the seal between the membrane and the acrylic support (Fig. 6). 6.2. Osmotically induced net water fluxes in mature and immature emptied-out oocytes In these experiments, both sides of the membrane were initially equilibrated in Barth’s solution, and oocyte volume remained constant during several minutes. Subsequently the external bath was replaced by a hypotonic one (40 mOsm) obtained by dilution of the Barth’s solution. The presence of the hypotonic medium in the bextracellular sideQ caused the extension of the oocyte membrane towards the chamber representing the medium where the cell resides (Fig. 7), reflecting water movement across the cellular membrane following the osmotic gradient. Fig. 8 shows the time course of fluid transfer assuming a spherical shape for the emptied-out oocyte. The osmotic permeability coefficient was estimated from the initial slope of the curve as described in Methods. The obtained value when measuring mature oocytes in the previously described condition was
Fig. 7. Effect of changing the osmotic pressure on the membrane position of a stage IV oocyte. When the isotonic medium is replaced by a hypotonic solution, an increase in the bintracellularQ volume displaces immature oocyte membrane (Stage IV oocyte). The edges of two acquired images were superimposed to detect membrane position changes after 120 s. (Bar = 0.1 mm).
196
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
Fig. 8. Relative volume as a function of time assuming a spherical shape for the emptied-out oocyte. This profile corresponds to a typical experiment done with a stage IV oocyte (immature) in the presence of a hyposmotic gradient (120 mOsm). The P OSM value (4.78 10 3 cm s 1) was calculated from the slope obtained with the four first points of the curve in the presence of the hyposmotic gradient.
(1.72 F 0.58) 10 3 cm s 1 (n = 6). Although it is traditionally accepted that native mature oocyte membranes possess low water permeability little is known about permeability values in stage IV oocytes membranes. When immature oocytes were tested in similar conditions the observed mean value (5.18 F 0.59) 10 3 cm s 1 (n = 5) was significantly higher ( p b 0.005) (Fig. 9). Controls made on mature and immature emptiedout oocytes changing the initial solution by an isotonic one did not show any volume changes during experimental time (data not shown).
7. Discussion The main goal of this work was to introduce a new experimental set up for the measurement of water permeability properties of the Xenopus oocyte membrane while controlling media composition on both sides of the membrane. The here-described method gave accurate permeability measurements and it represents an advance over the classical method using Xenopus oocytes. 7.1. Effects of applied hydrostatic gradients Previously published data indicated that oocytes are weakly elastic and that a large part of their resistance to expansion resides in the vitelline envelope [14]. Without the vitelline
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
197
6
POSM (cm.sec-1) x 10-3
5
4
3
2
1
0 Stage IV
Stage VI
Fig. 9. Osmotically measured permeabilities in mature and immature emptied-out oocytes. The bars represent values of P OSM F SEM measured on native mature (stage VI, n = 6) and immature (stage IV, n = 5) oocytes. P b 0.005.
structure, membrane capacitance measurements indicate that the oocyte area is at least 5 times larger than a comparable smooth sphere. A large excess membrane area is consistent with the presence of an extensive membrane folding and the presence of microvilli, i.e. a membrane reserve that allows the cell to tolerate large deformations with no stress in the lipid bilayer. However, a major morphological difference between oocytes on stage V and VI is the reduction in the number and size of the surface microvilli in the latter [15]. We here describe a mean volume increase of 27% before disruption when oocytes swelled by rising hydrostatic pressure from the intracellular side of the membrane. Kelly reported a similar value (24 F 3%) during maximal osmotic swelling [15]. Interestingly Zhang et al. [16] show that the repeated microinjection of solution in an oocyte promotes its inflation and a subsequent sudden release of cytoplasmic material around the injection pipette area (where the oocyte was damaged) and a decrease in cell volume. In our experimental condition, using partially emptied-out oocytes, the observation is coincident, i.e. the oocytes broke next to the site where the membrane was fixed with a violent explosion and the evident sudden release of the oocyte content. We can suppose that the applied treatment decreased the membrane elasticity next to the site where it was fixed. Therefore, the membrane was more rigid in this site and less resistant to stretch, and the rupture occurred in some point next to this ring. We also observed the subsequent decrease in oocyte volume. 7.2. Osmotic gradients and leakage effect After closing the chamber, the oocyte was inflated to a position showing a bnormalQ degree of inflation. To initially position the membrane the internal hydrostatic pressure was increased injecting solution with the help of a Hamilton pipette. This process also allowed us to observe if there was any leakage in the membrane or in the seal between the membrane
198
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
and the acrylic. When leakage was present, we observed the slowly release of the internal solution when the internal pressure increased gradually. In this case, oocytes were discarded. The lack of volume change along time was considered as a proof of absence of leaking in the studied membrane. If any leakage were present around the seal, then we could not observe any response to the osmotic gradient. Control experiments were also made to check a possible water flux across the glue by closing a hole in an acrylic support with cyanoacrylate. In this condition, no water movement was detected (data not shown). In addition, at the beginning of our osmotic experiments, the oocyte membrane was far away of the tolerated maximal distension. This was verified by applying a hydrostatic pressure from the intracellular side using the Hamilton syringe at the end of each osmotic experiment. This measurement also allowed us to calculate the maximal volume the oocytes could reach before disruption. Although the relative maximal volume change tolerated before rupture of the membrane was similar in mature and immature emptied-out oocytes, P OSM was significantly different between them. This would indicate that the response to the osmotic gradient is not related to the stretching capacity of the membrane. 7.3. Water permeability measurements The water permeability of the native mature oocytes is accepted as low [7]. Xenopus oocytes expressing aquaporins give a permeability coefficient ~0.02 cm/s for AQP1 compared with ~0.001 cm/s for controls without AQP1 [17]. Our new permeability measurements did not differ significantly from the values reported by the mentioned methods, indicating that manipulations related to mounting and emptying-out the oocyte did not affect its water permeability properties. This also indicates that the oocyte membrane is the only significant barrier when water is moving into the oocyte. It is conventionally accepted that Xenopus oocytes possess few endogenous water channels. Our results, showing that immature oocytes have higher water permeability than mature ones, suggest that water channels could be lost during arrival to maturity as demonstrated for mammalian oocytes [18]. This must be confirmed in molecular biology experiments (out of the scope of this work). As stated in Introduction, volumetric techniques were initially employed to estimate water permeability in cells [1]. Volumetric methods were also used to measure water permeability across epithelial barriers [19,20] and we previously developed an experimental device designed to improve the measurement of water transfers across epithelia [21]. That set up allowed the first measurements of transepithelial water permeability in aquaporin transfected cells cultures [22]. Undoubtedly, a quantum leap to understand the biophysical basis of water transfer across biological membranes was the development of artificial lipid bilayers. These experimental models allowed not only to measure water permeability of pure lipid membranes but also to analyze the new properties induced by introducing artificial channels into the bilayer [23]. The generally accepted and applied premises on the water pathway in cells arose from these experiments. More recently, Saparov et al. [24] used the bplanar lipid bilayer approachQ for measuring water permeability on AQP1 containing membranes, while controlling the medium composition on both sides of the channel. The
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
199
water permeability was indirectly estimated from the diluting effect of water transfer on ionic concentration measuring electrical parameters. The advantage on using biological and not artificial membranes is evident because of the presence of more genuine conditions at the experimental time. This is the case of using biological vesicles [25] or stopped-flow techniques [26]. We have earlier mentioned the valuable approaches of Dascal et al. [10] and Taglialatela et al. [11] working with mature X. laevis oocytes. The main scope of the new technique was to control for the first time the pressure and media composition on both sides of the Xenopus oocyte membrane to measure water movements, avoiding the uncertainty on the oocyte internal composition and with the advantage of using not artificial membranes to do biophysical measurements. 7.4. Simplified description of the method and its (future) applications We here described a new experimental set up where the cell membrane of an emptiedout X. laevis oocyte is used as a diaphragm between two chambers. This allows doing biophysical measurements of water movements across the biological membrane controlling pressure and media composition on both sides of it. First, the oocyte was fixed to an acrylic support. In a second step the oocyte was perforated across a hole present on the mentioned support and fully or partially emptied-out. The rest of the cytoplasmic membrane remained intact separating two independent and rigid chambers. In its final configuration the oocyte membrane was part of a closed compartment and osmotic net water movements could induce swelling or shrinking of it. Oocyte volume changes were followed by video-microscopy and digitally recorded. In this manner, water permeability could be continuously monitored. The new set up opens a novel way to biophysically measure water permeability of oocytes membranes expressing aquaporins (or other transporters), solutes movements, to study stretch activated channels, the side-specific action of inhibitors or drugs and, in union with electrical measurements, to study the connected water–solute movements. The new methodology can also be used with other amphibian big oocytes, for example oocytes from Bufo arenarum toad. Some of these topics are the objectives of our future work.
Acknowledgements The authors wish to thank Dr. Osvaldo Uchitel, Dr. Carlos D’Attellis and Dr. Vladimir Flores who kindly helped them with their valuable ideas and Dr. Ana Bele´n Elgoyhen who generously gave biological material for experiments. This work was supported by grants from Universidad Nacional de Buenos Aires and Fundacio´n Antorchas.
References [1] Paganelli CV, Solomon AK. The rate of exchange of tritiated water across the human red cell membrane. J Gen Physiol 1957;41:259 – 77.
200
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
[2] Zadunaisky JA, Parisi MN, Montoreano R. Effect of antidiuretic hormone on permeability of single muscle fibres. Nature 1963;200:365 – 6. [3] Gurdon JB, Lane CD, Woodland HR, Mairbaix G. Use of frog eggs and oocytes for the study of messenger RNA and its translation in living cells. Nature 1971;233:177 – 82. [4] Brown DD. A tribute to the Xenopus laevis oocyte and egg. J Biol Chem 2004;279:45291 – 9. [5] Capurro C, Ford P, Ibarra C, Ripoche P, Parisi M. Water permeability properties of the ovarian oocytes from Bufo arenarum and Xenopus laevis: a comparative study. J Membr Biol 1994;138:151 – 7. [6] Preston GM, Carroll TP, Guggino WB, Agre P. Appearance of water channels in Xenopus oocytes expressing red cell CHIP28 protein. Science 1992;256:385 – 7. [7] Zhang R, Logee K, Verkman AS. Expression of mRNA coding for kidney and red cell water channels in Xenopus oocytes. J Biol Chem 1990;265:15375 – 8. [8] Ford P, Amodeo G, Capurro C, Ibarra C, Dorr R, Ripoche P, et al. Progesterone inhibition of water permeability in Bufo arenarum oocytes and urinary bladder. Am J Physiol 1996;270:F880–5. [9] Cameron IL, Kanal KM, Keener CR, Fullerton GD. A mechanistic view of the non-ideal osmotic and motional behavior of intracellular water. Cell Biol Int 1997;21:99 – 113. [10] Dascal N, Chilcott G, Lester HA. Recording of voltage and Ca(2+)-dependent currents in Xenopus oocytes using an intracellular perfusion method. J Neurosci Methods 1991;39:29 – 38. [11] Taglialatela M, Toro L, Stefani E. Novel voltage clamp to record small, fast currents from ion channels expressed in Xenopus oocytes. Biophys J 1992;61:78 – 82. [12] Rosenberg PA, Finkelstein A. Water permeability of gramicidin A-treated lipid bilayer membranes. J Gen Physiol 1978;72:341 – 50. [13] Dumont JN. Oogenesis in Xenopus laevis (Daudin): I. Stages of oocyte development in laboratory maintained animals. J Morphol 1972;136:153 – 80. [14] Kelly SM, Jia YL, Macklem PT. Measurement of elastic properties of Xenopus oocytes. Comp Biochem Physiol A, Physiol 1997;118:607 – 13. [15] Kelly SM, Macklem PT. Direct measurement of intracellular pressure. Am J Physiol 1991;260:C652–7. [16] Zhang Y, Hamill OP. On the discrepancy between whole cell and membrane patch mechanosensitivity in Xenopus oocytes. J Physiol 2000;523:101 – 15. [17] Engel A, Stahlberg H. Aqualglyceroporins: channel proteins with a conserved core, multiple functions, and variable surfaces. Int Rev Cyt 2002;215:75 – 104. [18] Ford P, Merot J, Jawerbaum A, Gimeno MA, Capurro C, Parisi M. Water permeability in rat oocytes at different maturity stages: aquaporin-9 expression. J Membr Biol 2000;176:151 – 8. [19] Bourguet J, Jard S. Un dispositif automatique de mesure et d’enregistrement du flux net d’eau a` travers la peu et la vessie des amphibiens. Biochim Biophys Acta 1964;88:442 – 4. [20] Parisi M, Bourguet J. The single file hypothesis and the water channels induced by antidiuretic hormone. J Membr Biol 1983;71:189 – 93. [21] Dorr RA, Kierbel A, Vera J, Parisi M. A new data-acquisition system for the measurement of the net water flux across epithelia. Comput Methods Programs Biomed 1997;53:9 – 14. [22] Toriano R, Ford P, Rivarola V, Tamarappoo BK, Verkman AS, Parisi M. Reconstitution of a regulated transepithelial water pathway in cells transfected with AQP2 and an AQP1/AQP2 hybrid containing the AQP2-C terminus. J Membr Biol 1998;161:141 – 9. [23] Finkelstein A. Water movement through lipid bilayers, pores, and plasma membranes: theory and reality. USA7 John Wiley and Sons; 1987. [24] Saparov SM, Kozono D, Rothe U, Agre P, Pohl P. Water and ion permeation of aquaporin-1 in planar lipid bilayers. Major differences in structural determinants and stoichiometry. J Biol Chem 2001;276:31515 – 20. [25] Marinelli RA, Tietz PS, Caride AJ, Huang BQ, LaRusso NF. Water transporting properties of hepatocyte basolateral and canalicular plasma membrane domains. J Biol Chem 2003;278:43157 – 62. [26] Suga S, Maeshima M. Water channel activity of radish plasma membrane aquaporins heterologously expressed in yeast and their modification by site-directed mutagenesis. Plant Cell Physiol 2004;45:823 – 30.
New method to measure water permeability in emptied-out Xenopus oocytes controlling conditions on both sides of the membrane Marcelo Ozu, Ricardo Dorr, Mario Parisi* Laboratorio de Biomembranas, Departamento de Fisiologı´a, Facultad de Medicina, Universidad de Buenos Aires, Paraguay 2155 piso 7, (C1121ABG) Buenos Aires, Repu´blica Argentina Received 7 January 2005; received in revised form 19 April 2005; accepted 27 April 2005
Abstract Membrane water permeability is habitually calculated from volume changes in Xenopus laevis oocytes during external osmotic challenges. Nevertheless, this approach is limited by the uncertainty on the oocyte internal composition. To circumvent this limitation a new experimental set up is introduced where the cell membrane of an emptied-out oocyte was mounted as a diaphragm between two chambers. In its final configuration the oocyte membrane was part of a closed compartment and net water movements induced swelling or shrinking of it. Volume changes were followed by videomicroscopy and digitally recorded. In this manner, water movements could be continuously monitored while controlling chemical composition and hydrostatic pressure on both sides of the membrane. Using this novel experimental approach an increasing hydrostatic pressure gradient was applied to both mature (stage VI) and immature (stage IV) oocytes. The relative maximal volume change tolerated before disruption was similar in both cases (1.26 F 0.07 and 1.27 F 0.03 respectively) and similar to those previously reported under maximal osmotic stress. Nevertheless the osmotic permeability coefficient ( P OSM) in mature oocytes ((1.72 F 0.58) 10 3 cm s 1; n = 6) was significantly lower than in immature oocytes ((5.18 F 0.59) 10 3 cm s 1, n = 5; p b 0.005). D 2005 Elsevier B.V. All rights reserved. Keywords: Xenopus oocyte; Water permeability; Perfusion
* Corresponding author. E-mail address: [email protected] (M. Parisi). 0165-022X/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jbbm.2005.04.007
188
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
1. Introduction Volumetric techniques have been widely employed to measure net water transfers in isolated cells, either indirectly using light scattering techniques [1] or directly measuring cell volume changes [2]. Between the latter, Xenopus laevis oocyte, a system recurrently utilized to express heterologous proteins [3,4], was used to measure water permeability. This is the case when a pore-forming antibiotic as amphotericin B was inserted in the oocyte membrane [5] or before and after expression of aquaporins (AQP) in its membrane [6]. Osmotic permeability variation during an external osmotic challenge is calculated from volume changes in oocytes, and it is typically monitored using video-microscopy. This methodology has been widely employed since it was described [7] and it has been useful to identify water channels in living beings, including plants, mammals, other amphibians and their oocytes [8]. However, the previous described approaches using Xenopus oocytes are limited because of the incertitude on the oocyte internal composition, which does not allow the possibility of permeability measurements controlling the media composition on both sides of the membrane. Furthermore, in general it is assumed that essentially all the water in cells has the same ideal motional and colligative properties as water does in bulk liquid state [9], but oocyte cytoplasmic content is far from being an ideal water reservoir. Some efforts were made to control the compounds of intracellular and extracellular solutions in living oocytes, as in the method described by Dascal et al. [10] for internal perfusion of X. laevis oocytes. Also Taglialatela et al. [11] developed a technique for voltage-clamping Xenopus oocytes in which part of the membrane is isolated by a vaseline gap and the cytoplasmic fluid is exchanged by cutting or permeating the remaining membrane. However, no method was developed to our present knowledge to measure water permeability with the full control of compounds on both sides of the oocyte membrane. Now we are introducing a new set up where the cytoplasmic membrane of an emptiedout oocyte was mounted as a diaphragm between two independent chambers, and where water permeability could be continuously monitored controlling chemical compounds and hydrostatic pressure on both sides of the cell membrane. Our goal is to approximate the experimental conditions employed when measuring water fluxes across artificial lipid membranes doped with water permeable structures [12]. The here presented results, made in mature and immature oocytes, validate the new experimental approach.
2. Materials and methods 2.1. Immature and mature oocytes isolation Adult female X. laevis were kept in tanks containing filtered water and were fed twice a week. For the surgery the frog was anesthetized by hypothermia (40–45 min) induced by placing the animal in ice. A 1 cm incision was made in the abdominal wall, and a lobe of ovary containing oocytes was excised. The piece was rinsed several times with OR-2 solution (for solutions compositions see below) until the solution was clear. The oocytes were separated in small groups (10–20 oocytes per group) with fine forceps and washed several times with OR-2 solution. The oocytes were incubated with 20 ml of OR-2 solution plus 2
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
189
mg/ml of Colagenase A for 1 h at 18 8C with gentle shaking. Cells were washed 5 times with OR-2 solution plus Bovine Serum Albumin (BSA) (0.1 g / 100 ml) and then incubated with 50 ml of a high potassium solution allowing gentle shaking for 1 h. Every 15 min the oocytes were passed through a plastic pipette to help their separation. After that, the oocytes were washed 5 times with Barth’s solution plus BSA (0.1 g / 100 ml). Finally, the oocytes were placed in Barth’s containing gentamicine (1 Ag/ml) overnight at 18 8C. The selected cells for measurements were between 0.8 and 1.3 mm in diameter. According to the classification established by Dumont [13] the oocytes were catalogued as belonging to stages IV and VI. 2.2. Experimental chamber description The original proposed technique implied to glue an oocyte to a support to be placed in a modified Ussing chamber. The experimental set up where the Xenopus oocyte was attached to perform measurements had four acrylic pieces as shown in Fig. 1. The oocyte was glued to the side A1 of the piece A, which had a hole from side to side according to the profile indicated in detail in Fig. 2a. A fine layer of silicone grease was carefully extended on the external face of the hole on surface A1 without obtruding the internal channel that connects both sides of A (Fig. 2b). This silicone layer allowed to accommodate the whole oocyte and to stabilize it on the acrylic piece (Fig. 2c and d). In addition, the silicone layer was used to limit and isolate the surface of the oocyte membrane that will be in contact with the cyanoacrylate glue utilized to fix the oocyte to the acrylic support. The cyanoacrylate adhesive was applied from the inside of the channel (Fig. 2e). The exposed surface of the oocyte was continuously maintained under Barth’s solution, allowing 1 h to complete the solidification of the cyanoacrylate. Subsequently, with the help of a sharp metal needle that was introduced from side A2 (Fig. 2f), the oocyte membrane was broken allowing the exposure and the elimination of the cytoplasm through the channel which was previously filled with Barth’s solution. The exit of the cytoplasm from the core of the oocyte was improved with a micro-perfusion system which
Fig. 1. Experimental chamber parts. Piece A is the acrylic piece where the oocyte is glued. Piece B contains the core of the sealed chamber as a result of being hermetically attached to side A2 of piece A. Piece C represents the extracellular medium of the oocyte. Piece D has a glass window to maintain the focus during video-microscopy acquisition. All parts are in scale. (See next figures for further details).
190
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
Fig. 2. Methodology to glue and empty-out an oocyte. The oocyte was glued to piece A. An open duct allowed breaking a side of the cell membrane to empty-out the oocyte. X: length = 10 mm; Y: thickness = 2 mm; Z: external diameter of the pore = 2 mm; U: internal diameter of the pore: 0.8 mm; V: oocyte bed diameter = 1.5 mm. All parts are in scale. (See text for details).
allowed simultaneously injecting and removing liquid volume to/from the inside of the channel (Fig. 2g) until the partial or full emptying-out of the cell was completed (Fig. 2h). In a subsequent step, piece B of the experimental chamber was hermetically attached to the side A2 of the piece A where the emptied-out oocyte was glued (Fig. 3a). This configured a bsealed chamberQ where the internal pool (0.2 ml) represented the bintracellular mediumQ. The system was then coupled to piece C which pool represented the bextracellular mediumQ (Fig. 3b). This part was opened to the atmosphere and because of its large volume (3.4 ml) it could be considered as an infinite reservoir. A Hamilton 0.10 ml syringe (Microliter #710, Hamilton Company, USA), manipulated with the help of a translation stage, was used to inject or remove Barth’s solution to/from the sealed chamber to initially give to the oocyte a similar diameter to the one observed before manipulation. Even though the pressure in the intracellular side of the chamber was controlled via the Hamilton syringe, the presence of a drain plug was necessary to dissipate undesired hydrostatic pressure, which could damage membrane structure when the chamber was being closed. Piece D was positioned on the bundle formed by pieces A, B and C (Fig. 3c). The chimney allowed the control of the depth of the solution column to manage hydrostatic pressure. As the images were taken from the top, the glass window of piece C contacting the experimental solution ensured a correct optical focus of the oocyte membrane. A top side view of the bundled chamber (without piece D) is presented on Fig. 3d.
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
191
Fig. 3. Assembling and closing of the experimental chamber. The steps to close the experimental chamber (a, b and c) and a top side view (d) of the bundled chamber without piece D are shown. Both the drain duct and the duct where a Hamilton syringe is connected allow controlling the hydrostatic pressure inside the chamber. PA: piece A; PB: piece B; PC: piece C; PD: piece D. All parts are in scale.
2.3. Video-microscopy and image analysis An MC-350 CCD color video camera (O’Rite Technology Co., Ltd) was used to obtain images from a zoom stereo-microscope (SZ40, Olympus Co., Japan). The camera was connected through an USB port to a PC computer. A RGB24 24 bit (True color) capture format without compression was elected to obtain a 640 (horizontal) 480 (vertical) till 15
192
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
frames per second video AVI file. Single images (BMP format) were extracted from the video file using VirtualDub video capture/processing utility for 32-bit Microsoft Windows platforms (http://www.virtualdub.org). Canny Edge Detector (by Nikos Papamarkos, Democritus University of Thrace, Greece, http://ipml.ee.duth.gr/~papamark/), Scion Image (Scion Corporation, http://www.scioncorp.com/) and ImageJ (by Wayne Rasband, Research Services Branch, National Institute of Mental Health, Bethesda, Maryland, USA, http://rsb.info.nih.gov/ij/) image processing and analysis software for Windows were used to determine volume changes from oocytes membrane displacements. To quantify the observed changes the perimeter of the studied oocyte was identified using the specific edge detection software on images acquired at different times. Subsequently image math was done to overlay the images obtained in the different experimental conditions. Area differences were digitally quantified and data were used to calculate volume changes assuming a spherical morphology of the oocyte. 2.4. Estimation of the osmotic permeability coefficient The osmotic permeability coefficient ( P OSM) was estimated from POSM ¼ JW =VW dAdDOsm where J W is the volume of water (in Al) transferred in the unit of time (s) under an initial osmotic gradient DOsm (mol cm 3), V W is the partial molar volume of water (18 ml/mol) and A the observed area of the oocyte (mm2). J W values were obtained from the slope of the volume vs. time function.
3. Reagents Colagenase A and BSA were purchased from Sigma, gentamicine from GIBCO, silicone grease KLS-G4 from Silicon Argentina SRL and cyanoacrylate glue from Akapol S.A.C.I.F.I.A, Argentina.
4. Statistical analysis Data were presented as mean F SEM. P b 0.05 was considered statistically significant (Student’s t-test).
5. Solutions Barth’s solution (in mM): NaCl 88, KCl 1, NaHCO3 2.4, HEPES 10, Ca(NO3)2d 4H2O 0.33, CaCl2d 2H2O 0.41, MgSO4d 7H2O 0.82, pH 7.4. High K+ solution: 100 mM K2HPO4d 3H2O, BSA (0.1 g / 100 ml), pH 6.5. OR-2 solution (in mM): NaCl 82.5, KCl 2, HEPES 5, MgCl2d 6H2O 1, pH 7.5.
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
193
Fig. 4. A partially emptied-out oocyte glued to the experimental chamber. The channel connecting to the internal pool and the inner of a partially emptied-out stage IV oocyte are shown in the picture acquired by videomicroscopy (Bar = 0.1 mm).
Fig. 5. Effect of changing the hydrostatic pressure on the oocyte membrane position. An increase in the bintracellularQ hydrostatic pressure causes the displacement of the membrane towards the chamber representing the extracellular medium. The edges of two acquired images were superimposed to detect membrane displacements. Insert: original images superimposed (Bar = 0.1 mm).
194
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
6. Results Fig. 4 shows a picture of an emptied-out oocyte attached to the acrylic support showing the channel connectivity to the internal pool. Because the oocyte membrane was attached to a sealed and rigid compartment, any net fluid movement across it indicates a proportional swelling or shrinkage of the emptied-out oocyte. In the described set up membrane displacements were observed when a hydrostatic pressure was increased in the bintracellular sideQ of the membrane or when a non-isotonic media was placed in the bextracellularQ compartment. 6.1. Hydrostatic pressure gradients and the mechanical behavior of the oocyte membrane As stated in Methods a slight pressure was applied to the internal side of the emptiedout oocyte to reproduce its initial volume. The lack of change in volume as a function of time was a proof of the absence of leaking in the studied membrane. The micro syringe attached to the system allowed injecting known volumes of solution in the bintracellular chamberQ at very low rates. The rise of the hydrostatic pressure was visualized as a change in the position of the plasma membrane (Fig. 5). Emptied-out
Fig. 6. Effect of rising hydrostatic pressure until the rupture of the membrane. An excessive pressure increase in the bintracellularQ side of partially emptied-out oocytes causes the sudden release of cytoplasmic material from around the seal and the consequent decrease in cell volume. a. The volume of a mature emptied-out oocyte at the beginning of a hydrostatic experiment; b. The maximal volume the oocyte can reach before the rupture; c. The arrow indicates the point of rupture located around the ring of the seal; it can be observed that in this case the oocyte volume has decreased; d. The cytoplasmic material is released by the oocyte from the site of the rupture.
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
195
oocytes were able to increase their internal volume before they exploded. The relative maximal volume change tolerated before rupture when an increasing hydrostatic pressure gradient was applied was similar in mature (1.26 F 0.07) and immature oocytes (1.27 F 0.03). Oocytes usually broke close to the seal between the membrane and the acrylic support (Fig. 6). 6.2. Osmotically induced net water fluxes in mature and immature emptied-out oocytes In these experiments, both sides of the membrane were initially equilibrated in Barth’s solution, and oocyte volume remained constant during several minutes. Subsequently the external bath was replaced by a hypotonic one (40 mOsm) obtained by dilution of the Barth’s solution. The presence of the hypotonic medium in the bextracellular sideQ caused the extension of the oocyte membrane towards the chamber representing the medium where the cell resides (Fig. 7), reflecting water movement across the cellular membrane following the osmotic gradient. Fig. 8 shows the time course of fluid transfer assuming a spherical shape for the emptied-out oocyte. The osmotic permeability coefficient was estimated from the initial slope of the curve as described in Methods. The obtained value when measuring mature oocytes in the previously described condition was
Fig. 7. Effect of changing the osmotic pressure on the membrane position of a stage IV oocyte. When the isotonic medium is replaced by a hypotonic solution, an increase in the bintracellularQ volume displaces immature oocyte membrane (Stage IV oocyte). The edges of two acquired images were superimposed to detect membrane position changes after 120 s. (Bar = 0.1 mm).
196
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
Fig. 8. Relative volume as a function of time assuming a spherical shape for the emptied-out oocyte. This profile corresponds to a typical experiment done with a stage IV oocyte (immature) in the presence of a hyposmotic gradient (120 mOsm). The P OSM value (4.78 10 3 cm s 1) was calculated from the slope obtained with the four first points of the curve in the presence of the hyposmotic gradient.
(1.72 F 0.58) 10 3 cm s 1 (n = 6). Although it is traditionally accepted that native mature oocyte membranes possess low water permeability little is known about permeability values in stage IV oocytes membranes. When immature oocytes were tested in similar conditions the observed mean value (5.18 F 0.59) 10 3 cm s 1 (n = 5) was significantly higher ( p b 0.005) (Fig. 9). Controls made on mature and immature emptiedout oocytes changing the initial solution by an isotonic one did not show any volume changes during experimental time (data not shown).
7. Discussion The main goal of this work was to introduce a new experimental set up for the measurement of water permeability properties of the Xenopus oocyte membrane while controlling media composition on both sides of the membrane. The here-described method gave accurate permeability measurements and it represents an advance over the classical method using Xenopus oocytes. 7.1. Effects of applied hydrostatic gradients Previously published data indicated that oocytes are weakly elastic and that a large part of their resistance to expansion resides in the vitelline envelope [14]. Without the vitelline
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
197
6
POSM (cm.sec-1) x 10-3
5
4
3
2
1
0 Stage IV
Stage VI
Fig. 9. Osmotically measured permeabilities in mature and immature emptied-out oocytes. The bars represent values of P OSM F SEM measured on native mature (stage VI, n = 6) and immature (stage IV, n = 5) oocytes. P b 0.005.
structure, membrane capacitance measurements indicate that the oocyte area is at least 5 times larger than a comparable smooth sphere. A large excess membrane area is consistent with the presence of an extensive membrane folding and the presence of microvilli, i.e. a membrane reserve that allows the cell to tolerate large deformations with no stress in the lipid bilayer. However, a major morphological difference between oocytes on stage V and VI is the reduction in the number and size of the surface microvilli in the latter [15]. We here describe a mean volume increase of 27% before disruption when oocytes swelled by rising hydrostatic pressure from the intracellular side of the membrane. Kelly reported a similar value (24 F 3%) during maximal osmotic swelling [15]. Interestingly Zhang et al. [16] show that the repeated microinjection of solution in an oocyte promotes its inflation and a subsequent sudden release of cytoplasmic material around the injection pipette area (where the oocyte was damaged) and a decrease in cell volume. In our experimental condition, using partially emptied-out oocytes, the observation is coincident, i.e. the oocytes broke next to the site where the membrane was fixed with a violent explosion and the evident sudden release of the oocyte content. We can suppose that the applied treatment decreased the membrane elasticity next to the site where it was fixed. Therefore, the membrane was more rigid in this site and less resistant to stretch, and the rupture occurred in some point next to this ring. We also observed the subsequent decrease in oocyte volume. 7.2. Osmotic gradients and leakage effect After closing the chamber, the oocyte was inflated to a position showing a bnormalQ degree of inflation. To initially position the membrane the internal hydrostatic pressure was increased injecting solution with the help of a Hamilton pipette. This process also allowed us to observe if there was any leakage in the membrane or in the seal between the membrane
198
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
and the acrylic. When leakage was present, we observed the slowly release of the internal solution when the internal pressure increased gradually. In this case, oocytes were discarded. The lack of volume change along time was considered as a proof of absence of leaking in the studied membrane. If any leakage were present around the seal, then we could not observe any response to the osmotic gradient. Control experiments were also made to check a possible water flux across the glue by closing a hole in an acrylic support with cyanoacrylate. In this condition, no water movement was detected (data not shown). In addition, at the beginning of our osmotic experiments, the oocyte membrane was far away of the tolerated maximal distension. This was verified by applying a hydrostatic pressure from the intracellular side using the Hamilton syringe at the end of each osmotic experiment. This measurement also allowed us to calculate the maximal volume the oocytes could reach before disruption. Although the relative maximal volume change tolerated before rupture of the membrane was similar in mature and immature emptied-out oocytes, P OSM was significantly different between them. This would indicate that the response to the osmotic gradient is not related to the stretching capacity of the membrane. 7.3. Water permeability measurements The water permeability of the native mature oocytes is accepted as low [7]. Xenopus oocytes expressing aquaporins give a permeability coefficient ~0.02 cm/s for AQP1 compared with ~0.001 cm/s for controls without AQP1 [17]. Our new permeability measurements did not differ significantly from the values reported by the mentioned methods, indicating that manipulations related to mounting and emptying-out the oocyte did not affect its water permeability properties. This also indicates that the oocyte membrane is the only significant barrier when water is moving into the oocyte. It is conventionally accepted that Xenopus oocytes possess few endogenous water channels. Our results, showing that immature oocytes have higher water permeability than mature ones, suggest that water channels could be lost during arrival to maturity as demonstrated for mammalian oocytes [18]. This must be confirmed in molecular biology experiments (out of the scope of this work). As stated in Introduction, volumetric techniques were initially employed to estimate water permeability in cells [1]. Volumetric methods were also used to measure water permeability across epithelial barriers [19,20] and we previously developed an experimental device designed to improve the measurement of water transfers across epithelia [21]. That set up allowed the first measurements of transepithelial water permeability in aquaporin transfected cells cultures [22]. Undoubtedly, a quantum leap to understand the biophysical basis of water transfer across biological membranes was the development of artificial lipid bilayers. These experimental models allowed not only to measure water permeability of pure lipid membranes but also to analyze the new properties induced by introducing artificial channels into the bilayer [23]. The generally accepted and applied premises on the water pathway in cells arose from these experiments. More recently, Saparov et al. [24] used the bplanar lipid bilayer approachQ for measuring water permeability on AQP1 containing membranes, while controlling the medium composition on both sides of the channel. The
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
199
water permeability was indirectly estimated from the diluting effect of water transfer on ionic concentration measuring electrical parameters. The advantage on using biological and not artificial membranes is evident because of the presence of more genuine conditions at the experimental time. This is the case of using biological vesicles [25] or stopped-flow techniques [26]. We have earlier mentioned the valuable approaches of Dascal et al. [10] and Taglialatela et al. [11] working with mature X. laevis oocytes. The main scope of the new technique was to control for the first time the pressure and media composition on both sides of the Xenopus oocyte membrane to measure water movements, avoiding the uncertainty on the oocyte internal composition and with the advantage of using not artificial membranes to do biophysical measurements. 7.4. Simplified description of the method and its (future) applications We here described a new experimental set up where the cell membrane of an emptiedout X. laevis oocyte is used as a diaphragm between two chambers. This allows doing biophysical measurements of water movements across the biological membrane controlling pressure and media composition on both sides of it. First, the oocyte was fixed to an acrylic support. In a second step the oocyte was perforated across a hole present on the mentioned support and fully or partially emptied-out. The rest of the cytoplasmic membrane remained intact separating two independent and rigid chambers. In its final configuration the oocyte membrane was part of a closed compartment and osmotic net water movements could induce swelling or shrinking of it. Oocyte volume changes were followed by video-microscopy and digitally recorded. In this manner, water permeability could be continuously monitored. The new set up opens a novel way to biophysically measure water permeability of oocytes membranes expressing aquaporins (or other transporters), solutes movements, to study stretch activated channels, the side-specific action of inhibitors or drugs and, in union with electrical measurements, to study the connected water–solute movements. The new methodology can also be used with other amphibian big oocytes, for example oocytes from Bufo arenarum toad. Some of these topics are the objectives of our future work.
Acknowledgements The authors wish to thank Dr. Osvaldo Uchitel, Dr. Carlos D’Attellis and Dr. Vladimir Flores who kindly helped them with their valuable ideas and Dr. Ana Bele´n Elgoyhen who generously gave biological material for experiments. This work was supported by grants from Universidad Nacional de Buenos Aires and Fundacio´n Antorchas.
References [1] Paganelli CV, Solomon AK. The rate of exchange of tritiated water across the human red cell membrane. J Gen Physiol 1957;41:259 – 77.
200
M. Ozu et al. / J. Biochem. Biophys. Methods 63 (2005) 187–200
[2] Zadunaisky JA, Parisi MN, Montoreano R. Effect of antidiuretic hormone on permeability of single muscle fibres. Nature 1963;200:365 – 6. [3] Gurdon JB, Lane CD, Woodland HR, Mairbaix G. Use of frog eggs and oocytes for the study of messenger RNA and its translation in living cells. Nature 1971;233:177 – 82. [4] Brown DD. A tribute to the Xenopus laevis oocyte and egg. J Biol Chem 2004;279:45291 – 9. [5] Capurro C, Ford P, Ibarra C, Ripoche P, Parisi M. Water permeability properties of the ovarian oocytes from Bufo arenarum and Xenopus laevis: a comparative study. J Membr Biol 1994;138:151 – 7. [6] Preston GM, Carroll TP, Guggino WB, Agre P. Appearance of water channels in Xenopus oocytes expressing red cell CHIP28 protein. Science 1992;256:385 – 7. [7] Zhang R, Logee K, Verkman AS. Expression of mRNA coding for kidney and red cell water channels in Xenopus oocytes. J Biol Chem 1990;265:15375 – 8. [8] Ford P, Amodeo G, Capurro C, Ibarra C, Dorr R, Ripoche P, et al. Progesterone inhibition of water permeability in Bufo arenarum oocytes and urinary bladder. Am J Physiol 1996;270:F880–5. [9] Cameron IL, Kanal KM, Keener CR, Fullerton GD. A mechanistic view of the non-ideal osmotic and motional behavior of intracellular water. Cell Biol Int 1997;21:99 – 113. [10] Dascal N, Chilcott G, Lester HA. Recording of voltage and Ca(2+)-dependent currents in Xenopus oocytes using an intracellular perfusion method. J Neurosci Methods 1991;39:29 – 38. [11] Taglialatela M, Toro L, Stefani E. Novel voltage clamp to record small, fast currents from ion channels expressed in Xenopus oocytes. Biophys J 1992;61:78 – 82. [12] Rosenberg PA, Finkelstein A. Water permeability of gramicidin A-treated lipid bilayer membranes. J Gen Physiol 1978;72:341 – 50. [13] Dumont JN. Oogenesis in Xenopus laevis (Daudin): I. Stages of oocyte development in laboratory maintained animals. J Morphol 1972;136:153 – 80. [14] Kelly SM, Jia YL, Macklem PT. Measurement of elastic properties of Xenopus oocytes. Comp Biochem Physiol A, Physiol 1997;118:607 – 13. [15] Kelly SM, Macklem PT. Direct measurement of intracellular pressure. Am J Physiol 1991;260:C652–7. [16] Zhang Y, Hamill OP. On the discrepancy between whole cell and membrane patch mechanosensitivity in Xenopus oocytes. J Physiol 2000;523:101 – 15. [17] Engel A, Stahlberg H. Aqualglyceroporins: channel proteins with a conserved core, multiple functions, and variable surfaces. Int Rev Cyt 2002;215:75 – 104. [18] Ford P, Merot J, Jawerbaum A, Gimeno MA, Capurro C, Parisi M. Water permeability in rat oocytes at different maturity stages: aquaporin-9 expression. J Membr Biol 2000;176:151 – 8. [19] Bourguet J, Jard S. Un dispositif automatique de mesure et d’enregistrement du flux net d’eau a` travers la peu et la vessie des amphibiens. Biochim Biophys Acta 1964;88:442 – 4. [20] Parisi M, Bourguet J. The single file hypothesis and the water channels induced by antidiuretic hormone. J Membr Biol 1983;71:189 – 93. [21] Dorr RA, Kierbel A, Vera J, Parisi M. A new data-acquisition system for the measurement of the net water flux across epithelia. Comput Methods Programs Biomed 1997;53:9 – 14. [22] Toriano R, Ford P, Rivarola V, Tamarappoo BK, Verkman AS, Parisi M. Reconstitution of a regulated transepithelial water pathway in cells transfected with AQP2 and an AQP1/AQP2 hybrid containing the AQP2-C terminus. J Membr Biol 1998;161:141 – 9. [23] Finkelstein A. Water movement through lipid bilayers, pores, and plasma membranes: theory and reality. USA7 John Wiley and Sons; 1987. [24] Saparov SM, Kozono D, Rothe U, Agre P, Pohl P. Water and ion permeation of aquaporin-1 in planar lipid bilayers. Major differences in structural determinants and stoichiometry. J Biol Chem 2001;276:31515 – 20. [25] Marinelli RA, Tietz PS, Caride AJ, Huang BQ, LaRusso NF. Water transporting properties of hepatocyte basolateral and canalicular plasma membrane domains. J Biol Chem 2003;278:43157 – 62. [26] Suga S, Maeshima M. Water channel activity of radish plasma membrane aquaporins heterologously expressed in yeast and their modification by site-directed mutagenesis. Plant Cell Physiol 2004;45:823 – 30.