Search

Many vertebrate cells react to hypotonic conditions with swelling, followed by an active downregulation of the cell volume; a progress called regulatory volume decrease (RVD). While the actual process of volume decrease by loss of osmotically active molecules like K(+) and Cl(-), followed by water efflux has been extensively investigated, the signal for activation of RVD still remains obscure. Studies with different cell lines demonstrated a participation of the cation channel transient receptor potential vanilloid 4 (TRPV4) as well as the actin cytoskeleton in volume regulation. Therefore, we analyzed putative links between TRPV4 and F-actin in RVD in HaCaT keratinocytes and CHO cells. Laser scanning microscopy studies revealed a distinct colocalization of TRPV4 and actin in highly dynamic membrane structures, such as microvilli, filopodia and lamellipodia edges. After treatment of cells with the actin-destabilizing reagent latrunculin A, TRPV4 and F-actin no longer colocalized within the membrane. In accordance with these data, close interaction between TRPV4 and F-actin was revealed by FRAP and FRET studies. For functional analysis, CHO cells that endogenously do not express TRPV4, were transfected with recombinant TRPV4, which rendered them RVD-competent. Treatment with latrunculin A abolished both, RVD and the accompanying rise of [Ca(2+)](i) after hypotonic stress in TRPV4-transfected CHO cells. Taken together, our data demonstrate a functional interaction between TRPV4 and F-actin in sensing hypotonicity and the onset of RVD.

The in vitro interaction between the glycolytic enzyme glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and cytoskeletal elements is well documented. To verify this association within cells, the intracellular distribution of GAPDH under various metabolic conditions has been investigated in immunostained cells or cells expressing GAPDH as a GFP fusion protein. GAPDH was homogeneously distributed in the cytoplasm and no interaction of GAPDH with cytoskeletal elements, neither with microfilaments nor microtubules or intermediate filaments, was detectable. In living cells expressing GFP-GAPDH, stress fibres were excluded from the fluorescence. In contrast to proliferating cells, the cytoplasmic GAPDH of serum-depleted cells was not homogeneously distributed, but colocalised with stress fibres. The mechanism for stimulating this actin-binding affinity was independent of the NO-signalling pathway. The results support the idea of a specialised function for the interaction of GAPDH and cytoskeletal elements, rather than a general function, as e.g. microcompartmentalization of glycolytic enzymes., (Copyright 2002 Elsevier Science Ltd.)

The discovery of the green fluorescent protein (GFP) and its use as a marker for proteins in cells revolutionised cell biology. Among its applications are the intracellular localisation of proteins and the investigation of the organisation, regulation and dynamics of the cytoskeleton. GFP itself is considered to be an inert protein, homogeneously distributed within the cytoplasm. Here we investigated the intracellular distribution of GFP in an amphibian and in various mammalian cell lines (XTH2, CHO-K1, HaCaT, MDCK, NIH-3T3) by confocal laser scanning microscopy. After paraformaldehyde fixation GFP became associated with microfilaments in all the cell lines investigated. This interaction was not impaired by detergent treatment (1% Brij 58 for 10 min). In contrast to the F-actin binding of GFP in fixed cells, association of GFP with stress fibres was not detectable in living cells. The actin-binding property of GFP might contribute also to the interaction of fusion proteins with microfilaments. Thus, careful controls are unavoidable in investigating (weak) actin-binding proteins in fixed cells. Because no association of GFP with microfilaments was detectable in living cells, it is recommended to monitor the intracellular distribution of GFP-tagged proteins in vivo.

Polymerization and depolymerization of cytoskeletal elements maintaining cytoplasmic stiffness are key factors in the control of cell crawling. Rheometry is a significant tool in determining the mechanical properties of the single elements in vitro. Viscoelasticity of gels formed by these polymers strongly depends on both the length and the associations of the filaments (e.g. entanglements, annealings and side-by-side associations). Ultrasound attenuation is related to viscosity, sound velocity and supramolecular structures in the sample. In combination with a small glass fibre (2 mm x 50 microm), serving as a viscosity sensor, an acoustic microscope was used to measure the elasticity and acoustic attenuation of actin solutions. Changes in acoustic attenuation of polymerizing actin by far exceed the values expected from calculations based on changes in viscosity and sound velocity. During the lag-phase of actin polymerization, attenuation slightly decreases, depending on actin concentration. After the half-maximum viscosity is accomplished and elasticity turns into steady state, attenuation distinctly rises. Changes in ultrasound attenuation depend on actin concentration, and they are modulated by the addition of alpha-actinin, cytochalasin D and profilin. Thus absorption and scattering of sound on the polymerization of actin is related to the packing density of the actin net, entanglements and the length of the actin filaments. Shortening of actin filaments by cytochalasin D was also confirmed by electron micrographs and falling-ball viscosimetry. In addition to viscosity and elasticity, the attenuation of sound proved to be a valuable parameter in characterizing actin polymerization and the supramolecular associations of F-actin.

A nondestructive method to determine viscoelastic properties of gels and fluids involves an oscillating glass fiber serving as a sensor for the viscosity of the surrounding fluid. Extremely small displacements (typically 1-100 nm) are caused by the glass rod oscillating at its resonance frequency. These displacements are analyzed using a phase-sensitive acoustic microscope. Alterations of the elastic modulus of a fluid or gel change the propagation speed of a longitudinal acoustic wave. The system allows to study quantities as small as 10 microliters with temporal resolution >1 Hz. For 2-100 microM f-actin gels a final viscosity of 1.3-9.4 mPa s and a final elastic modulus of 2.229-2.254 GPa (corresponding to 1493-1501 m/s sound velocity) have been determined. For 10- to 100-microM microtubule gels (native, without stabilization by taxol), a final viscosity of 1.5-124 mPa s and a final elastic modulus of 2.288-2. 547 GPa (approximately 1513-1596 m/s) have been determined. During polymerization the sound velocity in low-concentration actin solutions increased up to +1.3 m/s (approximately 1.69 kPa) and decreased up to -7 m/s (approximately 49 kPa) at high actin concentrations. On polymerization of tubulin a concentration-dependent decrease of sound velocity was observed, too (+48 to -12 m/s approximately 2.3-0.1 MPa, for 10- to 100-microM tubulin). This decrease was interpreted by a nematic phase transition of the actin filaments and microtubules with increasing concentration. 2 mM ATP (when compared to 0.2 mM ATP) increased polymerization rate, final viscosity and elastic modulus of f-actin (17 microM). The actin-binding glycolytic enzyme hexokinase also accelerated the polymerization rate and final viscosity but elastic modulus (2.26 GPa) was less than for f-actin polymerized in presence of 0.2 mM ATP (2.28 GPa).

XTH-2 cells, a cell line derived from tadpole heart endothelial cells, were blocked at the end of the G1 phase of the cell cycle using desoxyguanosine (dG). Stress fibers are the dominant actin fibril structure in blocked cells. They disappear starting with the release of the dG block until the G2 phase of the cell cycle. In addition, after the release peripheral lamellae and microspikes appear, indicating increased surface motility of the cells. Concomitantly with these changes, net lactic acid production is increased while oxygen consumption remains constant. Increase in the content of F-actin per cell takes place only in subconfluent cultures. Similar morphological changes (organization of F-actin) have been induced by phorbol myristate acetate treatment. These are also accompanied by an increase in lactic acid production and unaltered oxygen consumption. Therefore, the changes in energy metabolism are supposed to result from the rearrangement of the F-actin network: The emanating loose fibrillar pattern provides an increased surface for the association with glycolytic enzymes, which enhances enzyme activity and represents a more motile state. The increased energy demand of the more motile structures is supplied predominantly by ATP derived from glycolysis.

In density-inhibited 3T3 fibroblast cultures F-actin is localized almost exclusively at the cell-to-cell borders. Depolymerization of microtubules results in a rearrangement of the F-actin pattern. Numerous actin bundles are formed, most of them are anchored to the cell borders. This rearrangement is preceded by a transient rise in F-actin content.

The microcompartmentation of aldolase and glyceraldehyde-3-phosphate-dehydrogenase (GAPDH) was investigated in four different cell types (3T3 cells, SV 40 transformed 3T3 cells, mouse fibroblasts, chick embryo cardiomyocytes) combining cell permeabilization and indirect immunofluorescence technique. Permeabilization of the cells prior to fixation released the soluble fractions, whilst the total amount of enzymes was preserved in nonpermeabilized cells. Both enzymes exist in a soluble as well as in a structure-bound form. The soluble fraction of aldolase and GAPDH is distributed homogeneously throughout the cytoplasm, excluding the nucleus and vesicles. The permeabilization-resistant form is associated with the actin cytoskeleton. A considerable amount of both enzymes is located in the perinuclear region and cannot be attributed to a definite structure. Comparing the staining patterns of aldolase and GAPDH in four different cell types we found that the distribution of the enzymes corresponds with diverse forms of actin cytoskeletal organization of these cells. The codistribution is maintained in cells treated with cytochalasin D.

Effects of cytochalasins on actin polymerization state in living cells were measured using fluorimetry of TRITC-phalloidin bound to F-actin. Normal (3T3) and tumour (SV-3T3, B16 melanoma, and Ehrlich ascites) cells were treated with cytochalasin B and cytochalasin D (1 microgram/ml). Three effects of cytochalasins were demonstrated--depolymerization of F-actin, promotion of polymerization, and redistribution of actin without change in polymerization state. Occurrence of a given effect was dependent on cell type, cell density, cytochalasin concentration and type. This indicates that cells from different lines, and even the same cells in different culture conditions may differ significantly in their state of actin polymerization, which we suppose is the cause of their different reactions to cytochalasins. Accordingly, caution should be taken in generalizing the results concerning the effect of cytochalasis on the polymerization state of actin.

The physiological significance of the association of glycolytic enzymes with actin fibrils was investigated in cell culture. Cytochalasin D (CD) was used to induce the known actin-based sequence of events in a culture of an endothelial-cell line (XTH-2) derived from hearts from tadpoles of Xenopus laevis. 1 min following addition of CD, ruptures in the cortical fibrillar meshwork and in stress fibres are seen. At the same time the cellular ATP level decreases by ca. 25%. This and the following reactions resulting in a kind of arborization depend on a continuous supply with metabolic energy. As shown by measurements of oxygen consumption, cells with intact energy metabolism provide the ATP needed from glycolysis; ATP produced by oxidative phosphorylation is not utilized as long as lactate dehydrogenase (LDH) reoxidizes NADH2. After inhibition of LDH, respiration in XTH-2 cells doubles. CD treatment induces a transient increase in oxygen consumption, indicating an increased energy supply by respiration. From these results we conclude: The energy needed by the actomyosin system is - under normal metabolic conditions -supplied from ATP phosphorylated in glycolysis. The processes of energy metabolism seem to be highly compartmentalized; ATP is not a parameter that is kept constant in time intervals of minutes up to one hour.

The dependence of the arrangement of fibrillar actin in cultured endothelial cells on metabolic conditions was investigated with cellular elements derived from the heart of Xenopus laevis tadpoles. Either primary culture or an established cell line (XTH-2) were used in these studies. The metabolic stage of the cells was influenced by inhibiting respiration and lactate production. The actin pattern was revealed either by indirect immunofluorescence or by tetramethylrhodaminyl (TRITC)-phalloidin fluorescence. Total block of energy supply causes in all cases a distinct loss of actin fibrils, while inhibition of respiration alone increases the variability of actin organization. In primary XTH cells but not in XTH-2 cells cyanide disintegrates most of the actin fibres during 3 h of treatment. This effect is independent of the inhibition of respiration, since actin gels prepared from skeletal muscle also undergo destruction in the presence of cyanide. It is concluded that the actin fibrils of the primary cells and the established line behave differently to changing metabolic conditions and to application of KCN.

In XTH-2 cells, a line derived from tadpole heart endothelia, changes of the F-actin pattern have been followed after injection of calcium and after permeabilization of the cells to calcium using the ionophore A 23187 and an external pCa of 6.0. On the dorsal surface numerous microvilli appear, the stress fibres show an interrupted pattern and F-actin accumulates at the cell periphery. These reactions are interpreted to result from contractions. Finally most of the F-actin becomes depolymerized (in about 30 min). In addition actin containing accumulations are developed in the perinuclear area. In G1 cells stress fibres are less sensitive to enhanced calcium, they persist for more than 30 min in pCa 6.0. Taking into account a relatively high cytosolic calcium concentration in many malignantly transformed cells, their more rounded appearance and lack of an ordered array of actin fibrils could be closely related phenomena.

A method is described for the determination of cellular F-actin content using fluorometry of TRITC-phalloidin at very high dilution. In this case no saturation of all binding sites available is reached, however the staining is highly specific and the specificity is not affected by the preparative procedure as may be the case at high concentrations of TRITC-phalloidin. The method is based on calculation of fluorescence intensity at equilibrium conditions from measurements at two different numbers of exchange of the staining solution by using Lineweaver-Burk plotting. The relative content of F-actin has been determined for three established cell lines, an amphibian cell line XTH-2 of endothelial origin, 3T3 cells and SV 40 transformed 3T3 cells. In the two non transformed lines F-actin decreases with increasing cell density starting with the onset of confluency of the culture. SV 40 transformed 3T3 cells generally contain less F-actin and do not show any significant point of change. The decrease in F-actin with increasing cell density is accompanied by a disappearance of stress fibres. SV 40 3T3 cells generally are devoid of stress fibres. The observations are discussed considering a possible involvement of F-actin in growth control.