Biochimica et Biophysica Acta 1848 (2015) 2337–2343
Contents lists available at ScienceDirect
Biochimica et Biophysica Acta
j ourna l homepage: www.e lsev ie r .com/ locate /bbamem
Role of cytoskeleton network in anisosmotic volume changes of intactand permeabilized A549 cells
Alexandra Platonova a,b, Olga Ponomarchuk a,b, Francis Boudreault a, Leonid V. Kapilevich c,Georgy V. Maksimov b, Ryszard Grygorczyk a,d,⁎, Sergei N. Orlov a,b,c,⁎⁎a Research Centre, Centre hospitalier de l'Université de Montréal (CRCHUM), Montreal, QC, Canadab Faculty of Biology, M.V. Lomonosov Moscow State University, Moscow, Russiac Tomsk State University, Tomsk, Russiad Department of Medicine, University of Montreal, Montreal, QC, Canada
⁎ Correspondence to: S.N. Orlov, Laboratory of BiologicaM.V. Lomonosov Moscow State University, Vorob'evy Gor⁎⁎ Correspondence to: R. Grygorczyk, CRCHUM— Tour V(Québec) H2X 0A9, Canada.
Recently we found that cytoplasm of permeabilized mammalian cells behaves as a hydrogel displaying intrinsicosmosensitivity. This study examined the role of microfilaments and microtubules in the regulation of hydrogelosmosensitivity, volume-sensitive ion transporters, and their contribution to volume modulation of intact cells.We found that intact and digitonin-permeabilized A549 cells displayed similar rate of shrinkage triggered byhyperosmotic medium. It was significantly slowed-down in both cell preparations after disruption of actinmicrofilaments by cytochalasin B, suggesting that rapidwater release by intact cytoplasmic hydrogel contributesto hyperosmotic shrinkage. In hyposmotic swelling experiments, disruption ofmicrotubules by vinblastine atten-uated the maximal amplitude of swelling in intact cells and completely abolished it in permeabilized cells. Theswelling of intact cells also triggered ~10-fold elevation of furosemide-resistant 86Rb+ (K+) permeability andthe regulatory volume decrease (RVD), both ofwhichwere abolished by Ba2+. Interestingly, RVD and K+ perme-ability remained unaffected in cytocholasin/vinblastine treated cells demonstrating that cytoskeleton disruptionhas no direct impact on Ba2+-sensitive K+-channels involved in RVD. Our results show, for the first time, that thecytoskeleton network contributes directly to passive cell volume adjustments in anisosmotic media via themodulation of the water retained by the cytoplasmic hydrogel.
The architecture of nucleated cells is supported by a three-dimensional (3D) cytoskeleton formed by actin-containing microfila-ments, intermediate filaments, and microtubules [1]. Several researchteams proposed that cell volume alterations result in cytoskeleton re-arrangement that, in turn, affects the conformation and functional activ-ity of diverse cytoskeleton-associated ion transporters and other pro-teins involved in regulatory volume decrease and increase (RVD andRVI, respectively) [2–5]. This hypothesis was supported by the observa-tions listed below. First, in several types of mammalian cells, shrinkageled to ligand-independent activation of cytoskeleton-associated cyto-kine receptors [6–8]. Second, the addition of F-actin-disrupting (cyto-chalasin B, latrunculin) and microtubule depolymerization agents
lMembranes, Faculty of Biology,y 1/12, Moscow 119899, Russia.iger, 900, rue St-Denis,Montréal
. Grygorczyk),
(colchicine, vinblastine) affected the activity of cell volume-sensitiveproteins, including Rho family GTPases, phospholipase A2, and myosinlight chain kinase (for review, refer to [9]). Third, in an overwhelmingnumber of cells studied so far, cell swelling and shrinkage were linkedwith cellular F-actin decrease and increase, respectively [10–12]. Fourth,in fibroblasts and retinal pigment epithelial cells, hypotonic stress wasassociated with increased microtubule polymerization [13,14]. Fifth, inamphibian gall bladder cells [15], rabbit proximal tubules [16], Ehrilchascites tumor cells [17], PC12 andmammalian tumor cells [18,19], cyto-chalasin B affected RVD and/or RVI.
It should be underlined that the cell volumemeasurement methodsemployed in these studies had several limitations. For example, lightscattering, refractometry and Coulter electronic size techniques are ap-plicable only to suspended cells and work best with cells of simple, e.g.spherical, shape. Measurement of intracellular water volume withmembrane-permeable, non-metabolized compounds, such as [14C]-urea and methyl-D-[14C]-glucose, needs prolonged incubation for thesteady-state distribution of radioisotopes as well as extensive washing,which complicates precise kinetics studies. Reconstruction of single cellimages by laser interference or holographic microscopy cannot beemployed for volume measurement because the average refractive
2338 A. Platonova et al. / Biochimica et Biophysica Acta 1848 (2015) 2337–2343
index of the cytoplasm is affected by cell shrinkage. Temporal resolutionof confocal laser scanning microscopy, scanning ion conductance mi-croscopy and atomic force microscopy, the former techniques beinglimited and do not permit monitoring of rapid volume changes, wereoften fully completed within 2–3 min. Further constraints include pho-todynamic damage to cells during longer exposure to laser light. Forcomprehensive review, see [20,21].
Keeping the above-listed limitations in mind, we developed thedual-image surface reconstruction (DISUR) technique based on phase-contrast digital video microscopy, which allowed us to simultaneouslymeasure cell height, total surface area and volume of unperturbed,substrate-attached cells of relatively regular shape with temporal reso-lution of ~100 ms [22]. Taking this methodological approach, we dem-onstrated that plasma membrane permeabilization with digitonin oramphotericin B leads to dissipation of Donnan equilibrium and swellingof nucleated mammalian cells but does not affect their integrity.Unexpectedly, we found that permeabilized cells swell and shrink inhypo- and hypertonic solutions, respectively. Remarkably, osmolality-induced volume changes were larger than those observed with intactcells [23] suggesting an involvement of osmosensing properties of cyto-plasmic hydrogel in cell volume regulation.
In thepresent study,we compared the actions of cytoskeleton-activecompounds on volume perturbations in intact and permeabilized A549cells as well as on K+ (86Rb) fluxes playing a key role in RVD [24]. Ourresults demonstrate, for thefirst time, that disruption of the cytoskeletalnetwork by cytochalasin B and vinblastine affects the osmosensingfunction of cytoplasmic hydrogel but did not abolish RVD andswelling-induced K+ fluxes.
2. Methods
Human lung carcinoma A549 cells were grown in Dulbecco's modi-fied Eagle's medium supplemented with 10% fetal bovine serum,2 mM L-glutamine, 50 U/ml penicillin and 100 μg/ml streptomycin.The cytoskeleton was visualized by double staining of actin filamentsand microtubules. Briefly, cells seeded on coverslips were washedwith phosphate-buffered saline (PBS: 150 mM NaCl, 5 mM Na-phosphate buffer, pH 7.4), fixed with 4% formaldehyde in PBS for10 min, washed with PBS, permeabilized for 1 min with PBS containing0.5% triton X-100, washed with PBS and incubated for 1 h with mono-clonal anti-alpha tubulin (DM1A) antibody (Sigma-Aldrich, St. Louis,MO) in PBS containing 1% bovine serum albumin (BSA). The cellswere then washed with PBS, and microtubules were stained withFITC-conjugated anti-mouse-IgG antibody in PBS containing 1% BSA.To visualize microfilaments and the nucleus, the cells were stainedwith Alexa Fluo488 phalloidin and 4′,6-diamidino-2-phenylindole, re-spectively, as described in detail elsewhere [25–27]. Actin filamentsand microtubules were disrupted by 20–30-min treatment with20 μM cytochalasin B and 10 μM vinblastine, respectively.
Cell volume was measured in substrate-attached cells with an im-proved version of the DISUR technique described in detail previously[23]. Cells seeded on coverslips were mounted in a custom-madeflow-through imaging chamber and perfused during ~25 min at 1–2ml/minwith HEPES-buffered isotonic medium B (B-iso) at 37 °C. Isos-motic medium (B-iso, ~310 mOsm) was composed of 135 mM NaCl,5 mM KCl, 1.2 mM MgSO4, 1.3 mM CaCl2, 1.2 mM Na2HPO4, 10 mM D-glucose, 10 mM 2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonicacid (HEPES); pH 7.4, adjusted with NaOH. To trigger swelling, cellswere incubated in hyposmotic medium B (B-hypo, ~180 mOsm) inwhich [NaCl] was decreased to 70 mM. Cell shrinkage was caused bytheir perfusion with hyperosmotic medium B (B-hyper, 510 mOsm)contained 200 mM mannitol.
To estimate the relative impact of membrane-bound transportersand cytoplasmic biogel in volume adjustment, A549 cells were perme-abilized by ~2-min exposure to 5 μg/ml (~4 μM) of digitonin (Sigma-Aldrich) in intracellular-like solution (ILS, ~304 mOsm) containing
10 mM NaCl, 110 mM KCl, 5 mM MgCl2, 1 mM Na-ATP, 1 mM EGTA,11 mM dithiothreitol, 25 imidazole and 10 mM 2-[[1,3-dihydroxy-2-(hydroxymethyl)propan-2-yl]amino]ethanesulfonic acid (pH 7.1),followed by washing with digitonin-free ILS. ILS osmolality was attenu-ated by decreasing KCl concentration to 70mM(ILS-hypo, ~224mOsm)and increased by adding 200 mM mannitol (ILS-hyper, ~504 mOsm).The values of medium osmolality were calculated as for ideal solutions.As verified by trypan blue staining, ~90% of A549 cells were perme-abilized by such treatment. In separate experiments, by includingtrypan blue at different time points after digitonin treatment, wefound that ~70% of cells remained permeabilized for at least 30 min,long enough to study the properties of such permeabilized cells.Because some cells re-sealed their plasma membranes during the 30–40 min period, trypan blue staining was always undertaken at the endof each experiment, to verify that they remained permeabilizedthroughout the experiment. Cells that failed this test were rejectedfrom analysis.
3D reconstruction of cell shape was achieved with 2 conventionalmicroscopy cell images acquired in 2 perpendicular directions. Verticalmounting of the coverslip permitted visualization of the lateral shapeof the cell of interest close to the lower edge via a ×20 phase contrastobjective. In addition, top-view images of the same cell were recordedvia a second microscope objective (×10) attached to a phototube andmounted on amechanical micromanipulator (MX 100; Newport Instru-ments,Mississauga, Canada) perpendicularly to the first objective. Side-and top-view cell images were acquired simultaneously at 10- to 60-sintervals prior to osmolality challenge and at 5- to 30-s intervals duringthe challenge. Images of the cell side-view profile and top view of thecell base were recorded for 100 ms with two independent miniaturecharge-coupled device cameras (Moticam 350; Motic Instruments,Richmond, BC, Canada) and Motic software at 0.14-lm/pixel and 0.18-lm/pixel resolutions for side-view and top-view images, respectively.The images were saved on a hard disk and analyzed off-line by DISURtechnique [22]. This technique generates a set of topographical curvesof the cell surface from its digitized profile and base outline. Cell volume,surface and height were calculated from such a reconstructed cell topo-graphical model. All calculations were made entirely with Mat Lab(MathWorks, Natick, MA). To visualize the reconstructed cell model in3D perspective, data obtained with the help of the DISUR techniquewere used to generate a 2D matrix containing approximate z coordi-nates of points on the membrane surface. The 3D perspective of themodel cell was then plottedwithMat Lab or ORIGIN (Microcal Software,Northampton, MA).
Intracellular water volume in cells seeded in 12-well plates wasmeasured as [14C]-urea available space according to a previously-described protocol [28] and calculated as V = Ac / Amm, where Ac wasthe radioactivity of the cells after incubation with 2 μCi/ml [14C]-urea(dpm), Am was the radioactivity of the incubation medium (dmp/μl),and m was protein content in the cell lysate (mg).
To determine cell membrane K+ permeability, 86Rb uptake wasmeasured in cells growing in 24-well plates, washed twice with2 ml of medium containing 150 mM NaCl, 1 mM MgCl2, 1 mM CaCl2and 10 mM HEPES-tris buffer (pH 7.4, room temperature) and incu-bated for 30 min at 37 °C in 1 ml of B-iso medium. The preincubationmediumwas then replaced by 0.5 ml of the samemedium containing0.5–1 μCi/ml 86RbCl, 10 μM ouabain, 10 μM bumetanide or 1 mMBaCl2. The osmolality of this medium was decreased by reducingNaCl concentration. Preliminary experiments demonstrated that86Rb uptake was linear up to 20 min. This considered, isotope uptakewas terminated in 5 min by adding 2 ml of ice-cold medium contain-ing 100 mM MgCl2 and 10 mM HEPES-tris buffer (pH 7.4). The cellswere then transferred on ice, washed 4 times with 2 ml of the sameice-cold medium and lysed with 1 ml of 1% SDS/4 mM EDTA mixture.Radioactivity of the cell lysate was measured with a liquid scintilla-tion analyzer, and ion uptake was calculated as V = A / amt, whereA was radioactivity in the sample (cpm), a was the specific
enitsalbnivBnisalahcotyclortnoc
Fig. 1. Representative micrographs of microfilaments (red) and microtubules (green) in control A549 cells and cells treated for 20 min with 20 μM cytochalasin B or 10 μM vinblastine.
2339A. Platonova et al. / Biochimica et Biophysica Acta 1848 (2015) 2337–2343
radioactivity of 86Rb (K+) in the incubation medium, m was proteincontent in the sample (mg) and t was incubation time (min).Na+,K+,2Cl− cotransport activity was quantified as the ouabain-resistant, bumetanide-sensitive component of the 86Rb influx rate.Previously, we reported that, side-by-side with K+ channels, Ba2+
dose-dependently inhibited Na+,K+,2Cl− cotransport [29]. Withthese data in mind, the activity of K+ channels was estimated as(ouabain + bumetanide)-resistant, Ba2+-sensitive 86Rb influx.
Chemicals were procured from Gibco BRL (Gaithersburg, MO),Calbiochem (La Jolla, CA), Sigma (St. Louis, MO) and Anachemia (Mon-treal, QC, Canada). 86RbCl and [14C]-urea were obtained fromPerkinElmer (Waltham, MA, USA) and Isotope (St. Petersburg, Russia).
50
60
70
80
90
100
110
120
130
140
150
160
60
70
80
90
100
110B-iso B-hyper B-iso
B-iso B-hypo B-iso
V0V1
Vmin V2
V3
1/2
V0
V1
Vmax
V2
V3
A
C
Cel
l vol
ume,
%C
ell v
olum
e, %
τ
Fig. 2. Volume changes triggered by hyperosmotic (A, B) and hyposmotic (C, D) solutions in intmedium B (B-iso, ~300mOsm) and exposed to hyperosmotic B containing 200mMmannitol (Bcells were permeabilized by 5 μg/ml digitonin (dig) in isosmotic low-Na+, high-K+ intracellu200 mMmannitol (ILS-hyper, ~500mOsm) or hyposmotic ILS containing 70mMKCl (ILS-hypowith experiments shown in panels B and D, respectively. The mean value of cell volume docum
3. Results
Fig. 1 shows that microtubules (presented in green) were uniformlydistributed in the cytoplasmof A549 cells,whereas actinmicrofilaments(presented in red) were mainly located in peripheral regions neighbor-ing the plasma membrane, forming the so-called cortical cytoskeletonor membrane carcass. Consistently with previous reports [25,30], 20-min exposure to 20 μM cytochalasin B sharply decreased the projectedarea of A549 cells, slightly affected themicrotubule network and causedaccumulation of actin aggregates in the cortical and central regions ofthe cytoplasm, indicated by bright red spots. Unlike cytochalasin B,10 μM vinblastine disrupted the microtubules and increased actin
80
100
120
140
160
180
200
220
240
80
100
120
140
160
180
200
220 B-iso ILS-iso ILS-isoILS-hyper
B-iso ILS-iso ILS-isoILS-hypo
10 min
V0V1
V2
V3
Vmin
V4
dig
dig
V0
V1
V2
V3
V4
V5
B
D
Cel
l vol
ume,
%C
ell v
olum
e, %
act (A, C) and permeabilized (B, D) A549 cells. Intact cells were preincubated in isosmotic-hyper, ~500mOsm) or hyposmotic B containing 70mMNaCl (B-hypo, ~150mOsm). Thelar-like solution (ILS-iso, ~300 mOsm) and then exposed to hyperosmotic ILS containing, ~150mOsm). Four independent experiments shown in panels A and C were randomizedented during the first 5 min of incubation (Vo) was taken as 100%.
2340 A. Platonova et al. / Biochimica et Biophysica Acta 1848 (2015) 2337–2343
filament content in the cytoplasm aswell as randomly-distributed actinaggregates. Considering these observations, we undertook 20-mintreatment with cytochalasin B and vinblastine to examine the role of3D cytoskeleton in the volume responses of intact and permeabilizedA549 cells.
In intact A549 cells, elevation of medium osmolality achieved by theaddition of 200 mMmannitol to isosmotic medium B, culminated in at-tenuation of cell volume by ~35% (Fig. 2A). Consistent with previousfindings [23], we did not detect in these cells any significant RVI duringthe next 15min of incubation. In contrast to rapid hyperosmotic shrink-age, the transfer of cells to an isosmotic environment resulted in slowrestoration of cell volume that was not completed in 15min. Disruptionof microtubules by vinblastine did not alter the kinetics of cell volumeperturbations triggered by hyperosmotic solution (Fig. 3A), whereasdisassembly of actin microfilaments in the presence of cytochalasin Bsignificantly increased the time of half-maximal shrinkage (τ1/2) from113 ± 19 to 542 ± 58 s (Table 1).
As noted in earlier studies [23,31], plasma membrane perme-abilizationwith digitonin leads to ~2-fold elevation of A549 cell volume(Fig. 2B). This is attributed to a new Donnan equilibrium between fixednegative charges of the polyelectrolyte cytoplasmic gel network and
50
100
150
200
250
control
vinblastin
cytochalaA
C
Cel
l vol
ume,
%
50
100
150
200
osi-Bosi-B B-hyper
osi-Bosi-B B-hypo
1
Fig. 3. Effect of cytochalasin B and vinblastine on cell volume changes triggered by hyperosmoticExperiments shown in panels A and C were randomized with experiments shown in panels B avolume documented during the first 5 min of incubation was taken as 100%. Time windows fo
attracted diffusible counter ions that is established after removal of plas-ma membrane permeability barrier. The resulting elevated concentra-tion of inorganic counter ions leads to increased intra-gel osmoticpressure, water influx and swelling [5,21]. Heightened ILS osmolalityby the incorporation of 200 mM mannitol decreased permeabilizedcell volume by 70–80% (Δ=V3− Vmin, Table 2), in contrast with atten-uation of cell volume by 30–40% in intact A549 cells (Fig. 2A, Table 1).The augmented cell volume change values triggered by hyperosmoticmedium in permeabilized cells were consistent with previous resultsobtained in A549, 16HBE14o and HL-60 cells, which were attributedto high water-binding capacity of cytoplasmic gel [23]. Vinblastine didnot significantly influence the dynamics of volume perturbations inpermeabilized cells, whereas cytochalasin B decreased the amplitudeof swelling evoked by alteration of Donnan equilibrium measured asΔ=V3 − V1 (Fig. 3B, Table 2). This indicates reduced water absorptioncapacity of cytoplasmic gel after disruption of actin filaments. Consis-tent with this notion, in hyperosmotic medium cytocholasin-treatedcells retained less water, displaying smaller volume compared to un-treated cells (Fig. 3B). Similarly to intact cells, cytochalasin B reducedby ~5-fold the rate of permeabilized cell shrinkage evoked by hyperton-ic solution (parameter τ1/2, Table 2), suggesting that besides actin
e
sin B
D
B-iso
ILS-iso
ILS
-iso
ILS-hyper
ILS
-iso
B-iso
ILS-iso ILS-hypo
0 min
(A, B) and hyposmotic (C, D) solutions in intact (A, C) and permeabilized (B, D) A549 cells.nd D, respectively. Means ± S.E. from 4 experiments are reported. The mean value of cellr digitonin treatment are shown by gray bars. For other details, see Fig. 2 legend.
Table 3Effect of hyposmotic medium, cytochalasin B and vinblastine on intact A549 cell volume.
Themean cell volume value documented during the first 5min of incubationwas taken as100%. For time points selected for V1, Vmax, V2, and V3measurements, see Fig. 2C.Means±S.E. are reported from experiments performed in quadruplicate. *p b 0.02 compared to thecontrols.
Table 1Effect of hyperosmotic medium, cytochalasin B and vinblastine on volume of intact A549cells.
Themean cell volume value documented during the first 5min of incubationwas taken as100%. For time points selected for V1, Vmin, V2 and V3 measurements, see Fig. 2A. τ1/2 cor-responds to the time of half-maximal cell volume attenuation caused by transfer of cellsfrom isosmotic to hyperosmotic medium. Means ± S.E. are reported from experimentsperformed in quadruplicate. *p b 0.002 and **p b 0.001 compared to control values.
2341A. Platonova et al. / Biochimica et Biophysica Acta 1848 (2015) 2337–2343
filaments-dependent water binding by cytoplasmic gel, other watercompartments contribute to cell volume changes.
In contrast to sustained hyperosmotic shrinkage, hypotonic swellingof intact A549 cells was followed by rapid RVD and almost completenormalization of cell volume in 30 min of incubation (Fig. 2C). Wenoted that disruption of microfilaments and microtubules decreasedthemaximal amplitude of hyposmotic shrinkage during RVDmeasuredas Δ=V1 − Vmax (Table 3), consistent with reduced water binding ca-pacity of cytoplasmic gel in vinblastine/cytocholasin-treated cells. Nei-ther cytochalasin B nor vinblastine abolished RVD in hypotonically-swollen intact cells (Fig. 3C). As detected in an overwhelming numberof cell types studied so far [10,12], reperfusion of hypotonically-swollen A549 cells in isosmotic medium rapidly attenuated cell volumeby 30–40% (so-called post-RVD/RVI protocol). Both compounds de-creased extent of shrinkage triggered by reperfusion of A549 cellswith isosmotic medium, estimated as parameter Δ = V2 − V3
(Table 3), consistent with lower water content of cytoplasmic gel inthese cells.
Fig. 2D shows that permeabilization of A549 cells significantly in-creased the amplitude of hyposmotic swelling (parameter Δ=V4 −V3, Table 4) compared to intact cells (parameterΔ=V1−Vmax, Table 3).This is attributed to high water binding capacity by cytoplasmic gel andunrestricted counter ion and water influx in permeabilized cells [21,23].As predicted, plasmamembrane permeabilization by digitonin complete-ly abolished RVD mediated by the efflux of intracellular osmolytes alongtheir electrochemical gradients. Both cytochalasin and vinblastine elimi-nated swelling of permeabilized cells in hypotonic medium (Fig. 3D) aswell as volume alterations evoked by reperfusion of permeabilized cellswith isosmotic ILS and estimated as parameter Δ = V4 − V5 (Table 4).This demonstrates that water binding by intact cytoplasmic gel is amajor water compartment participating in hypotonic swelling.
Previously, we demonstrated that in A549 cells hyposmotic swellingactivates Ba2+-sensitive K+ channels [24]. In this study, we examinedthe regulation of these channels by cytoskeleton-active compounds.
Table 2Effect of hyperosmotic medium, cytochalasin B and vinblastine on volume of perme-abilized A549 cells.
Themean cell volume value documented during the first 5min of incubationwas taken as100%. For time points selected for V1, V2, V3, Vmin and V4 measurements, see Fig. 2B. τ1/2corresponds to the time of half-maximal cell volume attenuation caused by transfer ofcells from isosmotic to hyperosmotic medium. Means ± S.E. are reported from experi-ments performed in quadruplicate. *p b 0.02, **p b 0.002 and ***p b 0.001 compared tocontrol values.
We found that ~10-fold increment of K+ (86Rb) permeability triggeredby hyposmotic medium was completely abolished by 1 mM BaCl2,slightly diminished by vinblastine and was insensitive to the presenceof cytochalasin B (Fig. 4).
Keeping inmind the kinetic data on cell volumemodulation in intactA549 cells obtained by DISUR technique, we employed the [14C]-ureameasurement of intracellular water volume in 5 and 30 min of incuba-tion in hyposmotic medium for comparative analysis of the action ofBa2+ and cytoskeleton-active compounds on RVD. Fig. 5 shows that in-tracellular water volume was increased up-to 40–50% after 5 min of in-cubation in hypotonic medium compared to isotonic medium (brokenline). Consistently with DISUR data, we found that the volume of intra-cellular water was almost completely restored in 30 min of incubationin hypotonic medium. The normalization of intracellular water volumewas abolished by Ba2+ but preserved in the presence of cytochalasinB and vinblastine. The data show that despite elimination ofcytoskeleton-associated water binding compartment in vinblastine/cytochalasin-treated cells, they retain ability to near-normal watercontent restoration after hypotonic swelling. This process involvesBa2+-sensitive passive movement of K+ ions across the plasmamembrane and osmotically-obliged water and is independent of thecytoskeleton.
4. Discussion
The data obtained in this study prompted us to conclude that the cy-toskeleton network contributes to the behavior of cytoplasmic hydrogelas a sensor of extracellular osmolality. This conclusion is supported byseveral observations.
First, disruption of actin microfilaments by cytochalasin B sharplydecreased the rate of shrinkage triggered by perfusion of A549 cellswith hyperosmotic medium (Fig. 3A, Table 1). Importantly, this actionof cytochalasin B was preserved in permeabilized cells (Fig. 3B,Table 2), demonstrating that attenuation of the shrinkage rate after mi-crofilament disruption is caused by modification of hydrogel's waterbinding/retention properties rather than by the reduced water perme-ability of plasmamembranes. Second, disruption ofmicrotubules by vin-blastine lowered the maximal amplitude of hyposmotic swelling in
Table 4Effect of hyposmotic medium, cytochalasin B and vinblastine on permeabilized A549 cellvolume.
Themean cell volume value documented during the first 5min of incubationwas taken as100%. For time points selected for V1, V2, V3, V4 and V5 measurements, see Fig. 2D. Means±S.E. are reported from experiments performed in quadruplicate. *pb 0.02, **p b 0.01 and***p b 0.002 compared to control values.
control
cytochalasin
vinblastine
0
10
20
30
40
K+
influ
x, n
mol
/mg
ptot
ein/
min
50
60Ba2+
isosmotic medium hyposmotic medium
*
Fig. 4. Effect of 1 mM Ba2+, 20 μM cytochalasin B and 10 μM vinblastine on the rate of K+
influx, measured as 86Rb+ uptake, in intact A549 cells in control (B-iso) and hyposmotic(B-hypo) media. Means ± S.E. from 6 independent experiments are reported. *p b 0.001compared to control values.
2342 A. Platonova et al. / Biochimica et Biophysica Acta 1848 (2015) 2337–2343
intact cells (Fig. 3C, Table 3) and completely abolished this action ofhyposmotic medium on the volume of permeabilized A549 cells(Fig. 3D, Table 4). This demonstrates that water binding by intact cyto-plasmic gel plays a major role in hypotonic swelling. Third, cytochalasinB and vinblastine attenuated the shrinkage of intact swollen cellstriggered by their reperfusion in isosmotic medium (Fig. 3C, Table 3),consistent with reduced water binding capacity of cytoplasmic gel invinblastine/cytocholasin-treated cells.
A comparative analysis of the action of cytochalasin and vinblastineon intact and permeabilized cells suggests that cytoskeleton affects cellvolume via regulation of cytoplasmwater binding/retention capacity. Inaccordance with this working hypothesis, in cells with intact cytoskele-ton there is more structured gel-associated water which can rapidly re-spond to osmotic challenges. The process is likely linkedwith expulsionof excess water from densely-packed actin molecules within fibers.
2.0
2.5
3.0
3.5
Intr
acel
lula
r w
ater
, μl/m
g pr
ot
5 min 30 min
4.0
control
cytochalasin
vinblastine
Ba2+
*
Fig. 5. Effect of 1 mM Ba2+, 20 μM cytochalasin B and 10 μM vinblastine on the volume ofintracellularwater in intact A549 cells after 5 and 30min of incubation in hyposmoticme-dium.Means±S.E. from4 independent experiments are reported.Mean value of intracel-lular water volume after 30 min of incubation of cells in isosmotic medium is shown bybroken line. *p b 0.005 compared to control values.
Therefore, F-actin disruption by cytochalasin B during hypertonic cellshrinkage could counteract actin condensation/polymerization that, inturn, decreases the maximal size and slows down the osmolality-dependent volume changes. This hypothesis is consistentwith elevationof the F- to G-actin ratio seen in shrunken smooth muscle cells and dis-ruption by cytochalasin B the maximal amplitude of contractile re-sponses triggered by to osmotic challenges [26,32]. Importantly,similar to major extracellular fluids ILS should contain proteins orother high-molecular weight solutes which do not penetrate throughdigitonin-openedmembrane pores. Such solutes would counterbalanceisosmotic cell swelling by awell-known colloid-osmoticmechanism [5].Thus, additional experiments should be performed to clarify the relativecontribution of low- and high-molecular weight osmolytes in volumeregulation in permeabilized cells.
We found that disruption of the cytoskeleton network by these com-pounds did not abolish the RVD observed in intact A549 cells duringtheir prolonged incubation in hypotonic medium (Figs. 3C and 5).Both increment of furosemide-resistant K+ fluxes in swollen cells(Fig. 4) and restoration of intracellular water volume in 30 min of incu-bation in hyposmotic medium was completely suppressed by 1 mMBa2+ thus showing a key role of Ba2+-sensitive K+ channels in RVD.We did not detect any actions of cytochalasin B on the activation ofBa2+-sensitive K+ channels in swollen cells (Fig. 4) whereas the slightinhibition of these channels by vinblastine was probably caused by de-creased amplitude of the maximal swelling seen in 5 min of incubationof vinblastine-treated A4549 cells in hypotonic medium (Figs. 3C and5). These findings contradict the inhibition of RVD by microfilament-and microtubule-disrupting agents reported in Jurket cells [33], am-phibian gall bladder cells [15], rabbit proximal tubules [16], PC12 cellsand several mammalian tumor cell lines [17–19], but are consistentwith negative results obtained in studies on the actions of these com-pounds on RVD in human neutrophils [33] and HL-60 cells [33–35].Very modest inhibitory effects of cytochalasin B on RVD kinetic param-eters were detected in rabbit proximal tubule epithelial cells [36], PC12cells [37] and trout hepatocytes [38]. Moreover, in bovine articularchondrocytes, actin cytoskeleton depolymerization by latrunculineven accelerated post-RVD RVI [39]. This discrepancy might be a conse-quence of cell type-specific organization of the cytoskeleton networkand/or the actions of cytoskeleton active compounds [33].
5. Conclusion
Our results show, for the first time, that the cytoskeleton networkcontributes directly to osmosensing and passive cell volume adjust-ments in anisosmotic media via the modulation of the water that isbound/retained by the cytoplasmic hydrogel. However, the disruptionof hydrogel osmosensing function by cytoskeleton-active compoundsdid not abolished RVD in A549 cells, thus implicating alternative/com-plementary osmosensory mechanisms of cell volume regulation medi-ated by plasma membrane osmolyte transporters.
Transparency document
The Transparency document associated with this article can befound, in the online version.
Acknowledgments
This studywas supported in part by grants from theNatural Sciencesand Engineering Research Council of Canada, the Russian Foundation forFundamental Research #15-04-00101 and grant from the RussianScientific Foundation #14-15-00006. The authors thank Mr. Ovid DaSilva for manuscript editing.
2343A. Platonova et al. / Biochimica et Biophysica Acta 1848 (2015) 2337–2343
References
[1] T. Kreis, R.E. Vale, Guidebook to the Cytoskeletal and Motor Proteins, Oxford Univer-sity Press, New York, USA, 1999.
[2] E.K. Hoffmann, J.W. Mills, Membrane Events Involved in Volume Regulation, Topicsin Membranes, Academic Press, USA, 1999. 123–195.
[3] J.H. Henson, Relations between the actin cytoskeleton and cell volume regulation,Microsc. Res. Tech. 47 (1999) 155–162.
[4] H.F. Cantiello, Role of actin filament organization in cell volume and ion channel reg-ulation, J. Exp. Zool. 279 (1997) 425–435.
[5] Y. Okada, Ion channels and transporters involved in cell volume regulation and sen-sor mechanisms, Cell Biochem. Biophys. 41 (2015) 233–258.
[6] C. Rosette, M. Karin, Ultraviolet light and osmotic stress: activation of the JNK cas-cade through multiple growth factor and cytokine receptors, Science 274 (1996)1194–1197.
[7] D. Sheikh-Hamad, M.C. Gustin, MAP kinases and the adaptive response to hyperto-nicity: functional preservation from yeast to mammals, Am. J. Physiol. Ren. Physiol.287 (2004) F1102–F1110.
[8] C. Di Ciano-Oliveira, G. Sirokmany, K. Szaszi, W.T. Arthur, A. Masszi, M. Peterson,O.D. Rotstein, A. Kapus, Hyperosmotic stress activates Rho: differential involvementin Rho kinase-dependent MLC phosphorylation and NKCC activation, Am. J. Physiol.Cell Physiol. 285 (2003) C555–C566.
[9] S.F. Pedersen, E.K. Hoffmann, J.W. Mills, The cytoskeleton and cell volume regula-tion, Comp. Biochem. Physiol. A Mol. Integr. Physiol. 130 (2001) 385–390.
[10] F. Lang, G. Busch, M. Ritter, H. Volkl, S. Waldegger, E. Gulbins, D. Haussinger, Func-tional significance of cell volume regulatory mechanisms, Physiol. Rev. 78 (1998)247–306.
[11] A.A. Mongin, S.N. Orlov, Mechanisms of cell volume regulation and possible natureof the cell volume sensor, Pathophysiology 8 (2001) 77–88.
[13] B. Jensen, J.M. Schroder, L.B. Pedersen, S.F. Pedersen, Osmotic stress induces rapidcytoskeleton reorganization in human retinal pigment epithelial cells in a mannerdependent on the microtubule plus-end trafficking proteins EB1 and EB3, ActaPhysiol (Oxf.) 198 (2010) 212.
[14] S.F. Pedersen, A. Kapus, E.K. Hoffmann, Osmosensory mechanisms in cellular andsystematic volume regulation, J. Am. Soc. Nephrol. 22 (2011) 1587–1597.
[15] J.K. Foskett, K.R. Spring, Involvement of calcium and cytoskeleton in gallbladder ep-ithelial cell volume regulation, Am. J. Physiol. 248 (1985) C27–C36.
[16] M.A. Linshaw, T.J. Macalister, L.W. Welling, C.A. Bauman, G. Hebert, G.P. Downey,E.W.Y. Koo, A.I. Gotlieb, Role of cytoskeleton in isotonic volume control of rabbitproximal tubules, Am. J. Physiol. 261 (1991) F60–F69.
[17] J.W. Mills, S.F. Pedersen, P.S. Walmod, E.K. Hoffmann, Effect of cytochalasins on F-actin and morphology of Ehrlich ascites tumor cells, Exp. Cell Res. 261 (2000)209–219.
[18] M. Cornet, E. Delpire, R. Gilles, Study of microfilament network during volume reg-ulation process of cultured PC12 cells, Pflugers Arch. 410 (1987) 223–225.
[19] M. Cornet, I.H. Lambert, E.K. Hoffmann, Relation between cytoskeleton, hypo-osmotic treatment and volume regulation in Ehrlich ascites tumor cells, J. Membr.Biol. 131 (1993) 55–66.
[20] A.I. Yusipovich, M.V. Zagubizhenko, G.G. Levin, A. Platonova, E.Y. Parshina, R.Grygorczyk, G.V. Maksimov, A.B. Rubin, S.N. Orlov, Laser interference microscopyof amphibian erythrocytes: impact of cell volume and refractive index, J. Microsc.244 (2011) 229.
[21] R. Grygorczyk, F. Boudreault, A. Platonova, S.N. Orlov, Salt and osmosensing: role ofcytoplasmic hydrogel, Pflugers Arch. — Eur. J. Physiol. 467 (2015) 475–487.
[22] F. Boudreault, R. Grygorczyk, Evaluation of rapid volume changes of substrate-adherent cells by conventional microscopy 3D imaging, J. Microsc. 215 (2004)302–312.
[23] J. Fels, S.N. Orlov, R. Grygorczyk, The hydrogel nature of mammalian cytoplasm con-tributes to osmosensing and extracellular pH sensing, Biophys. J. 96 (2009)4276–4285.
[24] A. Platonova, F. Boudreault, L.V. Kapilevich, G.V. Maksimov, O. Ponomarchuk, R.Grygorczyk, S.N. Orlov, Temperature-induced inactivation of cytoplsmic biogelosmosensing properties is associated with suppression of regulatory volume de-crease in A549 cells, J. Membr. Biol. 247 (2014) 571–579.
[25] Y. Ujihara, H. Miyazaki, S. Wada, Morphological study of fibroblasts treated with cy-tochalasin D and colchicine using a confocal laser scanning microscopy, J. Physiol.Sci. 58 (2008) 499–506.
[26] S.V. Koltsova, S.V. Gusakova, Y.J. Anfinogenova, M.B. Baskakov, S.N. Orlov, Vascularsmooth muscle contraction evoked by cell volume modulation: role of cytoskeletonnetwork, Cell. Physiol. Biochem. 21 (2008) 29–36.
[27] S.N. Orlov, N. Thorin-Trescases, N.O. Dulin, T.-V. Dam, M.A. Fortuno, J. Tremblay, P.Hamet, Activation of cAMP signaling transiently inhibits apoptosis in vascularsmooth muscle cells in a site upstream of caspase 3, Cell Death Differ. 6 (1999)661–672.
[28] S.N. Orlov, J. Tremblay, P. Hamet, Cell volume in vascular smoothmuscle is regulatedby bumetanide-sensitive ion transport, Am. J. Physiol. 270 (1996) C1388–C1397.
[29] F. Gagnon, N.O. Dulin, J. Tremblay, P. Hamet, S.N. Orlov, ATP-induced inhibition ofNa+, K+, Cl- cotransport in Madin–Darby canine kidney cells: lack of involvementof known purinoceptor-coupled signaling pathways, J. Membr. Biol. 167 (1999)193–204.
[30] D.E. Ingber, D. Prusty, Z. Sun, H. Betensky, H. Wang, Cell shape, cytoskeletal mechan-ics, and cell cycle control in angiogenesis, J. Biomech. 28 (1995) 1471–1484.
[31] S.V. Koltsova, A. Platonova, G.V. Maksimov, A.A. Mongin, R. Grygorczyk, S.N. Orlov,Activtion of P2Y receptors causes strong and persistant shrinkage of C11-MDCKrenal epithelial cells, Am. J. Physiol. Cell Physiol. 301 (2011) C403–C412.
[32] T. Sasahara, K. Yayama, T. Tahara, H. Onoe, H. Okamoto, Na+/H+ exchanger inhib-itor augments hyperosmolarity-induced vasoconstriction by enhancing actin poly-merization, Vasc. Pharmacol. 59 (2013) 120–126.
[33] G.P. Downey, S. Grinstein, A. Sue-A-Quan, B. Czaban, C.K. Chan, Volume regulation inleukocytes: requirement for an intact cytoskeleton, J. Cell. Physiol. 163 (1995)96–104.
[34] K.R. Hallows, C.H. Packman, P.A. Knauf, Acute cell volume changes in anisotonicmedia affect F-actin content of HL-60 cells, Am. J. Physiol. 261 (1991) C1154–C1161.
[35] K.R. Hallows, F.Y. Law, C.H. Packman, P.A. Knauf, Changes in cytoskeletal actin con-tent, F-actin distribution, and surface morpholohy during Hl-60 cell volume regula-tion, J. Cell. Physiol. 167 (1996) 60–71.
[36] M.A. Linshaw, C.A. Fogel, G.P. Downey, E.W.Y. Koo, A.I. Gotlieb, Role of cytoskeletonin volume regulation of rabbit proximal tubule in dilute medium, Am. J. Physiol. 262(1992) F144–F150.
[37] M. Cornet, J. Ubl, H.A. Kolb, Cytoskeleton and ion movements during volume regu-lation in cultured PC12 cells, J. Membr. Biol. 133 (1993) 161–170.
[38] H.L. Ebner, A. Cordas, D.E. Pafundo, P.J. Schwarzbaum, B. Pelster, G. Krumschnabel,Importance of cytoskeletal elements in volume regulatory responses of trout hepa-tocytes, Am. J. Physiol. Regul. Integr. Comp. Physiol. 289 (2005) R877–R890.
[39] M.J. Kerrigan, C.S. Hook, A. Qusous, A.C. Hall, Regulatory volume increase (RVI) by insitu and isolated bovine articular chondrocytes, J. Cell. Physiol. 209 (2006) 481–492.
