Involvement of Gelsolin in Cadmium-Induced Disruption of the Mesangial
http://www.100md.com
《毒物学科学杂志》
Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, M5S 1A8, Canada
ABSTRACT
Cadmium (Cd2+) is known to cause a selective disruption of the filamentous actin cytoskeleton in the smooth muscle-like renal mesangial cell. We examined the effect of Cd2+ on the distribution of the actin-severing protein, gelsolin. Over 8 h, CdCl2 (10 μM) caused a progressive shift of gelsolin from a diffuse perinuclear and cytoplasmic distribution to a pattern decorating F-actin filaments. Over this time filaments were decreased in number in many cells, and membrane ruffling was initiated. Western blotting and 125I-F-actin gel overlays demonstrated an increase in actin-binding gelsolin activity in the cytoskeletal fraction of cell extracts following Cd2+ treatment. In in vitro polymerization assays, gelsolin acted as a nucleating factor and increased the rate of polymerization. Cytosolic extracts also increased the polymerization rate. Addition of Cd2+ together with gelsolin further increased the rate of polymerization. Gelsolin enhanced depolymerization of purified actin, and Cd2+ partially suppressed this effect. However, cytoskeletal extracts from Cd2+-treated cells also markedly increased depolymerization, suggesting further that Cd2+ may activate cellular component(s) such as gelsolin for actin binding. We conclude that a major effect of Cd2+ on the mesangial cell cytoskeleton is manifest through activating the association of gelsolin with actin, with gelsolin's severing properties predominating under conditions found in Cd2+-treated cells.
Key Words: F-actin; actin depolymerization; cadmium toxicity.
INTRODUCTION
Cadmium (Cd2+) is a divalent metal ion that exhibits potent toxicity to most cells and organisms (Bhattacharyya et al., 2000; Goering et al., 1995; Nordberg et al., 1992). While the proximal tubule is a major target of Cd2+ toxicity, damage to the glomerulus is also significant (Jrup, 2002). Workers exposed to Cd2+ show urinary markers characteristic of glomerular injury (Fels et al., 1994; Roels et al., 1993). Occupational exposure to Cd2+ is associated with glomerular injury and decreased glomerular filtration rate (Jrup et al., 1993, 1995). The mesangial cell is a contractile smooth muscle-like cell of the glomerulus involved in regulating the glomerular filtration rate. Cadmium causes rearrangement of the actin cytoskeleton in cultured mesangial cells (Wang et al., 1996) and stimulates them to contract (Barrouillet et al., 1999a,b; L'Azou et al., 2002a,b). It has been suggested that Cd2+-induced mesangial cell contraction underlies the changes in intrarenal hemodynamics and glomerular filtration rate that precede proteinuria (Barrouillet et al., 1999b; Jrup et al., 1993, 1995; Roels et al., 1993). The contribution of cytoskeletal disruption has not been studied in detail.
Previously we showed that exposure of mesangial cells to Cd2+ was accompanied by a loss of filamentous (F-) actin structure and organization (Wang et al., 1996; Wang and Templeton, 1996). Exposure of mesangial cells to Cd2+ resulted in F-actin depolymerization without a compensatory increase in cellular G-actin levels (Wang et al., 1996). These smooth muscle-like cells showed some specificity for this effect of Cd2+, a number of other divalent metals having no effect on F-actin at the same concentration as Cd2+. The cytoskeleton of smooth muscle cells in general seems to be particularly susceptible to disruption by Cd2+ (Templeton, 2000). The polymerization of F-actin from monomeric, globular (G-) actin subunits is regulated by a myriad of proteins, many of them Ca2+-dependent (Ayscough, 1998; Carpenter, 2000; Vandekerchove, 1993; Welch and Mullins, 2002). Further, the redox status of the cell, reflected in part in the thiol content of its F-actin cytoskeleton, is another determinant of cytoskeletal structure (Stournaras, 1990; Valentin-Ranc and Carlier, 1991).
The biological interactions of Cd2+ are dominated by two over-riding chemical features. First, its strong propensity for sulfur as a ligand disrupts the structure and function of a number of proteins and enzymes (Beyersmann and Hechtenberg, 1997; Bhattacharyya et al., 2000 Goering et al., 1995; Waalkes et al., 1992) by poisoning protein thiol groups. Second, it has an ionic radius (0.97 ) very close to that of Ca2+ (0.99 ), and so potentially can interfere with Ca2+-dependent signaling processes; it can substitute for Ca2+ in vitro, activating Ca2+-dependent forms of protein kinase C and supporting calmodulin signaling (Mazzei et al., 1984; Suzuki et al., 1985). Thus, because of its interactions with thiol groups and substitution for Ca2+, it is not unexpected that Cd2+ will influence cytoskeletal integrity. The depolymerizing effects of Cd2+ on F-actin were, though, found to be independent of its effects on cytosolic [Ca2+], and independent of direct effects on Ca2+-binding proteins (Wang et al., 1996). In fact, Cd2+ stimulated actin polymerization in a reconstituted in vitro assay system. However, extracts made from cells that had been treated with Cd2+ mimicked the depolymerizing effects of Cd2+ on actin in cultured cells (Wang and Templeton, 1996). And, exposing mesangial cells to Cd2+ influenced actin-binding activity of protein(s) with molecular weights of approximately 90 kDa and 45 kDa, differentially in cytosolic and cytoskeletal-associated fractions. The indication is that Cd2+ does not act directly on the cytoskeleton, but affects the expression and (or) activity of proteins that may regulate the status of F-actin.
Gelsolin represents a family of ca. 90 kDa monomeric actin-binding proteins present in most animal cells, that sever F-actin and cap the quickly growing barbed end, thus favouring depolymerization from the pointed end (Ayscough, 1998; Sun et al., 1999; Yin and Stossel, 1979). Cardiac myocytes from gelsolin -/- mice show increased F-actin bundling of stress fibers in culture (Lader et al., 1999), whereas overexpression of gelsolin enhances cytoskeletal reorganization and fibroblast motility (Cunningham et al., 1991). Gelsolin binding to actin is regulated by Ca2+ and polyphosphoinositide-4,5-bisphosphate (Gremm and Wegner, 2000; Kwiatkowski, 1999; Sun et al., 1999). At sub-micromolar Ca2+ concentrations, Ca2+-regulated capping and severing activities can occur independently of one another (Bryan and Coluccio, 1985; Gremm and Wegner, 2000). Gelsolin can also bind to G-actin monomers (Gremm and Wegner, 1999, 2000) and may facilitate nucleation in some circumstances.
We have pursued effects of Cd2+ on the mesangial cell cytoskeleton in order both to elucidate potential mechanisms of cadmium toxicity and to gain further insight into features controlling the F-actin–G-actin equilibrium in these cells. The present study was undertaken to evaluate the effects of Cd2+ on gelsolin expression and binding in mesangial cells, and in particular to evaluate gelsolin's involvement in Cd2+-induced disruption of cellular F-actin content.
MATERIALS AND METHODS
Cell culture conditions and treatment.
Rat mesangial cells were isolated as described previously (Wang et al., 1994) and were grown in RPMI 1640 medium without antibiotics, supplemented with 10% (v/v) fetal bovine serum, in a humidified atmosphere of 5% CO2 at 37°C. Cells growing on 10-cm dishes at 70% confluence, or on cover slips, were made quiescent by serum deprivation (0.2% serum for 48 h). Quiescent cells were treated with 10 μM CdCl2 in RPMI 1640 for up to 8 h. All experiments were performed with cells between passages 5 and 20. Cytosolic and cytoskeletal-enriched fractions were prepared by lysis of cells in actin polymerization buffer (see below) containing 0.2% Triton X-100 as described previously (Wang and Templeton, 1996).
Actin purification.
Rabbit skeletal muscle actin was prepared according to Selden et al. (2000). Briefly, G-actin was extracted from skeletal muscle acetone powder (prepared as described previously; Wang and Templeton, 1996) with buffer G (2 mM Tris-HCl, pH 8.0 at 25°C, with 0.1 mM CaCl2 and 0.5 mM dithiothreitol [DTT]) containing 0.2 mM ATP. The mixture was centrifuged at 10,000 x g for 20 min and the supernatant was further centrifuged at 150,000 x g for 30 min at 4°C. The supernatant was polymerized by addition of 3 M KCl to a final concentration of 0.8 M and left over night at 4°C with very slow stirring. The solution was centrifuged at 150,000 x g for 2 h and the F-actin pellet was soaked in buffer G for 2 h on ice. Pellets were picked up with a glass rod, homogenized carefully in a Dounce homogenizer, and dialysed against 500 ml of buffer G for 48 h. The dialysed solution was centrifuged at 150,000 x g for 2 h at 4°C. The polymerization step was repeated twice and G-actin concentration was determined spectrophotometrically at 290 nm using an absorption coefficient of = 26,600 M–1 cm–1. G-actin was further purified by gel filtration on a 2.6 x 100 cm Sephadex G-150 column pre-equilibrated with buffer G. G-actin (5 mg/ml) was eluted at a flow rate of 1 ml/min. Five-milliliter fractions were collected and the absorption at 280 nm was monitored. Fractions corresponding to G-actin (Mr = 43 kD) were pooled and stored at 4°C for up to 2 weeks.
Immunoblotting and immunoprecipitation.
Mesangial cell extracts were prepared by harvesting cells at the indicated times after addition of 10 μM CdCl2 and lysis in buffer A (20 mM Tris-HCl, pH 7.4, with 2 mM MgCl2, 138 mM KCl, 1 mM ATP, and 0.2 % [v/v] Triton X-100). The lysates were cleared by centrifugation (10,000 x g, 15 min) and the pellets were resuspended in buffer A without Triton X-100, and either used immediately or stored at –80°C. SDS-PAGE was performed on (4–20)% gradient gels. Proteins were transferred to polyvinylidene difluoride (PVDF) membranes, and probed with polyclonal anti-gelsolin antibody (C-20) and secondary horseradish peroxidase-conjugated anti-goat IgG, both from Santa Cruz Biotechnology (Santa Cruz, CA). Bands were visualized by enhanced chemiluminescence detection kit (ECL+, Amersham) according to the manufacturer's instructions. For immunoprecipitation, mesangial cells were treated with CdCl2 and rinsed with ice-cold PBS, harvested, and lysed in immunoprecipitation buffer C (buffer A + 2 μg/ml aprotinin, 2 μg/ml leupeptin, 1 mM phenylmethylsulfonyl fluoride, 10 mM NaF, 1 mM Na3VO4). Samples containing 500 μg protein were incubated overnight at 4°C with anti-gelsolin antibody (1 μg/sample). Immune complexes were isolated by the addition of protein G Sepharose (2 h, room temperature) and washed five times with buffer C. Proteins bound to the Sepharose beads were released by boiling in SDS-PAGE sample buffer for 5 min, resolved by gradient SDS-PAGE (4–20%), transferred to PVDF, blocked with Membrane blocking agent (Amersham), and probed with anti-gelsolin antibody or 125I-F-actin.
Indirect immunofluorescence and confocal microscopy.
Mesangial cells treated as described above were washed twice in PBS and fixed with 3.8% para-formaldehyde (4°C, 40 min). After two washes with PBS, cells were permeabilized with 0.2% Triton X-100 in PBS and blocked in PBS containing 5% bovine serum albumin for 1 h at room temperature, followed by a further 1 h incubation with anti-gelsolin antibody. After three 10-min washes with PBS, cells were exposed for 1 h to secondary antibody conjugated to fluorescein isothiocyanate (FI-5000; Vector Laboratories, Burlingame, CA). Cells were rinsed three times with PBS and incubated with Alexa Fluor 568-labeled phalloidin (Molecular Probes; Eugene, OR), following the manufacturer's instructions. Following six washes in PBS samples were mounted in 4',6-diamidino-2-phenylindol dihydrochloride (DAPI) containing Vectashield mount medium (Vector Laboratories). Photomicrographs were obtained with a Nikon Eclipse E-600 microscope equipped with a Hamamatsu digital camera (Hamamatsu Inc., Japan), using Simple PCI software (Compix Inc., PA). Confocal microscopy was performed with a Zeiss LSM 410 inverted laser scanning microscope equipped with flourescein and rhodamine filters and a Zeiss 63 x 1.40 Plan-Apochromat oil immersion lens. Image analysis was performed with ImageJ 1.24t software (National Institutes of Health, U.S.). The results are representative of at least three independent experiments conducted in triplicate.
125I-F-actin preparation.
G-actin (1 mg/ml) prepared as described above was labeled with 250 μCi 125I (carrier-free NaI; ICN Pharmaceuticals, Costa Mesa, CA) in the presence of an Iodo-Bead (Pierce Biotechnology, Rockford, IL). Radiolabeled G-actin was separated from unincorporated 125I by gel filtration at room temperature through a 5 ml PD-10 column (Amersham Bioscience, Uppsala, Sweden) equilibrated with buffer B (2 mM Tris-HCl, pH 8.3, with 0.2 mM CaCl2, 0.2 mM ATP, and 0.5 mM DTT) at 0°C. 125I-G-actin was polymerized by gentle mixing with 0.25 volume of 100 mM PIPES-NaOH buffer, pH 7.0, containing 250 mM KCl, 10 mM MgCl2, and 0.25 mM CaCl2 at room temperature for 1 h. 125I-F-actin was collected by centrifugation at 150,000 x g for 2 h at 4°C. The pellet was resuspended in 500 μl of buffer B and phalloidin was added to give a final concentration of 40 μM. The mixture was then incubated for another 15 min at room temperature and the phalloidin-stabilized 125I-F-actin was stored at 4°C.
125I-F-actin overlay-assay.
125I-F-actin overlay assays were performed essentially as described (Luna, 1998). The blocked PVDF membranes were incubated overnight without disturbance with 50–100 μg/ml 125I-F-actin stabilized with 40 μM phalloidin. PVDF membranes were washed five times for 2 min with TBST (10 mM Tris-HCl, pH 7.5, with 90 mM NaCl and 0.5% [v/v] Tween-20) and exposed to film at –80°C for 2–24 h.
Pyrene labeling of actin.
G-actin (3 mg/ml in buffer G) was dialysed against buffer G (with 0.2 mM ATP but without DTT) for 48 h at 4°C. Pyrene iodoacetamide (Molecular Probes, Eugene, OR) was dissolved in dimethylsulfoxide and added to G-actin at a 10:1 molar ratio. The mixture was left overnight with continuous stirring at 4°C in the dark. G-actin-pyrene was then dialysed against buffer G for 48 h at 4°C. To remove unreacted excess pyrene, the mixture was chromatographed on a PD-10 Sephadex column and 500 μl fractions were collected after elution with buffer G. The concentration of labeled G-actin was calculated from the absorption at 290 nm after correction for pyrene absorption at 344 nm by subtracting 0.127 A344 from the measured A290. Concentration ratios [pyrene]/[actin] were calculated using 344 = 2.22 x 104 M–1 cm–1 for pyrene. The molar ratio of pyrene to actin monomer in the labeled product was between 0.7 and 1.0 in all experiments.
Polymerization and depolymerization assays.
Kinetic measurements following the changes in pyrene-labeled actin fluorescence were initiated in polymerization buffer (20 mM Tris-HCl, pH 7.4, containing 2 mM MgCl2, 138 mM KCl, and 1 mM ATP). Pyrene-G-actin (2 μM) was used in all experiments which followed polymerization rates, and for the initiation of depolymerization pyrene-F-actin was initially diluted to 200 nM in the same polymerization buffer. To study the effects of Cd2+ and gelsolin, pyrene-labeled actin was added to polymerization buffer containing various concentrations of Cd2+ and/or gelsolin, and the fluorescence intensity of pyrene (excitation at 365 nm, emission at 385 nm) was recorded for up to 10 min on a Fluorolog spectrofluorimeter (J Y Horiba, Edison, NJ). The conditions for actin polymerization and depolymerization in the presence of cytosolic and cytoskeletal-enriched fractions (5 μg/μl final protein concentration) were as described previously (Wang and Templeton, 1996). All measurements were performed at 37°C.
RESULTS
Immunohistochemical Localization of Gelsolin
Cells were seeded on a glass surface, starved for 48 h in 0.2% serum, and then treated with 10 μM CdCl2 for up to 8 h. Longer exposures and higher concentrations result in some cell death with loss of filaments in all cells (not shown; see Wang et al., 1996), whereas shorter exposures resulted in less cytoskeletal disruption. Fluorescence photomicrographs of phalloidin-stained cells (Figs. 1a and 1b) show a disruption of actin filaments in many cells after Cd2+ treatment, confirming earlier observations (Wang et al., 1996). The response is heterogeneous at a given time point, with not all cells showing loss of filaments by 8 h, and we observed that this was dependent on cell density. In a typical experiment, cultures plated at 104 cells cm–2 (as in Fig. 1) showed significant disruption of filaments in 40% of cells (n = 293) while at 105 cells cm–2 only 18% (n = 356) were affected. Based on these observations, we examined further the distribution of gelsolin in cells prior to Cd2+ treatment, and after 8 h of Cd2+ exposure in cells with and without a remaining F-actin cytoskeleton.
Gelsolin distributes in a diffuse, punctate pattern of staining throughout the cytoplasm and especially associated with the nucleus of control cells not exposed to Cd2+ (Fig. 1c). Following 2 h of Cd2+ exposure, cells showed prominent staining for gelsolin in the cytoplasm and evidence of some filamentous arrangement (Fig. 1d). After 8 h of Cd2+ treatment, gelsolin was predominantly localized along any remaining F-actin filaments (Fig. 1e). In cells in which the F-actin filaments remained more or less intact after 8 h of Cd2+, strong colocalization of gelsolin with actin was observed (Figs. 1f–1h). Scanning across magnified fields of confocal images of cytosolic regions (Figs. 1i and 1j) confirmed a shift of gelsolin from a diffuse distribution before Cd2+ treatment, to take on a periodicity matching that of F-actin at 8 h. Images and analyses in Figures 1c–1j are representative of multiple fields from at least 10 different experiments. (Note that panels i and j are produced from the high resolution confocal images in panels c and e, and are magnified to allow visualization of the pixelated images.) The percentage of the cells showing localization of gelsolin to F-actin fibers steadily increased with the time of Cd2+ treatment (not shown). These observations suggest that gelsolin redistribution from the nucleus and cytoplasm to the actin cytoskeleton can be triggered by Cd2+ treatment. In addition, Cd2+ treatment and accompanying stress fiber reorganization induces areas of membrane ruffling, or lamellipodia, rich in gelsolin (Fig. 2). These are seen rarely in control cells (an example is shown in Fig. 2a), but are frequent in Cd2+-treated cells (Fig. 2b). Gelsolin localization to the membrane ruffles was detected in over 60% of the cells after 8 h of Cd2+exposure, where it colocalized with actin filaments (Fig. 2c).
Western Blotting and Gel Overlays
To confirm the Cd2+-dependent redistribution of gelsolin binding activity, cytosolic and cytoskeletal fractions from control and Cd-treated cells were subjected to Western blotting. Two actin-binding protein bands with Mr values of about 90 and 40 kDa were detected (Fig. 3). The band with Mr = 90 kD corresponds to full-length gelsolin and the 40 kDa band may be attributed to a cleaved gelsolin peptide (see Discussion) recognized by the antibody against a C-terminal peptide; this lower band is also present in a purified gelsolin standard. Although no change in total gelsolin mRNA in whole cell extracts was found (data not shown), the intensity of both gelsolin protein bands increased in the cytoskeletal fraction in a time-dependent manner following Cd2+ treatment, consistent with the results of immunofluorescence. The putative cleaved fragment at 40 kDa was absent from cytosol before Cd2+ treatment but was apparent in this fraction at and after 4 h.
To demonstrate that these bands represented actin-binding activity, cell fractions were electrophoresed and blotted with [125I]-F-actin. Despite the high background inherent in this technique, at least five actin-binding bands were identified (numbered 1–5, Fig. 4). Two of these (bands 2 and 5) were prominent in the gelsolin standard, representing intact gelsolin (band 2) and probably a cleaved fragment containing the C-terminal epitope (band 5). The nature of the other bands was not pursued in this study, including the very prominent band 4 which is restricted to the cytoskeletal fraction and increases markedly with Cd2+ treatment. Immunoprecipitates with anti-gelsolin antibody were prepared to confirm the identity of bands 2 and 5. [125I]-F-actin overlays of gelsolin immunoprecipitates showed only bands 2 and 5 (Fig. 4c). These bands show little or no change in binding activity in the cytosolic fraction after 8 h of treatment with Cd2+. However, actin-binding activity of gelsolin, especially of the 40 kDa fragment, increases in the cytoskeletal fraction after Cd2+ exposure, consistent with the results of immunofluorescence and Western blotting.
Polymerization/Depolymerization Studies
We previously showed that Cd2+ both facilitates polymerization and suppresses depolymerization of actin in an in vitro assay (Wang and Templeton, 1996), both counter to the loss of actin filaments from the cell. Therefore, direct interaction of Cd2+ with F-actin/G-actin is unlikely to account for Cd2+-induced loss of filaments in vivo. However, the above results raise the possibility that Cd2+ may have different effects when gelsolin is present, and we therefore studied the polymerization and depolymerization of purified actin in the presence of Cd2+ and gelsolin together. Pyrene-labeled G-actin was dissolved in polymerization buffer at 2 μM and polymerization was initiated by the addition of ATP, resulting in an apparent first-order rate of increase in fluorescence intensity (Fig. 5a). When gelsolin was included in the reaction solution, a sigmoidal polymerization curve was observed, representing a lag phase indicative of a nucleation process followed by a linear phase of polymerization at a markedly increased rate. We confirmed our previous observation (Wang et al., 1996) that inclusion of Cd2+ in the polymerization mixture increased the rate of polymerization at concentrations in the 200 μM range, though concentrations of 10–100 nM were without significant effect in the absence of gelsolin (not shown). When low concentrations of Cd2+ (20–80 nM) were added to a polymerization mixture containing 5 nM gelsolin, there was an indication that the lag phase was decreased and the rate of polymerization was unaffected or increased (Fig. 5a), although not in any apparent concentration-dependent manner. These results are presented to demonstrate that neither Cd2+ nor gelsolin, alone or in combination, drive depolymerization of the filaments.
To study depolymerization, pyrene-labeled F-actin was diluted to 200 nM (based on the G-actin subunit) by the addition of buffer. This is below the critical concentration for polymerization, and spontaneous depolymerization was observed by fluorescence quenching. Depolymerization was enhanced by inclusion of gelsolin in the dilution buffer (Fig. 5b), and increased with increasing gelsolin:actin ratio (not shown), consistent with well documented capping at the barbed end and dissociation from the pointed end. Inclusion of Cd2+ in the gelsolin mixture prior to addition of actin slowed the rate of depolymerization in a concentration-dependent manner, though even at a ratio of Cd2+:gelsolin of 10:1 depolymerization was still more rapid than in the absence of gelsolin.
Because Cd2+ exposure facilitated a localization of gelsolin with the cytoskeleton in cultured cells, cytosolic and cytoskeletal cell extracts from control and Cd2+-treated cells were examined for their ability to affect depolymerization of purified F-actin. Cytosolic extracts from control cells did not affect the spontaneous depolymerization of actin below its critical concentration (Fig. 6a). Cytoskeletal extracts from control cells increased the rate of depolymerization (Fig. 6b), consistent with the presence of protein(s) in this fraction that facilitate depolymerization. When obtained from Cd2+-treated cells, both cytosolic and cytoskeletal fractions increased the rate of depolymerization.
DISCUSSION
Rat mesangial cells have previously been shown to undergo disruption of F-actin after treatment with Cd2+ (Barrouillet et al., 1999b; L'Azou et al., 2002b; Wang et al., 1996; Wang and Templeton, 1996). The present study corroborates these findings and shows that Cd2+ treatment results in F-actin-gelsolin interactions that accompany changes in F-actin stress fiber organization and distribution, cell-cell contacts, and organization of membrane ruffles. A redistribution of gelsolin into the detergent-insoluble cytoskeletal fraction on Western blots and immunolocalization of gelsolin to the actin cytoskeleton in those cells where actin filaments remained, indicates that Cd2+ treatment favors association of gelsolin with actin filaments. This association is seen as early as 2 h after exposure to Cd2+ and precedes gross actin disruption; gelsolin remains associated with residual shortened actin filaments throughout the process, leading to the eventual disappearance of many fibers by 8 h. Thus, gelsolin is a leading candidate for mediating the effects of Cd2+ on the cytoskeleton.
Disruption of F-actin by enhanced association with gelsolin is consistent with the well-described capping and/or severing properties of gelsolin (Ayscough, 1998; Gremm and Wegner, 2000; Kinosian et al., 1998; Lagarrigue et al., 2003; Sun et al., 1999). Nevertheless, kinetic analysis indicates that complex mechanisms are involved. Thus, gelsolin is also an effective nucleating factor for G-actin polymerization under some circumstances (Burtnick et al., 1997; Ditsch and Wegner, 1994; Feinberg et al., 1997). This effect dominates in the assay with purified actin, e.g., at a gelsolin:G-actin ratio of 1:400 (Fig. 5a) where an initial lag phase followed by markedly enhanced polymerization is apparent. The inclusion of Cd2+ at Cd:gelsolin ratios of 4:1–16:1 does not qualitatively alter this behavior, although complex effects independent of the ratio support the view that Cd2+ influences the nature of the interaction of gelsolin with actin. Cadmium ion itself can increase the rate of G-actin polymerization in the reconstituted system (Wang and Templeton, 1996), but only at concentrations much greater than achieved in the cell (Wang et al., 1996). Thus, it appears that the cellular context is required for Cd2+'s effects on actin. Other cellular proteins may influence the response of actin to gelsolin, and local compartmental effects may alter nucleation. Nevertheless, the effects of Cd2+ in the cell appear to involve gelsolin.
Cadmium concentrations in the 10 to 50 μM range are typical in cell culture studies, and we chose to use 10 μM in this study based on our earlier demonstration of positive effects on signaling (Ding and Templeton, 2000) and the cytoskeleton (Wang et al., 1996) with minimal toxicity in mesangial cells. How do these concentrations relate to realistic exposures in vivo Blood cadmium concentrations in an unexposed non-smoker are typically less than 0.01 μM. However, environmental exposures in industrial society have been reported to produce a renal cortical cadmium concentration of 35 μg/g wet wt (310 μM based on homogeneous distribution with a tissue density of 1.0) (Hotz et al., 1999). Of course much of the latter is buffered by cytosolic metallothionein, so we previously measured cytosolic ionic Cd2+ using fluorophores (Wang et al., 1996). Exposure of cultured mesangial cells to 0.1 μM CdCl2 for 4 h or 10 μM CdCl2 for 8 h produced comparable cytosolic concentrations of 1.2 and 1.5 pM, respectively. Therefore, the conditions used in the present study probably produce realistic intracellular concentrations of Cd2+ after buffering by metallothionein. Furthermore, while the timing of chronic occupational or environmental exposure cannot be reproduced in cultured cells, the use of the cultures allows a mechanistic insight into a phenomenon that may result in long-term cumulative damage from repetitive daily exposures.
We also noted that more severe disruption of actin filaments occurred at lower cell density. Although the reason for this is unclear, no difference in gelsolin immunofluorescence was apparent in cultures of different density, and a more general effect on toxicity is likely. We have observed a similar enhancement of iron toxicity in several lines of cultured cells at lower cell density (Z. Popovic and D. M. Templeton, unpublished). Possibly, cells proliferating at lower density are more susceptible to uptake or chemical damage, or perhaps some degree of protection is provided in more confluent cultures. Cell-cell contacts may decrease membrane permeability or stabilize cytoskeletal structures, and more confluent cultures may expose less surface area per cell to the medium. Whatever the explanation, the observation underscores the need to control cell density as one variable in mechanistic studies with cultured cells.
While the signal(s) effecting gelsolin redistribution in Cd2+-treated cells is (are) unclear, the Ca2+ dependence of gelsolin activation is well known. Binding of two Ca2+ ions at sites of moderate affinity (Kd in the μM range) is necessary for activation of actin binding (Weeds et al., 1988), and loss of one site by mutation in domain 2 leads to increased proteolysis and an amyloid-like syndrome (Kazmirski et al., 2002). Calcium binding produces large conformational changes that expose actin-binding sites (Gremm and Wegner, 1999; Hellweg et al., 1993). Gelsolin binds to both G-actin monomer and to actin filaments (Bryan, 1988; Gremm and Wegner, 1999; Pope et al., 1991), perhaps accounting, respectively, for nucleation and severing. Stepwise formation of gelsolin:actin monomer 1:1 and 1:2 complexes is cooperative and Ca2+-dependent, and chelation of Ca2+ favors dissociation to the 1:1 state (Gremm and Wegner, 1999). Thus, many of the effects of Cd2+ on the actin cytoskeleton may arise from effects on Ca2+-gelsolin interactions. Cadmium might activate gelsolin for F-actin binding and severing in the cell by substituting directly for Ca2+, or alternatively by displacing Ca2+ from other cellular sites and making it available to gelsolin. However, when G-actin is the predominant species in localized cell compartments, or as is the case in the in vitro polymerization assay, activation of gelsolin may on balance favor nucleation and polymerization. This may account for the persistence of actin filaments, decorated with gelsolin, in many Cd2+-treated cells. It is interesting to note that Cd2+ can substitute for Ca2+ in the gelsolin crystal structure at a Ca2+-binding site in domain 2 that is involved in gelsolin stabilization and perhaps activation (Kazmirski et al., 2002).
Our immunoblots and 125I-F-actin overlay experiments indicate a lower molecular mass species of gelsolin present in mesangial cells that is also present in the commercial gelsolin standard. This band associates prominently with the cytoskeletal fraction of Cd2+-treated cells, and corresponds to the approximately 45 kDa band reported in our earlier study (Wang and Templeton, 1996). Gelsolin cleavage may result from caspase-3 activation as shown by several groups (Azuma et al., 1998; Kothakota et al., 1997; Sun et al., 1999). Kothakota et al. (1997) have demonstrated that caspase-3 cleaves gelsolin within segment 3 at Asp352/Gly353 to generate peptides with apparent electrophoretic masses of 40 and 48 kDa. The N-terminal fragment is able to sever F-actin in a Ca2+-independent manner. Alternatively, conformational changes induced in the gelsolin molecule by Ca2+ unmask relatively specific tryptic cleavage sites that generate fragments of 70, 45, and 30 kDa (Khaitlina and Hinssen, 2002). Ca2+-dependent specific cleavage patterns of gelsolin have also been observed with plasmin (Wen et al., 1996). In light of these observations, Robinson et al. (2001) proposed that the fully activated gelsolin molecule is functionally susceptible to proteolysis. Therefore, gelsolin may be bound to the microfilament system as a combination of full-length gelsolin and one or more proteolysed fragments, with disruption of the microfilament system dependent upon all forms of gelsolin. Cadmium appears to favor cytoskeletal association of both full-length gelsolin and the 40 kDa peptide.
The intracellular distribution of gelsolin has long been a matter of debate (Carron et al., 1986). In fibroblasts it has been localized to the cell cortex and regions of adhesion. A homogeneous cytoplasmic localization of gelsolin has also been reported (Carron et al., 1986; Cooper et al., 1987), whereas in human gingival fibroblasts the distribution of gelsolin depends on their state of mobility. In migrant cells, it has been found diffusely distributed through the cytoplasm, whereas in non-migrating cells it has been localized along the stress fibers (Arora and McCulloch, 1996). Our results demonstrate that mesangial cell gelsolin alters its intracellular location from nucleus and cytosol in response to Cd2+ treatment. The view that gelsolin can actively participate in F-actin disruption and ruffle formation is supported by the demonstration that fibroblasts from gelsolin-null mice exhibit reduced membrane ruffling and an increased amount of F-actin (Azuma et al., 1998; Kwiatkowski, 1999; Witke et al., 1995). The highly dynamic nature of the ruffling process suggests that rapid F-actin assembly and disassembly are taking place, and this may be connected with increased gelsolin accumulation in the region of ruffles following Cd2+ treatment.
In summary, Cd2+ causes gelsolin to associate with the actin cytoskeleton, which is subsequently disrupted with filament loss and membrane ruffling. In polymerization assays, gelsolin present in extracts from Cd2+-treated cells acts as a nucleating factor for actin polymerization, and mixing Cd2+ with exogenous gelsolin does not diminish this effect. However, gelsolin also causes a large increase in the depolymerization rate of actin dissolved below its critical concentration. Although Cd2+ partially reverses this, depolymerization is still enhanced in the presence of gelsolin and Cd2+. It is apparent, then, that a major effect of Cd2+ in cultured cells is to cause a redistribution of gelsolin to the cytoskeleton, where its capping and severing properties predominate in the cellular context.
ACKNOWLEDGMENTS
This work was supported by an operating grant from the Canadian Institutes of Health Research. Conflict of interest: none declared.
REFERENCES
Arora, P. D., and McCulloch, C. A. (1996). Dependence of fibroblast migration on actin severing activity of gelsolin. J. Biol. Chem. 271, 20516–20523.
Ayscough, K. R. (1998). In vivo functions of actin binding proteins. Curr. Opin. Cell Biol. 10, 102–111.
Azuma, T., Witke, W., Stossel, T. P., Hartwig, J. H., and Kwiatkowski, D. J. (1998). Gelsolin is a downstream effector of rac for fibroblast motility. EMBO J. 17, 1362–1370.
Barrouillet, M. P., Moiret, A., and Cambar, J. (1999a). Protective effect of polyphenols against cadmium-induced glomerular mesangial cell myocontracture. Arch. Toxicol. 73, 485–488.
Barrouillet, M. P., Potier, M., and Cambar, J. (1999b). Cadmium nephrotoxicity assessed in isolated rat glomeruli and cultured mesangial cells: Evidence for contraction of glomerular cells. Exp. Nephrol. 7, 251–258.
Beyersmann, D., and Hechtenberg, S. (1997). Cadmium, gene regulation, and cellular signalling in mammalian cells. Toxicol. Appl. Pharmacol. 144, 247–261.
Bhattacharyya, M. H., Wilson, A. K., Rajan, S. S., and Jonah, M. (2000). Biochemical pathways in cadmium toxicity. In Molecular Biology and Toxicology of Metals (R. K. Zalups and J. Koropatnick, Eds.), pp. 34–74. Taylor and Francis, London.
Bryan, J. (1988). Gelsolin has three actin-binding sites. J. Cell Biol. 106, 1553–1562.
Bryan, J., and Coluccio, L. M. (1985). Kinetic analysis of F-actin depolymerization in the presence of platelet gelsolin and gelsolin complexes. J. Cell Biol. 101, 1236–1244.
Burtnick, L. D., Koepf, E. K., Grimes, J., Jones, E. Y., Stuart, D. I., McLaughlin, P. J., and Robinson, R. C. (1997). The crystal structure of plasma gelsolin: Implications for actin severing, capping, and nucleation. Cell 90, 661–670.
Carpenter, C. L. (2000). Actin cytoskeleton and cell signaling. Crit. Care Med. 28(Suppl. 4), N94–99.
Carron, C. P., Hwo, S. Y., Dingus, J., Benson, D. M., Meza, I., and Bryan, J. (1986). A re-evaluation of cytoplasmic gelsolin localization. J. Cell. Biol. 102, 237–245.
Cooper, J. A., Bryan, J., Schwab, B., III, Frieden, C., Loftus, D. J., and Elson, E. L. (1987). Microinjection of gelsolin into living cells. J. Cell. Biol. 104, 491–501.
Cunningham, C. C., Stossel, T. P., and Kwiatkowski, D. J. (1991). Enhanced motility in NIH-3T3 fibroblasts that overexpress gelsolin. Science 251, 1233–1236.
Ding, W., and Templeton, D. M. (2000). Activation of parallel mitogen-activated protein kinase cascades and induction of c-fos by cadmium. Toxicol. Appl. Pharmacol. 162, 93–99.
Ditsch, A., and Wegner, A. (1994). Nucleation of actin polymerization by gelsolin. Eur. J. Biochem. 224, 223–227.
Feinberg, J., Mery, J., Heitz, F., Benyamin, Y., and Roustan, C. (1997). Conformational and functional studies of three gelsolin subdomain-1 synthetic peptides and their implication in actin polymerization. Biopolymers 41, 647–655.
Fels, L. M., Bundschuh, I., Gwinner, W., Jung, K., Pergande, M., Graubaum, H.-J., Price, R. G., Taylor, S. A., De Broe, M. E., Nuyts, G. D., et al. (1994). Early urinary markers of target nephron segments as studied in cadmium toxicity. Kidney Int. 46, S81–S88.
Goering, P. L., Waalkes, M. P., and Klaassen, C. D. (1995). Toxicology of cadmium. In Handbook of Experimental Pharmacology, Vol. 115. Toxicology of Metals – Biochemical Aspects (R. A. Goyer and M. G. Cherian, Eds.), pp. 189–214. Springer-Verlag, Berlin.
Gremm, D., and Wegner, A. (1999). Co-operative binding of Ca2+ ions to the regulatory binding sites of gelsolin. Eur. J. Biochem. 262, 330–334.
Gremm, D., and Wegner, A. (2000). Gelsolin as a calcium-regulated actin filament-capping protein. Eur. J. Biochem. 267, 4339–4345.
Hellweg, T., Hinssen, H., and Eimer, W. (1993). The Ca2+-induced conformational change of gelsolin is located in the carboxy-terminal half of the molecule. Biophys. J. 65, 799–805.
Hotz, P., Buchet, J. P., Bernard, A., Lison, D., and Lauwerys, R. (1999). Renal effects of low-level environmental cadmium exposure: 5-year follow-up of a subcohort from the Cadmibel study. Lancet 354, 1508–1513.
Jrup, L. (2002). Cadmium overload and toxicity. Nephrol. Dial. Transplant 17(Suppl. 2), 35–39.
Jrup, L., Persson, B., Edling, C., and Elinder, C. G. (1993). Renal function impairment in workers previously exposed to cadmium. Nephron 64, 75–81.
Jrup, L., Persson, B., and Elinder, C. G. (1995). Decreased glomerular filtraion rate in solderers exposed to cadmium. Occup. Environ. Med. 52, 818–822.
Kazmirski, S. L., Isaacson, R. L., An, C., Buckle, A., Johnson, C. M., Daggett, V., and Fersht, A. R. (2002). Loss of a metal-binding site in gelsolin leads to familial amyloidosis–Finnish type. Nature Structural Biol. 9, 112–116.
Khaitlina, S., and Hinssen, H. (2002). Ca-dependent binding of actin to gelsolin. FEBS Lett. 521, 14–18.
Kinosian, H. J., Newman, J., Lincoln, B., Selden, L. A., Gershman, L. C., and Estes, J. E. (1998). Ca2+ regulation of gelsolin activity: binding and severing of F-actin. Biophys. J. 75, 3101–3109.
Kothakota, S., Azuma, T., Reinhard, C., Klippel, A., Tang, J., Chu, K., McGarry, T. J., Kirschner, M. W., Koths, K., Kwiatkowski, D. J., et al. (1997). Caspase-3-generated fragments of gelsolin: Effector of morphological change in apoptosis. Science 278, 294–298.
Kwiatkowski, D. J. (1999). Functions of gelsolin: motility, signaling, apoptosis, cancer. Current Opin. Cell Biol. 11, 103–108.
L'Azou, B., Dubus, I., Ohayon-Courtes, C., Labouyrie, J. P., Perez, L., Pouvreau, C., Juvet, L., and Cambar, J. (2002a). Cadmium induces direct morphological changes in mesangial cell culture. Toxicology 179, 233–245.
L'Azou, B., Henge-Napoli, M. H., Minaro, L., Mirto, H., Barrouillet, M. P., and Cambar, J. (2002b). Effects of cadmium and uranium on some in vitro renal targets. Cell Biol. Toxicol. 18, 505–516.
Lader, A. S., Kwiatkowski, D. J., and Cantiello, H. F. (1999). Role of gelsolin in the actin filament regulation of cardiac L-type calcium channels. Am. J. Physiol. 277, C1277–C1283.
Lagarrigue, E., Maciver, S. K., Fattoum, A., Benyamin, Y., and Roustan, C. (2003). Co-operation of domain-binding and calcium-binding sites in the activation of gelsolin. Eur. J. Biochem. 270, 2236–2243.
Luna, E. J. (1998). F-actin blot overlays. Methods Enzymol. 298, 32–42.
Mazzei, G. J., Girard, P. R., and Kuo, J. F. (1984). Environmental pollutant Cd2+ biphasically and differentially regulates myosin light chain kinase and phospholipid/Ca2+-dependent protein kinase. FEBS Lett. 173, 124–128.
Nordberg, G. F., Herber, R. F. M., and Alessio, L. (Eds.) (1992). Cadmium in the Human Environment: Toxicity and Carcinogenicity. International Agency for Research on Cancer, Lyon.
Pope, B., Way, M., and Weeds, A. G. (1991). Two of the three actin-binding domains of gelsolin bind to the same subdomain of actin. Implications of capping and severing mechanisms. FEBS Lett. 280, 70–74.
Robinson, R. C., Choe, S., and Burtnick, L. D. (2001). The disintegration of a molecule: The role of gelsolin in FAF, familial amyloidosis (Finnish type). Proc. Natl. Acad. Sci. U.S.A. 98, 2117–2118.
Roels, H., Bernard, A. M., Cardenas, A., Buchet, J. P., Lauwerys, R. R., Hotter, G., Ramis, I., Mutti, A., Franchini, I., Stolte, H., et al. (1993). Markers of early renal changes induced by industrial pollutants. Application to workers exposed to cadmium. Br. J. Ind. Med. 50, 37–48.
Selden, L. A., Kinosian, H. J., Estes, J. E., and Gershman, L. C. (2000). Cross-linked dimers with nucleating activity in actin prepared from muscle acetone powder. Biochemistry 39, 64–74.
Stournaras, C. (1990). Exposure of thiol groups and bound nucleotide in G-actin: Thiols as an indicator of the native state of actin. Anticancer Res. 10, 1651–1659.
Sun, H. Q., Yamamoto, M., Mejillano, M., and Yin, H. (1999). Gelsolin, a multifunctional actin regulatory protein. J. Biol. Chem. 274, 33179–33182.
Suzuki, Y., Chao, S.-H., Zysk, J. R., and Cheung, W. Y. (1985). Stimulation of calmodulin by cadmium ion. Arch. Toxicol. 57, 205–211.
Templeton, D. M. (2000). Metal ions and the cytoskeleton. In Molecular Biology and Toxicology of Metals (R. F. Zalups and J. Loropatnick, Eds.), pp. 460–476. Taylor and Francis, London.
Valentin-Ranc, C., and Carlier, M.-F. (1991). Role of ATP-bound divalent metal ion in the conformation and function of actin. Comparison of Mg-ATP, Ca-ATP, and metal ion-free ATP-actin. J. Biol. Chem. 266, 7668–7675.
Vandekerchove, J. (1993). Actins. In Guidebook to the Cytoskeletal and Motor Proteins (T. Kreis and R. Vale, Eds.), pp. 13–15. Oxford University Press, Oxford.
Waalkes, M. P., Coogan, T. P., and Barter, R. A. (1992). Toxicological principles of metal carcinogenesis with special emphasis on cadmium. Crit. Rev. Toxicol. 22, 175–201.
Wang, A., Fan, M.-Y., and Templeton, D. M. (1994). Growth modulation and proteoglycan turnover in cultured mesangial cells. J. Cell. Physiol. 159, 295–310.
Wang, Z., Chin, T. A., and Templeton, D. M. (1996). Calcium-independent effects of cadmium on actin assembly in mesangial and vascular smooth muscle cells. Cell Motil. Cytoskeleton 32, 208–222.
Wang, Z., and Templeton, D. M. (1996). Cellular factors mediate cadmium-dependent actin depolymerization. Toxicol. Appl. Pharmacol. 139, 115–121.
Weeds, A. G., Gooch, J., Pope, B., and Harris, H. E. (1988). Preparation and characterization of pig plasma and platelet gelsolins. Eur. J. Biochem. 161, 69–76.
Welch, M. D., and Mullins, R. D. (2002). Cellular control of actin nucleation. Annu. Rev. Cell Dev. Biol. 18, 247–288.
Wen, D., Corina, K., Chow, E. P., Miller, S., Janmey, P. A., and Pepinsky, R. B. (1996). The plasma and cytoplasmic forms of human gelsolin differ in disulfide structure. Biochemistry 35, 9700–9709.
Witke, W., Sharpe, A. H., Hartwig, J. H., Azuma, T., Stossel, T. P., and Kwiatkowski, D. J. (1995). Hematostatic, inflammatory, and fibroblast responses are blunted in mice lacking gelsolin. Cell 81, 41–51.
Yin, H. N., and Stossel, T. P. (1979). Control of cytoplasmic gel-sol transition by gelsolin a calcium-dependent regulatory protein. Nature 281, 583–586.(Margarita D. Apostolova, Tania Christova)
ABSTRACT
Cadmium (Cd2+) is known to cause a selective disruption of the filamentous actin cytoskeleton in the smooth muscle-like renal mesangial cell. We examined the effect of Cd2+ on the distribution of the actin-severing protein, gelsolin. Over 8 h, CdCl2 (10 μM) caused a progressive shift of gelsolin from a diffuse perinuclear and cytoplasmic distribution to a pattern decorating F-actin filaments. Over this time filaments were decreased in number in many cells, and membrane ruffling was initiated. Western blotting and 125I-F-actin gel overlays demonstrated an increase in actin-binding gelsolin activity in the cytoskeletal fraction of cell extracts following Cd2+ treatment. In in vitro polymerization assays, gelsolin acted as a nucleating factor and increased the rate of polymerization. Cytosolic extracts also increased the polymerization rate. Addition of Cd2+ together with gelsolin further increased the rate of polymerization. Gelsolin enhanced depolymerization of purified actin, and Cd2+ partially suppressed this effect. However, cytoskeletal extracts from Cd2+-treated cells also markedly increased depolymerization, suggesting further that Cd2+ may activate cellular component(s) such as gelsolin for actin binding. We conclude that a major effect of Cd2+ on the mesangial cell cytoskeleton is manifest through activating the association of gelsolin with actin, with gelsolin's severing properties predominating under conditions found in Cd2+-treated cells.
Key Words: F-actin; actin depolymerization; cadmium toxicity.
INTRODUCTION
Cadmium (Cd2+) is a divalent metal ion that exhibits potent toxicity to most cells and organisms (Bhattacharyya et al., 2000; Goering et al., 1995; Nordberg et al., 1992). While the proximal tubule is a major target of Cd2+ toxicity, damage to the glomerulus is also significant (Jrup, 2002). Workers exposed to Cd2+ show urinary markers characteristic of glomerular injury (Fels et al., 1994; Roels et al., 1993). Occupational exposure to Cd2+ is associated with glomerular injury and decreased glomerular filtration rate (Jrup et al., 1993, 1995). The mesangial cell is a contractile smooth muscle-like cell of the glomerulus involved in regulating the glomerular filtration rate. Cadmium causes rearrangement of the actin cytoskeleton in cultured mesangial cells (Wang et al., 1996) and stimulates them to contract (Barrouillet et al., 1999a,b; L'Azou et al., 2002a,b). It has been suggested that Cd2+-induced mesangial cell contraction underlies the changes in intrarenal hemodynamics and glomerular filtration rate that precede proteinuria (Barrouillet et al., 1999b; Jrup et al., 1993, 1995; Roels et al., 1993). The contribution of cytoskeletal disruption has not been studied in detail.
Previously we showed that exposure of mesangial cells to Cd2+ was accompanied by a loss of filamentous (F-) actin structure and organization (Wang et al., 1996; Wang and Templeton, 1996). Exposure of mesangial cells to Cd2+ resulted in F-actin depolymerization without a compensatory increase in cellular G-actin levels (Wang et al., 1996). These smooth muscle-like cells showed some specificity for this effect of Cd2+, a number of other divalent metals having no effect on F-actin at the same concentration as Cd2+. The cytoskeleton of smooth muscle cells in general seems to be particularly susceptible to disruption by Cd2+ (Templeton, 2000). The polymerization of F-actin from monomeric, globular (G-) actin subunits is regulated by a myriad of proteins, many of them Ca2+-dependent (Ayscough, 1998; Carpenter, 2000; Vandekerchove, 1993; Welch and Mullins, 2002). Further, the redox status of the cell, reflected in part in the thiol content of its F-actin cytoskeleton, is another determinant of cytoskeletal structure (Stournaras, 1990; Valentin-Ranc and Carlier, 1991).
The biological interactions of Cd2+ are dominated by two over-riding chemical features. First, its strong propensity for sulfur as a ligand disrupts the structure and function of a number of proteins and enzymes (Beyersmann and Hechtenberg, 1997; Bhattacharyya et al., 2000 Goering et al., 1995; Waalkes et al., 1992) by poisoning protein thiol groups. Second, it has an ionic radius (0.97 ) very close to that of Ca2+ (0.99 ), and so potentially can interfere with Ca2+-dependent signaling processes; it can substitute for Ca2+ in vitro, activating Ca2+-dependent forms of protein kinase C and supporting calmodulin signaling (Mazzei et al., 1984; Suzuki et al., 1985). Thus, because of its interactions with thiol groups and substitution for Ca2+, it is not unexpected that Cd2+ will influence cytoskeletal integrity. The depolymerizing effects of Cd2+ on F-actin were, though, found to be independent of its effects on cytosolic [Ca2+], and independent of direct effects on Ca2+-binding proteins (Wang et al., 1996). In fact, Cd2+ stimulated actin polymerization in a reconstituted in vitro assay system. However, extracts made from cells that had been treated with Cd2+ mimicked the depolymerizing effects of Cd2+ on actin in cultured cells (Wang and Templeton, 1996). And, exposing mesangial cells to Cd2+ influenced actin-binding activity of protein(s) with molecular weights of approximately 90 kDa and 45 kDa, differentially in cytosolic and cytoskeletal-associated fractions. The indication is that Cd2+ does not act directly on the cytoskeleton, but affects the expression and (or) activity of proteins that may regulate the status of F-actin.
Gelsolin represents a family of ca. 90 kDa monomeric actin-binding proteins present in most animal cells, that sever F-actin and cap the quickly growing barbed end, thus favouring depolymerization from the pointed end (Ayscough, 1998; Sun et al., 1999; Yin and Stossel, 1979). Cardiac myocytes from gelsolin -/- mice show increased F-actin bundling of stress fibers in culture (Lader et al., 1999), whereas overexpression of gelsolin enhances cytoskeletal reorganization and fibroblast motility (Cunningham et al., 1991). Gelsolin binding to actin is regulated by Ca2+ and polyphosphoinositide-4,5-bisphosphate (Gremm and Wegner, 2000; Kwiatkowski, 1999; Sun et al., 1999). At sub-micromolar Ca2+ concentrations, Ca2+-regulated capping and severing activities can occur independently of one another (Bryan and Coluccio, 1985; Gremm and Wegner, 2000). Gelsolin can also bind to G-actin monomers (Gremm and Wegner, 1999, 2000) and may facilitate nucleation in some circumstances.
We have pursued effects of Cd2+ on the mesangial cell cytoskeleton in order both to elucidate potential mechanisms of cadmium toxicity and to gain further insight into features controlling the F-actin–G-actin equilibrium in these cells. The present study was undertaken to evaluate the effects of Cd2+ on gelsolin expression and binding in mesangial cells, and in particular to evaluate gelsolin's involvement in Cd2+-induced disruption of cellular F-actin content.
MATERIALS AND METHODS
Cell culture conditions and treatment.
Rat mesangial cells were isolated as described previously (Wang et al., 1994) and were grown in RPMI 1640 medium without antibiotics, supplemented with 10% (v/v) fetal bovine serum, in a humidified atmosphere of 5% CO2 at 37°C. Cells growing on 10-cm dishes at 70% confluence, or on cover slips, were made quiescent by serum deprivation (0.2% serum for 48 h). Quiescent cells were treated with 10 μM CdCl2 in RPMI 1640 for up to 8 h. All experiments were performed with cells between passages 5 and 20. Cytosolic and cytoskeletal-enriched fractions were prepared by lysis of cells in actin polymerization buffer (see below) containing 0.2% Triton X-100 as described previously (Wang and Templeton, 1996).
Actin purification.
Rabbit skeletal muscle actin was prepared according to Selden et al. (2000). Briefly, G-actin was extracted from skeletal muscle acetone powder (prepared as described previously; Wang and Templeton, 1996) with buffer G (2 mM Tris-HCl, pH 8.0 at 25°C, with 0.1 mM CaCl2 and 0.5 mM dithiothreitol [DTT]) containing 0.2 mM ATP. The mixture was centrifuged at 10,000 x g for 20 min and the supernatant was further centrifuged at 150,000 x g for 30 min at 4°C. The supernatant was polymerized by addition of 3 M KCl to a final concentration of 0.8 M and left over night at 4°C with very slow stirring. The solution was centrifuged at 150,000 x g for 2 h and the F-actin pellet was soaked in buffer G for 2 h on ice. Pellets were picked up with a glass rod, homogenized carefully in a Dounce homogenizer, and dialysed against 500 ml of buffer G for 48 h. The dialysed solution was centrifuged at 150,000 x g for 2 h at 4°C. The polymerization step was repeated twice and G-actin concentration was determined spectrophotometrically at 290 nm using an absorption coefficient of = 26,600 M–1 cm–1. G-actin was further purified by gel filtration on a 2.6 x 100 cm Sephadex G-150 column pre-equilibrated with buffer G. G-actin (5 mg/ml) was eluted at a flow rate of 1 ml/min. Five-milliliter fractions were collected and the absorption at 280 nm was monitored. Fractions corresponding to G-actin (Mr = 43 kD) were pooled and stored at 4°C for up to 2 weeks.
Immunoblotting and immunoprecipitation.
Mesangial cell extracts were prepared by harvesting cells at the indicated times after addition of 10 μM CdCl2 and lysis in buffer A (20 mM Tris-HCl, pH 7.4, with 2 mM MgCl2, 138 mM KCl, 1 mM ATP, and 0.2 % [v/v] Triton X-100). The lysates were cleared by centrifugation (10,000 x g, 15 min) and the pellets were resuspended in buffer A without Triton X-100, and either used immediately or stored at –80°C. SDS-PAGE was performed on (4–20)% gradient gels. Proteins were transferred to polyvinylidene difluoride (PVDF) membranes, and probed with polyclonal anti-gelsolin antibody (C-20) and secondary horseradish peroxidase-conjugated anti-goat IgG, both from Santa Cruz Biotechnology (Santa Cruz, CA). Bands were visualized by enhanced chemiluminescence detection kit (ECL+, Amersham) according to the manufacturer's instructions. For immunoprecipitation, mesangial cells were treated with CdCl2 and rinsed with ice-cold PBS, harvested, and lysed in immunoprecipitation buffer C (buffer A + 2 μg/ml aprotinin, 2 μg/ml leupeptin, 1 mM phenylmethylsulfonyl fluoride, 10 mM NaF, 1 mM Na3VO4). Samples containing 500 μg protein were incubated overnight at 4°C with anti-gelsolin antibody (1 μg/sample). Immune complexes were isolated by the addition of protein G Sepharose (2 h, room temperature) and washed five times with buffer C. Proteins bound to the Sepharose beads were released by boiling in SDS-PAGE sample buffer for 5 min, resolved by gradient SDS-PAGE (4–20%), transferred to PVDF, blocked with Membrane blocking agent (Amersham), and probed with anti-gelsolin antibody or 125I-F-actin.
Indirect immunofluorescence and confocal microscopy.
Mesangial cells treated as described above were washed twice in PBS and fixed with 3.8% para-formaldehyde (4°C, 40 min). After two washes with PBS, cells were permeabilized with 0.2% Triton X-100 in PBS and blocked in PBS containing 5% bovine serum albumin for 1 h at room temperature, followed by a further 1 h incubation with anti-gelsolin antibody. After three 10-min washes with PBS, cells were exposed for 1 h to secondary antibody conjugated to fluorescein isothiocyanate (FI-5000; Vector Laboratories, Burlingame, CA). Cells were rinsed three times with PBS and incubated with Alexa Fluor 568-labeled phalloidin (Molecular Probes; Eugene, OR), following the manufacturer's instructions. Following six washes in PBS samples were mounted in 4',6-diamidino-2-phenylindol dihydrochloride (DAPI) containing Vectashield mount medium (Vector Laboratories). Photomicrographs were obtained with a Nikon Eclipse E-600 microscope equipped with a Hamamatsu digital camera (Hamamatsu Inc., Japan), using Simple PCI software (Compix Inc., PA). Confocal microscopy was performed with a Zeiss LSM 410 inverted laser scanning microscope equipped with flourescein and rhodamine filters and a Zeiss 63 x 1.40 Plan-Apochromat oil immersion lens. Image analysis was performed with ImageJ 1.24t software (National Institutes of Health, U.S.). The results are representative of at least three independent experiments conducted in triplicate.
125I-F-actin preparation.
G-actin (1 mg/ml) prepared as described above was labeled with 250 μCi 125I (carrier-free NaI; ICN Pharmaceuticals, Costa Mesa, CA) in the presence of an Iodo-Bead (Pierce Biotechnology, Rockford, IL). Radiolabeled G-actin was separated from unincorporated 125I by gel filtration at room temperature through a 5 ml PD-10 column (Amersham Bioscience, Uppsala, Sweden) equilibrated with buffer B (2 mM Tris-HCl, pH 8.3, with 0.2 mM CaCl2, 0.2 mM ATP, and 0.5 mM DTT) at 0°C. 125I-G-actin was polymerized by gentle mixing with 0.25 volume of 100 mM PIPES-NaOH buffer, pH 7.0, containing 250 mM KCl, 10 mM MgCl2, and 0.25 mM CaCl2 at room temperature for 1 h. 125I-F-actin was collected by centrifugation at 150,000 x g for 2 h at 4°C. The pellet was resuspended in 500 μl of buffer B and phalloidin was added to give a final concentration of 40 μM. The mixture was then incubated for another 15 min at room temperature and the phalloidin-stabilized 125I-F-actin was stored at 4°C.
125I-F-actin overlay-assay.
125I-F-actin overlay assays were performed essentially as described (Luna, 1998). The blocked PVDF membranes were incubated overnight without disturbance with 50–100 μg/ml 125I-F-actin stabilized with 40 μM phalloidin. PVDF membranes were washed five times for 2 min with TBST (10 mM Tris-HCl, pH 7.5, with 90 mM NaCl and 0.5% [v/v] Tween-20) and exposed to film at –80°C for 2–24 h.
Pyrene labeling of actin.
G-actin (3 mg/ml in buffer G) was dialysed against buffer G (with 0.2 mM ATP but without DTT) for 48 h at 4°C. Pyrene iodoacetamide (Molecular Probes, Eugene, OR) was dissolved in dimethylsulfoxide and added to G-actin at a 10:1 molar ratio. The mixture was left overnight with continuous stirring at 4°C in the dark. G-actin-pyrene was then dialysed against buffer G for 48 h at 4°C. To remove unreacted excess pyrene, the mixture was chromatographed on a PD-10 Sephadex column and 500 μl fractions were collected after elution with buffer G. The concentration of labeled G-actin was calculated from the absorption at 290 nm after correction for pyrene absorption at 344 nm by subtracting 0.127 A344 from the measured A290. Concentration ratios [pyrene]/[actin] were calculated using 344 = 2.22 x 104 M–1 cm–1 for pyrene. The molar ratio of pyrene to actin monomer in the labeled product was between 0.7 and 1.0 in all experiments.
Polymerization and depolymerization assays.
Kinetic measurements following the changes in pyrene-labeled actin fluorescence were initiated in polymerization buffer (20 mM Tris-HCl, pH 7.4, containing 2 mM MgCl2, 138 mM KCl, and 1 mM ATP). Pyrene-G-actin (2 μM) was used in all experiments which followed polymerization rates, and for the initiation of depolymerization pyrene-F-actin was initially diluted to 200 nM in the same polymerization buffer. To study the effects of Cd2+ and gelsolin, pyrene-labeled actin was added to polymerization buffer containing various concentrations of Cd2+ and/or gelsolin, and the fluorescence intensity of pyrene (excitation at 365 nm, emission at 385 nm) was recorded for up to 10 min on a Fluorolog spectrofluorimeter (J Y Horiba, Edison, NJ). The conditions for actin polymerization and depolymerization in the presence of cytosolic and cytoskeletal-enriched fractions (5 μg/μl final protein concentration) were as described previously (Wang and Templeton, 1996). All measurements were performed at 37°C.
RESULTS
Immunohistochemical Localization of Gelsolin
Cells were seeded on a glass surface, starved for 48 h in 0.2% serum, and then treated with 10 μM CdCl2 for up to 8 h. Longer exposures and higher concentrations result in some cell death with loss of filaments in all cells (not shown; see Wang et al., 1996), whereas shorter exposures resulted in less cytoskeletal disruption. Fluorescence photomicrographs of phalloidin-stained cells (Figs. 1a and 1b) show a disruption of actin filaments in many cells after Cd2+ treatment, confirming earlier observations (Wang et al., 1996). The response is heterogeneous at a given time point, with not all cells showing loss of filaments by 8 h, and we observed that this was dependent on cell density. In a typical experiment, cultures plated at 104 cells cm–2 (as in Fig. 1) showed significant disruption of filaments in 40% of cells (n = 293) while at 105 cells cm–2 only 18% (n = 356) were affected. Based on these observations, we examined further the distribution of gelsolin in cells prior to Cd2+ treatment, and after 8 h of Cd2+ exposure in cells with and without a remaining F-actin cytoskeleton.
Gelsolin distributes in a diffuse, punctate pattern of staining throughout the cytoplasm and especially associated with the nucleus of control cells not exposed to Cd2+ (Fig. 1c). Following 2 h of Cd2+ exposure, cells showed prominent staining for gelsolin in the cytoplasm and evidence of some filamentous arrangement (Fig. 1d). After 8 h of Cd2+ treatment, gelsolin was predominantly localized along any remaining F-actin filaments (Fig. 1e). In cells in which the F-actin filaments remained more or less intact after 8 h of Cd2+, strong colocalization of gelsolin with actin was observed (Figs. 1f–1h). Scanning across magnified fields of confocal images of cytosolic regions (Figs. 1i and 1j) confirmed a shift of gelsolin from a diffuse distribution before Cd2+ treatment, to take on a periodicity matching that of F-actin at 8 h. Images and analyses in Figures 1c–1j are representative of multiple fields from at least 10 different experiments. (Note that panels i and j are produced from the high resolution confocal images in panels c and e, and are magnified to allow visualization of the pixelated images.) The percentage of the cells showing localization of gelsolin to F-actin fibers steadily increased with the time of Cd2+ treatment (not shown). These observations suggest that gelsolin redistribution from the nucleus and cytoplasm to the actin cytoskeleton can be triggered by Cd2+ treatment. In addition, Cd2+ treatment and accompanying stress fiber reorganization induces areas of membrane ruffling, or lamellipodia, rich in gelsolin (Fig. 2). These are seen rarely in control cells (an example is shown in Fig. 2a), but are frequent in Cd2+-treated cells (Fig. 2b). Gelsolin localization to the membrane ruffles was detected in over 60% of the cells after 8 h of Cd2+exposure, where it colocalized with actin filaments (Fig. 2c).
Western Blotting and Gel Overlays
To confirm the Cd2+-dependent redistribution of gelsolin binding activity, cytosolic and cytoskeletal fractions from control and Cd-treated cells were subjected to Western blotting. Two actin-binding protein bands with Mr values of about 90 and 40 kDa were detected (Fig. 3). The band with Mr = 90 kD corresponds to full-length gelsolin and the 40 kDa band may be attributed to a cleaved gelsolin peptide (see Discussion) recognized by the antibody against a C-terminal peptide; this lower band is also present in a purified gelsolin standard. Although no change in total gelsolin mRNA in whole cell extracts was found (data not shown), the intensity of both gelsolin protein bands increased in the cytoskeletal fraction in a time-dependent manner following Cd2+ treatment, consistent with the results of immunofluorescence. The putative cleaved fragment at 40 kDa was absent from cytosol before Cd2+ treatment but was apparent in this fraction at and after 4 h.
To demonstrate that these bands represented actin-binding activity, cell fractions were electrophoresed and blotted with [125I]-F-actin. Despite the high background inherent in this technique, at least five actin-binding bands were identified (numbered 1–5, Fig. 4). Two of these (bands 2 and 5) were prominent in the gelsolin standard, representing intact gelsolin (band 2) and probably a cleaved fragment containing the C-terminal epitope (band 5). The nature of the other bands was not pursued in this study, including the very prominent band 4 which is restricted to the cytoskeletal fraction and increases markedly with Cd2+ treatment. Immunoprecipitates with anti-gelsolin antibody were prepared to confirm the identity of bands 2 and 5. [125I]-F-actin overlays of gelsolin immunoprecipitates showed only bands 2 and 5 (Fig. 4c). These bands show little or no change in binding activity in the cytosolic fraction after 8 h of treatment with Cd2+. However, actin-binding activity of gelsolin, especially of the 40 kDa fragment, increases in the cytoskeletal fraction after Cd2+ exposure, consistent with the results of immunofluorescence and Western blotting.
Polymerization/Depolymerization Studies
We previously showed that Cd2+ both facilitates polymerization and suppresses depolymerization of actin in an in vitro assay (Wang and Templeton, 1996), both counter to the loss of actin filaments from the cell. Therefore, direct interaction of Cd2+ with F-actin/G-actin is unlikely to account for Cd2+-induced loss of filaments in vivo. However, the above results raise the possibility that Cd2+ may have different effects when gelsolin is present, and we therefore studied the polymerization and depolymerization of purified actin in the presence of Cd2+ and gelsolin together. Pyrene-labeled G-actin was dissolved in polymerization buffer at 2 μM and polymerization was initiated by the addition of ATP, resulting in an apparent first-order rate of increase in fluorescence intensity (Fig. 5a). When gelsolin was included in the reaction solution, a sigmoidal polymerization curve was observed, representing a lag phase indicative of a nucleation process followed by a linear phase of polymerization at a markedly increased rate. We confirmed our previous observation (Wang et al., 1996) that inclusion of Cd2+ in the polymerization mixture increased the rate of polymerization at concentrations in the 200 μM range, though concentrations of 10–100 nM were without significant effect in the absence of gelsolin (not shown). When low concentrations of Cd2+ (20–80 nM) were added to a polymerization mixture containing 5 nM gelsolin, there was an indication that the lag phase was decreased and the rate of polymerization was unaffected or increased (Fig. 5a), although not in any apparent concentration-dependent manner. These results are presented to demonstrate that neither Cd2+ nor gelsolin, alone or in combination, drive depolymerization of the filaments.
To study depolymerization, pyrene-labeled F-actin was diluted to 200 nM (based on the G-actin subunit) by the addition of buffer. This is below the critical concentration for polymerization, and spontaneous depolymerization was observed by fluorescence quenching. Depolymerization was enhanced by inclusion of gelsolin in the dilution buffer (Fig. 5b), and increased with increasing gelsolin:actin ratio (not shown), consistent with well documented capping at the barbed end and dissociation from the pointed end. Inclusion of Cd2+ in the gelsolin mixture prior to addition of actin slowed the rate of depolymerization in a concentration-dependent manner, though even at a ratio of Cd2+:gelsolin of 10:1 depolymerization was still more rapid than in the absence of gelsolin.
Because Cd2+ exposure facilitated a localization of gelsolin with the cytoskeleton in cultured cells, cytosolic and cytoskeletal cell extracts from control and Cd2+-treated cells were examined for their ability to affect depolymerization of purified F-actin. Cytosolic extracts from control cells did not affect the spontaneous depolymerization of actin below its critical concentration (Fig. 6a). Cytoskeletal extracts from control cells increased the rate of depolymerization (Fig. 6b), consistent with the presence of protein(s) in this fraction that facilitate depolymerization. When obtained from Cd2+-treated cells, both cytosolic and cytoskeletal fractions increased the rate of depolymerization.
DISCUSSION
Rat mesangial cells have previously been shown to undergo disruption of F-actin after treatment with Cd2+ (Barrouillet et al., 1999b; L'Azou et al., 2002b; Wang et al., 1996; Wang and Templeton, 1996). The present study corroborates these findings and shows that Cd2+ treatment results in F-actin-gelsolin interactions that accompany changes in F-actin stress fiber organization and distribution, cell-cell contacts, and organization of membrane ruffles. A redistribution of gelsolin into the detergent-insoluble cytoskeletal fraction on Western blots and immunolocalization of gelsolin to the actin cytoskeleton in those cells where actin filaments remained, indicates that Cd2+ treatment favors association of gelsolin with actin filaments. This association is seen as early as 2 h after exposure to Cd2+ and precedes gross actin disruption; gelsolin remains associated with residual shortened actin filaments throughout the process, leading to the eventual disappearance of many fibers by 8 h. Thus, gelsolin is a leading candidate for mediating the effects of Cd2+ on the cytoskeleton.
Disruption of F-actin by enhanced association with gelsolin is consistent with the well-described capping and/or severing properties of gelsolin (Ayscough, 1998; Gremm and Wegner, 2000; Kinosian et al., 1998; Lagarrigue et al., 2003; Sun et al., 1999). Nevertheless, kinetic analysis indicates that complex mechanisms are involved. Thus, gelsolin is also an effective nucleating factor for G-actin polymerization under some circumstances (Burtnick et al., 1997; Ditsch and Wegner, 1994; Feinberg et al., 1997). This effect dominates in the assay with purified actin, e.g., at a gelsolin:G-actin ratio of 1:400 (Fig. 5a) where an initial lag phase followed by markedly enhanced polymerization is apparent. The inclusion of Cd2+ at Cd:gelsolin ratios of 4:1–16:1 does not qualitatively alter this behavior, although complex effects independent of the ratio support the view that Cd2+ influences the nature of the interaction of gelsolin with actin. Cadmium ion itself can increase the rate of G-actin polymerization in the reconstituted system (Wang and Templeton, 1996), but only at concentrations much greater than achieved in the cell (Wang et al., 1996). Thus, it appears that the cellular context is required for Cd2+'s effects on actin. Other cellular proteins may influence the response of actin to gelsolin, and local compartmental effects may alter nucleation. Nevertheless, the effects of Cd2+ in the cell appear to involve gelsolin.
Cadmium concentrations in the 10 to 50 μM range are typical in cell culture studies, and we chose to use 10 μM in this study based on our earlier demonstration of positive effects on signaling (Ding and Templeton, 2000) and the cytoskeleton (Wang et al., 1996) with minimal toxicity in mesangial cells. How do these concentrations relate to realistic exposures in vivo Blood cadmium concentrations in an unexposed non-smoker are typically less than 0.01 μM. However, environmental exposures in industrial society have been reported to produce a renal cortical cadmium concentration of 35 μg/g wet wt (310 μM based on homogeneous distribution with a tissue density of 1.0) (Hotz et al., 1999). Of course much of the latter is buffered by cytosolic metallothionein, so we previously measured cytosolic ionic Cd2+ using fluorophores (Wang et al., 1996). Exposure of cultured mesangial cells to 0.1 μM CdCl2 for 4 h or 10 μM CdCl2 for 8 h produced comparable cytosolic concentrations of 1.2 and 1.5 pM, respectively. Therefore, the conditions used in the present study probably produce realistic intracellular concentrations of Cd2+ after buffering by metallothionein. Furthermore, while the timing of chronic occupational or environmental exposure cannot be reproduced in cultured cells, the use of the cultures allows a mechanistic insight into a phenomenon that may result in long-term cumulative damage from repetitive daily exposures.
We also noted that more severe disruption of actin filaments occurred at lower cell density. Although the reason for this is unclear, no difference in gelsolin immunofluorescence was apparent in cultures of different density, and a more general effect on toxicity is likely. We have observed a similar enhancement of iron toxicity in several lines of cultured cells at lower cell density (Z. Popovic and D. M. Templeton, unpublished). Possibly, cells proliferating at lower density are more susceptible to uptake or chemical damage, or perhaps some degree of protection is provided in more confluent cultures. Cell-cell contacts may decrease membrane permeability or stabilize cytoskeletal structures, and more confluent cultures may expose less surface area per cell to the medium. Whatever the explanation, the observation underscores the need to control cell density as one variable in mechanistic studies with cultured cells.
While the signal(s) effecting gelsolin redistribution in Cd2+-treated cells is (are) unclear, the Ca2+ dependence of gelsolin activation is well known. Binding of two Ca2+ ions at sites of moderate affinity (Kd in the μM range) is necessary for activation of actin binding (Weeds et al., 1988), and loss of one site by mutation in domain 2 leads to increased proteolysis and an amyloid-like syndrome (Kazmirski et al., 2002). Calcium binding produces large conformational changes that expose actin-binding sites (Gremm and Wegner, 1999; Hellweg et al., 1993). Gelsolin binds to both G-actin monomer and to actin filaments (Bryan, 1988; Gremm and Wegner, 1999; Pope et al., 1991), perhaps accounting, respectively, for nucleation and severing. Stepwise formation of gelsolin:actin monomer 1:1 and 1:2 complexes is cooperative and Ca2+-dependent, and chelation of Ca2+ favors dissociation to the 1:1 state (Gremm and Wegner, 1999). Thus, many of the effects of Cd2+ on the actin cytoskeleton may arise from effects on Ca2+-gelsolin interactions. Cadmium might activate gelsolin for F-actin binding and severing in the cell by substituting directly for Ca2+, or alternatively by displacing Ca2+ from other cellular sites and making it available to gelsolin. However, when G-actin is the predominant species in localized cell compartments, or as is the case in the in vitro polymerization assay, activation of gelsolin may on balance favor nucleation and polymerization. This may account for the persistence of actin filaments, decorated with gelsolin, in many Cd2+-treated cells. It is interesting to note that Cd2+ can substitute for Ca2+ in the gelsolin crystal structure at a Ca2+-binding site in domain 2 that is involved in gelsolin stabilization and perhaps activation (Kazmirski et al., 2002).
Our immunoblots and 125I-F-actin overlay experiments indicate a lower molecular mass species of gelsolin present in mesangial cells that is also present in the commercial gelsolin standard. This band associates prominently with the cytoskeletal fraction of Cd2+-treated cells, and corresponds to the approximately 45 kDa band reported in our earlier study (Wang and Templeton, 1996). Gelsolin cleavage may result from caspase-3 activation as shown by several groups (Azuma et al., 1998; Kothakota et al., 1997; Sun et al., 1999). Kothakota et al. (1997) have demonstrated that caspase-3 cleaves gelsolin within segment 3 at Asp352/Gly353 to generate peptides with apparent electrophoretic masses of 40 and 48 kDa. The N-terminal fragment is able to sever F-actin in a Ca2+-independent manner. Alternatively, conformational changes induced in the gelsolin molecule by Ca2+ unmask relatively specific tryptic cleavage sites that generate fragments of 70, 45, and 30 kDa (Khaitlina and Hinssen, 2002). Ca2+-dependent specific cleavage patterns of gelsolin have also been observed with plasmin (Wen et al., 1996). In light of these observations, Robinson et al. (2001) proposed that the fully activated gelsolin molecule is functionally susceptible to proteolysis. Therefore, gelsolin may be bound to the microfilament system as a combination of full-length gelsolin and one or more proteolysed fragments, with disruption of the microfilament system dependent upon all forms of gelsolin. Cadmium appears to favor cytoskeletal association of both full-length gelsolin and the 40 kDa peptide.
The intracellular distribution of gelsolin has long been a matter of debate (Carron et al., 1986). In fibroblasts it has been localized to the cell cortex and regions of adhesion. A homogeneous cytoplasmic localization of gelsolin has also been reported (Carron et al., 1986; Cooper et al., 1987), whereas in human gingival fibroblasts the distribution of gelsolin depends on their state of mobility. In migrant cells, it has been found diffusely distributed through the cytoplasm, whereas in non-migrating cells it has been localized along the stress fibers (Arora and McCulloch, 1996). Our results demonstrate that mesangial cell gelsolin alters its intracellular location from nucleus and cytosol in response to Cd2+ treatment. The view that gelsolin can actively participate in F-actin disruption and ruffle formation is supported by the demonstration that fibroblasts from gelsolin-null mice exhibit reduced membrane ruffling and an increased amount of F-actin (Azuma et al., 1998; Kwiatkowski, 1999; Witke et al., 1995). The highly dynamic nature of the ruffling process suggests that rapid F-actin assembly and disassembly are taking place, and this may be connected with increased gelsolin accumulation in the region of ruffles following Cd2+ treatment.
In summary, Cd2+ causes gelsolin to associate with the actin cytoskeleton, which is subsequently disrupted with filament loss and membrane ruffling. In polymerization assays, gelsolin present in extracts from Cd2+-treated cells acts as a nucleating factor for actin polymerization, and mixing Cd2+ with exogenous gelsolin does not diminish this effect. However, gelsolin also causes a large increase in the depolymerization rate of actin dissolved below its critical concentration. Although Cd2+ partially reverses this, depolymerization is still enhanced in the presence of gelsolin and Cd2+. It is apparent, then, that a major effect of Cd2+ in cultured cells is to cause a redistribution of gelsolin to the cytoskeleton, where its capping and severing properties predominate in the cellular context.
ACKNOWLEDGMENTS
This work was supported by an operating grant from the Canadian Institutes of Health Research. Conflict of interest: none declared.
REFERENCES
Arora, P. D., and McCulloch, C. A. (1996). Dependence of fibroblast migration on actin severing activity of gelsolin. J. Biol. Chem. 271, 20516–20523.
Ayscough, K. R. (1998). In vivo functions of actin binding proteins. Curr. Opin. Cell Biol. 10, 102–111.
Azuma, T., Witke, W., Stossel, T. P., Hartwig, J. H., and Kwiatkowski, D. J. (1998). Gelsolin is a downstream effector of rac for fibroblast motility. EMBO J. 17, 1362–1370.
Barrouillet, M. P., Moiret, A., and Cambar, J. (1999a). Protective effect of polyphenols against cadmium-induced glomerular mesangial cell myocontracture. Arch. Toxicol. 73, 485–488.
Barrouillet, M. P., Potier, M., and Cambar, J. (1999b). Cadmium nephrotoxicity assessed in isolated rat glomeruli and cultured mesangial cells: Evidence for contraction of glomerular cells. Exp. Nephrol. 7, 251–258.
Beyersmann, D., and Hechtenberg, S. (1997). Cadmium, gene regulation, and cellular signalling in mammalian cells. Toxicol. Appl. Pharmacol. 144, 247–261.
Bhattacharyya, M. H., Wilson, A. K., Rajan, S. S., and Jonah, M. (2000). Biochemical pathways in cadmium toxicity. In Molecular Biology and Toxicology of Metals (R. K. Zalups and J. Koropatnick, Eds.), pp. 34–74. Taylor and Francis, London.
Bryan, J. (1988). Gelsolin has three actin-binding sites. J. Cell Biol. 106, 1553–1562.
Bryan, J., and Coluccio, L. M. (1985). Kinetic analysis of F-actin depolymerization in the presence of platelet gelsolin and gelsolin complexes. J. Cell Biol. 101, 1236–1244.
Burtnick, L. D., Koepf, E. K., Grimes, J., Jones, E. Y., Stuart, D. I., McLaughlin, P. J., and Robinson, R. C. (1997). The crystal structure of plasma gelsolin: Implications for actin severing, capping, and nucleation. Cell 90, 661–670.
Carpenter, C. L. (2000). Actin cytoskeleton and cell signaling. Crit. Care Med. 28(Suppl. 4), N94–99.
Carron, C. P., Hwo, S. Y., Dingus, J., Benson, D. M., Meza, I., and Bryan, J. (1986). A re-evaluation of cytoplasmic gelsolin localization. J. Cell. Biol. 102, 237–245.
Cooper, J. A., Bryan, J., Schwab, B., III, Frieden, C., Loftus, D. J., and Elson, E. L. (1987). Microinjection of gelsolin into living cells. J. Cell. Biol. 104, 491–501.
Cunningham, C. C., Stossel, T. P., and Kwiatkowski, D. J. (1991). Enhanced motility in NIH-3T3 fibroblasts that overexpress gelsolin. Science 251, 1233–1236.
Ding, W., and Templeton, D. M. (2000). Activation of parallel mitogen-activated protein kinase cascades and induction of c-fos by cadmium. Toxicol. Appl. Pharmacol. 162, 93–99.
Ditsch, A., and Wegner, A. (1994). Nucleation of actin polymerization by gelsolin. Eur. J. Biochem. 224, 223–227.
Feinberg, J., Mery, J., Heitz, F., Benyamin, Y., and Roustan, C. (1997). Conformational and functional studies of three gelsolin subdomain-1 synthetic peptides and their implication in actin polymerization. Biopolymers 41, 647–655.
Fels, L. M., Bundschuh, I., Gwinner, W., Jung, K., Pergande, M., Graubaum, H.-J., Price, R. G., Taylor, S. A., De Broe, M. E., Nuyts, G. D., et al. (1994). Early urinary markers of target nephron segments as studied in cadmium toxicity. Kidney Int. 46, S81–S88.
Goering, P. L., Waalkes, M. P., and Klaassen, C. D. (1995). Toxicology of cadmium. In Handbook of Experimental Pharmacology, Vol. 115. Toxicology of Metals – Biochemical Aspects (R. A. Goyer and M. G. Cherian, Eds.), pp. 189–214. Springer-Verlag, Berlin.
Gremm, D., and Wegner, A. (1999). Co-operative binding of Ca2+ ions to the regulatory binding sites of gelsolin. Eur. J. Biochem. 262, 330–334.
Gremm, D., and Wegner, A. (2000). Gelsolin as a calcium-regulated actin filament-capping protein. Eur. J. Biochem. 267, 4339–4345.
Hellweg, T., Hinssen, H., and Eimer, W. (1993). The Ca2+-induced conformational change of gelsolin is located in the carboxy-terminal half of the molecule. Biophys. J. 65, 799–805.
Hotz, P., Buchet, J. P., Bernard, A., Lison, D., and Lauwerys, R. (1999). Renal effects of low-level environmental cadmium exposure: 5-year follow-up of a subcohort from the Cadmibel study. Lancet 354, 1508–1513.
Jrup, L. (2002). Cadmium overload and toxicity. Nephrol. Dial. Transplant 17(Suppl. 2), 35–39.
Jrup, L., Persson, B., Edling, C., and Elinder, C. G. (1993). Renal function impairment in workers previously exposed to cadmium. Nephron 64, 75–81.
Jrup, L., Persson, B., and Elinder, C. G. (1995). Decreased glomerular filtraion rate in solderers exposed to cadmium. Occup. Environ. Med. 52, 818–822.
Kazmirski, S. L., Isaacson, R. L., An, C., Buckle, A., Johnson, C. M., Daggett, V., and Fersht, A. R. (2002). Loss of a metal-binding site in gelsolin leads to familial amyloidosis–Finnish type. Nature Structural Biol. 9, 112–116.
Khaitlina, S., and Hinssen, H. (2002). Ca-dependent binding of actin to gelsolin. FEBS Lett. 521, 14–18.
Kinosian, H. J., Newman, J., Lincoln, B., Selden, L. A., Gershman, L. C., and Estes, J. E. (1998). Ca2+ regulation of gelsolin activity: binding and severing of F-actin. Biophys. J. 75, 3101–3109.
Kothakota, S., Azuma, T., Reinhard, C., Klippel, A., Tang, J., Chu, K., McGarry, T. J., Kirschner, M. W., Koths, K., Kwiatkowski, D. J., et al. (1997). Caspase-3-generated fragments of gelsolin: Effector of morphological change in apoptosis. Science 278, 294–298.
Kwiatkowski, D. J. (1999). Functions of gelsolin: motility, signaling, apoptosis, cancer. Current Opin. Cell Biol. 11, 103–108.
L'Azou, B., Dubus, I., Ohayon-Courtes, C., Labouyrie, J. P., Perez, L., Pouvreau, C., Juvet, L., and Cambar, J. (2002a). Cadmium induces direct morphological changes in mesangial cell culture. Toxicology 179, 233–245.
L'Azou, B., Henge-Napoli, M. H., Minaro, L., Mirto, H., Barrouillet, M. P., and Cambar, J. (2002b). Effects of cadmium and uranium on some in vitro renal targets. Cell Biol. Toxicol. 18, 505–516.
Lader, A. S., Kwiatkowski, D. J., and Cantiello, H. F. (1999). Role of gelsolin in the actin filament regulation of cardiac L-type calcium channels. Am. J. Physiol. 277, C1277–C1283.
Lagarrigue, E., Maciver, S. K., Fattoum, A., Benyamin, Y., and Roustan, C. (2003). Co-operation of domain-binding and calcium-binding sites in the activation of gelsolin. Eur. J. Biochem. 270, 2236–2243.
Luna, E. J. (1998). F-actin blot overlays. Methods Enzymol. 298, 32–42.
Mazzei, G. J., Girard, P. R., and Kuo, J. F. (1984). Environmental pollutant Cd2+ biphasically and differentially regulates myosin light chain kinase and phospholipid/Ca2+-dependent protein kinase. FEBS Lett. 173, 124–128.
Nordberg, G. F., Herber, R. F. M., and Alessio, L. (Eds.) (1992). Cadmium in the Human Environment: Toxicity and Carcinogenicity. International Agency for Research on Cancer, Lyon.
Pope, B., Way, M., and Weeds, A. G. (1991). Two of the three actin-binding domains of gelsolin bind to the same subdomain of actin. Implications of capping and severing mechanisms. FEBS Lett. 280, 70–74.
Robinson, R. C., Choe, S., and Burtnick, L. D. (2001). The disintegration of a molecule: The role of gelsolin in FAF, familial amyloidosis (Finnish type). Proc. Natl. Acad. Sci. U.S.A. 98, 2117–2118.
Roels, H., Bernard, A. M., Cardenas, A., Buchet, J. P., Lauwerys, R. R., Hotter, G., Ramis, I., Mutti, A., Franchini, I., Stolte, H., et al. (1993). Markers of early renal changes induced by industrial pollutants. Application to workers exposed to cadmium. Br. J. Ind. Med. 50, 37–48.
Selden, L. A., Kinosian, H. J., Estes, J. E., and Gershman, L. C. (2000). Cross-linked dimers with nucleating activity in actin prepared from muscle acetone powder. Biochemistry 39, 64–74.
Stournaras, C. (1990). Exposure of thiol groups and bound nucleotide in G-actin: Thiols as an indicator of the native state of actin. Anticancer Res. 10, 1651–1659.
Sun, H. Q., Yamamoto, M., Mejillano, M., and Yin, H. (1999). Gelsolin, a multifunctional actin regulatory protein. J. Biol. Chem. 274, 33179–33182.
Suzuki, Y., Chao, S.-H., Zysk, J. R., and Cheung, W. Y. (1985). Stimulation of calmodulin by cadmium ion. Arch. Toxicol. 57, 205–211.
Templeton, D. M. (2000). Metal ions and the cytoskeleton. In Molecular Biology and Toxicology of Metals (R. F. Zalups and J. Loropatnick, Eds.), pp. 460–476. Taylor and Francis, London.
Valentin-Ranc, C., and Carlier, M.-F. (1991). Role of ATP-bound divalent metal ion in the conformation and function of actin. Comparison of Mg-ATP, Ca-ATP, and metal ion-free ATP-actin. J. Biol. Chem. 266, 7668–7675.
Vandekerchove, J. (1993). Actins. In Guidebook to the Cytoskeletal and Motor Proteins (T. Kreis and R. Vale, Eds.), pp. 13–15. Oxford University Press, Oxford.
Waalkes, M. P., Coogan, T. P., and Barter, R. A. (1992). Toxicological principles of metal carcinogenesis with special emphasis on cadmium. Crit. Rev. Toxicol. 22, 175–201.
Wang, A., Fan, M.-Y., and Templeton, D. M. (1994). Growth modulation and proteoglycan turnover in cultured mesangial cells. J. Cell. Physiol. 159, 295–310.
Wang, Z., Chin, T. A., and Templeton, D. M. (1996). Calcium-independent effects of cadmium on actin assembly in mesangial and vascular smooth muscle cells. Cell Motil. Cytoskeleton 32, 208–222.
Wang, Z., and Templeton, D. M. (1996). Cellular factors mediate cadmium-dependent actin depolymerization. Toxicol. Appl. Pharmacol. 139, 115–121.
Weeds, A. G., Gooch, J., Pope, B., and Harris, H. E. (1988). Preparation and characterization of pig plasma and platelet gelsolins. Eur. J. Biochem. 161, 69–76.
Welch, M. D., and Mullins, R. D. (2002). Cellular control of actin nucleation. Annu. Rev. Cell Dev. Biol. 18, 247–288.
Wen, D., Corina, K., Chow, E. P., Miller, S., Janmey, P. A., and Pepinsky, R. B. (1996). The plasma and cytoplasmic forms of human gelsolin differ in disulfide structure. Biochemistry 35, 9700–9709.
Witke, W., Sharpe, A. H., Hartwig, J. H., Azuma, T., Stossel, T. P., and Kwiatkowski, D. J. (1995). Hematostatic, inflammatory, and fibroblast responses are blunted in mice lacking gelsolin. Cell 81, 41–51.
Yin, H. N., and Stossel, T. P. (1979). Control of cytoplasmic gel-sol transition by gelsolin a calcium-dependent regulatory protein. Nature 281, 583–586.(Margarita D. Apostolova, Tania Christova)