Tài liệu Báo cáo khoa học: Interaction of the small heat shock protein with molecular mass 25 kDa (hsp25) with actin doc

Interaction of the small heat shock protein with molecular mass 25 kDa (hsp25) with actin Olesya O. Panasenko 1 , Maria V. Kim 1 , Steven B. Marston 2 and Nikolai B. Gusev 1 1 Department of Biochemistry, School of Biology, Moscow State University, Moscow, Russia; 2 Imperial College School of Medicine at National Heart and Lung Institute, Dovehose Street, London, UK The interaction of heat shock protein with molecular mass 25 kDa (HSP25) and its point mutants S77D + S81D (2D mutant) and S15D + S77D + S81D (3D mutant) with intact and thermally denatured actin was analyzed by means of fluorescence spectroscopy and ultracentrifugation. Wild type HSP25 did not affect the polymerization of intact actin. The HSP25 3D mutant decreased the initial rate without affecting the maximal extent of intact actin polymerization. G-actin heated at 40–45 °C was partially denatured, but retained its ability to polymerize. The wild type HSP25 did not affect polymerization of this partially denatured actin. The 3D mutant of HSP25 increased the initial rate of poly- merization of partially denatured actin. Heating at more than 55 °C induced complete denaturation of G-actin. Completely denatured G-actin cannot polymerize, but it aggregates at increased ionic strength. HSP25 and especially its 2D and 3D mutants effectively prevent salt-induced aggregation of completely denatured actin. It is concluded that the interaction of HSP25 with actin depends on the state of both actin and HSP25. HSP25 predominantly acts as a chaperone and preferentially interacts with thermally unfolded actin, preventing the formation of insoluble aggregates. Keywords: small heat shock protein; actin; thermal denaturation. Actin is the major component of the thin filaments of muscle cells and of the cytoskeleton system of nonmuscle cells. It is therefore a very abundant protein, and its concentration in smooth muscle is close to 800–900 l M [1]. Actin has a rather complex and labile tertiary structure [2,3]. Different types of stress can induce actin unfolding [4,5], aggregation of partially folded actin [5,6] and redistribution of actin inside the cell [7–9]. Accumulation of partially folded or aggregated proteins can induce significant damage to cells. This is especially important in the case of abundant proteins, such as actin. Therefore the cell has evolved different mechanisms to prevent the formation of insoluble aggregates, and heat shock proteins (HSPs) play an important role in this process. The data in the literature indicate that the small heat shock protein with molecular mass 25–27 kDa (HSP25) plays an important role in actin remodeling, contractility of different cell types and protection of the cytoskeleton under different unfavorable conditions [7,8,10]. Miron et al. [11,12] showed that avian HSP25 effectively inhibits actin polymerization and prevents gelation of actin induced by filamin and/or a-actinin. These observations were confirmed by Benndorf et al. [13], who showed that nonphosphoryl- ated monomers of HSP25 effectively inhibit actin polymerization, whereas phosphorylated monomers and nonphosphorylated multimers of HSP25 are ineffective in the regulation of actin polymerization. The protein seg- ments of monomeric HSP25 involved in the inhibition of actin polymerization were determined recently [14]. Although these data are of great interest, their application to cell physiology is questionable as under physiological conditions HSP25 forms high molecular mass oligomers that are in equilibrium with low molecular mass oligomers [15,16], but practically do not dissociate to monomers. The actin depolymerizing effect ascribed to HSP25 [11–14] contrasts with the stabilizing of microfilaments induced by HSP25 or its phosphorylated forms [7,17]. Moreover, recently Butt et al. [18] have shown that under in vitro conditions HSP25 either does not affect or even activates the polymerization of actin. To explain the contradictory results described in the literature we assumed that the mode of interaction is dependent both on the state of HSP25 and actin. In this paper we analyze the effect of recombinant avian HSP25 and its mutants mimicking phosphorylation on the heat- induced aggregation and polymerization of intact and partially denatured actin. Materials and methods Proteins HSP25 from chicken gizzard was purified by the procedure described previously [19]. Chicken HSP25 was cloned, Correspondence to N. B. Gusev, Department of Biochemistry, School of Biology, Moscow State University, Moscow 119992, Russia. Tel./Fax: + 7 095 939 2747, E-mail: NBGusev@mail.ru Abbreviations: ANS, 8-anilinonaphtalene-1-sulfonic acid; HSP, heat shock proteins; 1D mutant, chicken HSP25 with mutation S15D; 2D mutant, chicken HSP25 with mutation S77D + S81D; 3D mutant, chicken HSP25 with mutation S15D + S77D + S81D; MAPKAP-2, mitogen-activated protein kinase-activated protein kinase-2. (Received 15 October 2002, revised 25 December 2002, accepted 7 January 2003) Eur. J. Biochem. 270, 892–901 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03449.x expressed and purified as described by Bukach et al.[19], and Panasenko et al. [20]. Three point mutants of HSP25 with replacements S15D (1D mutant), S77D + S81D (2D mutant), and S15D + S77D + S81D (3D mutant) were obtained by the procedures published earlier [19]. Introduc- tion of extra negative charges in positions 15, 77 and 81 mimics phosphorylation of HSP25 by MAPKAP-2 kinase that may occur in vivo [16]. These mutations induce changes in the quaternary structure of avian HSP25 similar to those evoked by the corresponding mutations in mammalian HSP27 [16,20]. HSP25 may form oligomers of different sizes [10,15,16], therefore for all calculations we used a molecular mass for the monomeric proteins of 25 kDa. Rabbit skeletal actin was purified according to Pardee and Spudich [21]. G-actin in buffer G (5 m M Tris/HCl, pH 8.2, 0.2 m M ATP, 0.1 m M CaCl 2 ,0.5m M b-mercapto- ethanol, 1 m M NaN 3 ) was stored on ice and used within 10 days of purification. The purity of all proteins was checked by SDS gel electrophoresis [22]. Thermal denaturation was achieved by incubation of G-actin (usually 15 l M ) in buffer G for the required period of time at 43, 60 or 80 °C. Limited proteolysis of intact and heated actin was performed in buffer G at the weight ratio of actin/N-tosyl- L -phenylalanine chloromethyl ketone-trypsin (Sigma) equal to 100 : 1 or 50 : 1. After incubation at 25 °C for 0.25– 20 min phenylmethanesulfonyl fluoride was added to a final concentration of 0.5 m M . The samples were mixed with the sample buffer and after boiling were subjected to SDS gel electrophoresis [22]. After staining, the gels were scanned and evaluated by the ONEDSCAN program. The intensity of the band of unhydrolyzed actin was plotted against the time of incubation. Characterization of actin preparation quality Fluorescence parameters were used for estimation of nativity of actin preparations. Corrected spectra of actin fluorescence excited at 297 nm were recorded in the range 300–400 nm on the Hitachi F-3000 fluorescence spectro- photometer. Parameter A, introduced by Turoverov et al. [23] and equal to the ratio of intensities of fluorescence at 320 and 365 nm (I 320 /I 365 ) was determined for different preparations of actin. Parameter A reflects the polarity of the tryptophan environment. Any changes in actin structure influence this environment and affect parameter A. Prepa- rations of actin with A > 2.56 contain less than 2% of inactivated actin, whereas parameter A for inactivated and completely unfolded actin is equal to 1.3 and 0.4, respect- ively [5]. The interaction of intact and partially folded actin with the hydrophobic probe 8-anilinonaphtalene-1-sulfonic acid (ANS) (Sigma) was measured at 25 °C in buffer G. Samples containing 2.3 l M of intact or heated actin and 140 l M of ANS were excited at 390 nm, and spectra of fluorescence were recorded in the range 400–600 nm as before. Actin aggregation Light scattering, ultracentrifugation and size-exclusion chromatography were used to follow the process of actin aggregation. After heating in buffer G under different conditions G-actin (10–15 l M ) was cooled to 25 °C, and aggregation was initiated by the addition of KCl and MgCl 2 up to 50 m M and 2 m M , respectively. The increase of ionic strength promotes aggregation of thermally denatured actin [5]. Aggregation was followed by light scattering, measured at 560 nm, again using the Hitachi F-3000 fluorescence spectrophotometer. Ultracentrifugation was also used to follow aggregation of partially unfolded actin. In this case G-actin (15 l M )in bufferGwasheatedat60°C for 1 h. The samples were diluted with cold buffer G, cooled to 25 °C, and mixed either with buffer H (20 m M Tris/acetate pH 7.6, 10 m M NaCl, 0.1 m M EDTA, 15 m M 2-mercaptoethanol, 10% glycerol) or with different quantities of HSP25 in buffer H and incubation for 20 min at 25 °C. Buffer S (50 m M imidazole pH 7.6, 750 m M KCl, 10 m M MgCl 2 ,1m M ATP and 50 m M 2-mercaptoethanol) (1/5 of the sample volume) was added, and incubation was continued for 60 min at 25 °C. The samples obtained were then subjected to ultracentrifugation at 100 000 g for 1 h. The protein composition of the supernatant and pellet was determined by quantitative SDS gel electrophoresis [22]. Size exclusion chromatography of intact and heated actin was performed on Acta-FPLC (Amersham-Pharmacia Biotech.) using Superdex 200 10/30 column. The column was equilibrated and developed in buffer G. The samples (90 lL) of intact or heated actin (15 l M ) were loaded on the column and eluted with buffer G at the rate of 0.5 mLÆmin )1 . Actin polymerization The methods of fluorescent spectroscopy and ultracentri- fugation were used for registration of actin polymerization. In the first case F-actin was labeled by N-(1-pyrenyl)iodo- acetamide according to Kouyama and Mihashi [24]. After the removal of insoluble N-(1-pyrenyl)iodoacetamide by low-speed centrifugation (10 min, 10 000 g), the modified F-actin was collected by ultracentrifugation (1 h, 100 000 g). The pellet of modified F-actin was dissolved in buffer G, dialyzed against buffer G for 48 h and subjected to ultra- centrifugation. The supernatant contained G-actin with an extent of modification equal to 0.6–0.7 mol of N-(1-pyrenyl) iodoacetamide per mole of actin. Size exclusion chroma- tography of modified G-actin revealed that the sample does not contain high molecular mass aggregates and is free of fluorescent label, unattached to the protein. The sample of modifiedG-actinwasstoredoniceandusedwithin1week of purification. Polymerization of pyrene-labeled actin was measured according to Pollard [25] and Miron et al.[12]. Briefly, polymerization was performed in buffer G and was initiated by the addition of KCl and MgCl 2 up to 50 m M and 2 m M , respectively. Different quantities of actin nuclei (short fragments of F-actin) were added simultaneously with KCl and MgCl 2 if the initial rate of polymerization was measured. In the series of preliminary experiments we have shown that under the conditions used (1–4 l M of actin, containing 10–15% of pyrenyl-actin) there was no self- assembly of G-actin. We also observed proportionality of the initial rate of polymerization to nucleus concentration. In addition, the increase in fluorescence induced by actin polymerization was hyperbolic in time, and that above the critical concentration the rate of polymerization was linearly Ó FEBS 2003 Actin and small heat shock protein (Eur. J. Biochem. 270) 893 dependent on actin concentration. According to Pollard [25] fulfilment of all these criteria is desirable for proper measurement of the initial rate of actin polymerization. Polymerization of actin was performed at different actin concentrations both in the presence and in the absence of HSP25 or its mutant mimicking phosphorylation. In the series of separate experiments we measured the extent of actin polymerization. In this case, polymerization was initiated by the addition of only KCl and MgCl 2 (50 m M and 2 m M , respectively). In this case increase in fluorescence was sigmoidal in time and was followed for 60–90 min until it reached its maximal value. When ultracentrifugation was used for the measurement of actin polymerization, G-actin (2–8 l M in buffer G) was mixed with buffer H or with different quantities of HSP25 (or its mutants mimicking phosphorylation) in buffer H. The samples were incubated for 20 min at 25 °Cand polymerization was initiated by addition of buffer S (1/5 of the sample volume). After mixing, the samples were subjected to ultracentrifugation (1 h, 100 000 g), and the protein composition of both supernatant and pellet was determined by quantitative SDS gel electrophoresis. The quantity of actin in the pellet was plotted against the total quantity of actin in the sample. Results Heat denaturation of actin Before starting the investigation of the HSP25–actin inter- action it was desirable to characterize the properties of intact and heated actin. We were mainly interested in the irrever- sible changes in the structure of G-actin that were induced by heating. Therefore in all experiments the samples of actin in G-buffer were heated for 1 h at the appropriate temperature and after cooling different properties of actin were measured at 25 °C. In the first series of experiments we analyzed the effect of heating on intrinsic W fluorescence of actin. After recording corrected spectra of fluorescence, parameter A (equal to I 320 /I 365 ) was plotted against the temperature of incubation (Fig. 1). Parameter A was 2.65 for intact actin and decreased to 1.3 for actin heated at temperatures higher than 55 °C. Further increase of the temperature up to 80 °C had no effect on parameter A. The data presented agree with the results of Kuznetsova et al. [6] and Turoverov et al.[5] who showed that removal of Ca 2+ or addition of urea up to 4 M results in partial unfolding of actin that is accompanied by a decrease of parameter A from 2.5–2.6 to 1.2–1.3. Even prolonged heating at 80 °C does not induce complete unfolding, which is achieved only in the presence of 6–8 M urea or 4–5 M guanidine hydrochloride, and is characterized by parameter A equal to 0.4 [5,6]. As already mentioned, for many proteins partial unfold- ing is accompanied by self-aggregation. We used size-exclusion chromatography to follow heat-induced aggregation of actin. Under the conditions used, intact actin was eluted as a symmetrical peak with the maximum at 8.65 mL (Fig. 2). Heating of actin for 1 h at 43, 50 or 60 °C resulted in the appearance of an additional peak at 7.65 mL on the elution profile. Increase of the temperature of incubation was accompanied by the simultaneous increase of the peak eluted at 7.65 mL and decrease of the peak eluted at 8.65 mL. We were unable to determine the exact molecular mass of the protein species eluted in these two peaks because at low ionic strength of buffer G strongly acidic actin was partially excluded from Superdex 200. However, ovalbumin, having pI and molecular mass similar to that of actin, was eluted as a symmetrical peak with a maximum at 8.65 mL (data not shown). Therefore, we may conclude that intact, unheated actin is eluted as a monomer. Heating induces self-aggregation and the formation of actin oligomers that are eluted from the Superdex 200 column at 7.65 mL. Self-aggregation of actin can be due to the exposing of hydrophobic regions upon heating. To check this suggestion we analyzed the interaction of the hydrophobic probe ANS with intact and heated actin (Fig. 3). Under the conditions used, free ANS had a rather low intensity of fluorescence with a broad maximum at 510–530 nm (Fig. 3, curve 1). The addition of intact actin induced only a small increase in the fluorescence at 440–500 nm (Fig. 3, curve 2). Much Fig. 1. Effect of heating on the parameter A (I 320 /I 365 )ofactin.Actin (15 l M ) in buffer G was heated at different temperatures for 1 h. The intensity of fluorescence at 320 nm (I 320 ) and 365 nm (I 360 )excitedat 297 nm was used for the determination of parameter A. Fig. 2. Size-exclusion chromatography of actin on a Superdex 200 col- umn. Actin (15 l M ) was kept at 4 °C (1) or heated for 1 h at 43 (2), 50 (3) or 60 °C (4) in buffer G. After cooling, 90 lLofsamplewere loaded on the Superdex 200 column and eluted with buffer G at the rate of 0.5 mLÆmin )1 . For clearance the elution profiles are shifted from each other by 8 mAu. 894 O. O. Panasenko et al.(Eur. J. Biochem. 270) Ó FEBS 2003 higher fluorescence was observed if ANS was mixed with actin heated at 43 °C (Fig. 3, curve 3), and in this case the maximum fluorescence was shifted to 490–510 nm. The highest intensity of fluorescence, with a maximum at 465–475 nm, was observed when ANS was mixed with actin heated at 60 or 80 °C (Fig. 3, curves 4,5). The data presented indicate that heating is accompanied by the change in the hydrophobic environment of W, exposure of hydrophobic regions and self-aggregation of actin. Heating at any temperature higher than 55 °C induced similar effects on actin structure. Actin is fairly stable to trypsinolysis. The main fragment accumulated during incubation had an apparent molecular mass of 33 kDa (Fig. 4A) but even after prolonged incu- bation with trypsin more than 40% of the actin remained uncleaved (Fig. 4B, curve 1). If actin heated at 60 °Cwas subjected to trypsinolysis under the same conditions, the band of actin disappeared in the first 15–30 s (Fig. 4B, curve 2) and a number of faint bands with different molecular masses were accumulated in the incubation mixture (Fig. 4A). These results agree with other data in the literature [6] and indicate that heat-induced unfolding increased the susceptibility of actin to trypsinolysis. In contrast, if actin was heated for 1 h at 43 °C, it becomes more resistant to trypsinolysis than intact actin (Fig. 4B, curve 3). Three peptide bands with apparent molecular mass of 29, 31 and 33 kDa were accumulated during the early stages of trypsinolysis of actin heated at 43 °C (Fig. 4A). During the late stages of trypsinolysis, predominantly one major band with an apparent molecular mass of 33 kDa was detected in the incubation mixture (Fig. 4A). The data for limited trypsinolysis indicate that after heating at 43 °C actin acquires a structure different from that of the intact and thermally inactivated protein. In this state, the envi- ronment of W residues remains comparatively hydrophobic, actin weakly interacts with ANS and only a small portion of protein forms high molecular mass aggregates. Thermal unfolding can also affect actin polymerization. To investigate this possibility we heated actin containing 10% of pyrene-labeled protein for 1 h at 43 and 60 °Cand analyzed polymerization and aggregation induced by the addition of KCl and MgCl 2 (Fig. 5). As expected, salt addition induced rapid polymerization of control unheated actin (Fig. 5A, curve 1), that was accompanied by 14–16- fold increase in the fluorescence of the pyrene label attached to C373. Actin heated at 43 °C was also able to polymerize, although the rate and the extent of polymerization was slightly lower than in the case of unheated actin (Fig. 5A, curve 2). Heating at 60 °C completely prevented any increase in fluorescence induced by salt addition (Fig. 5A, curve 3). This indicates that actin heated at temperatures higher than 60 °C was not able to polymerize. Addition of salt can induce not only polymerization of intact actin, but can also promote aggregation of partially unfolded actin [5]. This process was followed by light scattering (Fig. 5B). Fig. 3. Fluorescence spectra of 140 l M ANS in the absence of added proteins (1) and in the presence of 2.25 l M of intact actin (2) or actin heated for 1 h at 43 (3), 60 (4) or 80 °C(5).All measurements were performed in G buffer at 25 °C and the fluorescence was excited at 350 nm. Fig. 4. Proteolytic fragmentation of intact actin and actin heated at 43 or 60 °C. (A) Limited proteolysis of actin by trypsin (weight ratio actin/N-tosyl- L -phenylalanine chloromethyl ketone–trypsin, 50 : 1). Actin (15 l M )inG-bufferwaskeptat4°C(gelsmarked4°C) or heated at 60 or 43 °C for 1 h (gels marked 60 °Cand43°C, respect- ively). After dilution up to 2 l M , actin was subjected to trypsinolysis that was performed under identical conditions at 25 °C. Aliquots were removed at the time indicated and the reaction was stopped by the addition of phenylmethanesulfonyl fluoride, followed by boiling in the sample buffer. The samples obtained were subjected to quantitative SDS gel electrophoresis. (B) Time course of the disappearance of the band of unhydrolyzed actin during limited trypsinolysis of intact actin (1) or actin preincubated at 60 °C (2) or 43 °C(3). Ó FEBS 2003 Actin and small heat shock protein (Eur. J. Biochem. 270) 895 Addition of the salts induced slow increase of the light scattering of actin heated at 60 °C (Fig. 5B, curve 3). As this sample of actin was unable to polymerize (Fig. 5A), the observed increase in light scattering can be due only to the slow aggregation of partially folded protein. In contrast, addition of the salts to intact actin or to actin heated at 43 °C induced rapid increase of the light scattering that coincides in time with polymerization (Fig. 5, curves 1,2). Summing up, we may conclude that heating of actin at 43 °C induced rather mild changes in the structure of actin. The samples of actin heated at 43 °Chaveamore hydrophilic environment of W residues than the intact protein, weakly bind ANS, contain rather small quantities of high molecular mass aggregates, are more resistant to trypsinolysis than intact actin and retain the ability to polymerize. Actin samples heated at temperatures higher than 55 °C are partially unfolded, strongly interact with ANS, are highly susceptible to trypsinolysis, contain high quantities of high molecular mass aggregates and are unable to polymerize. We assumed that intact and heated actin would interact differently with HSP25 so we analyzed the effect of HSP25 on the polymerization of intact actin and aggregation of heated actin. Interaction of HSP25 with intact actin In the first series of experiments we analyzed the effect of HSP25 on the apparent critical concentration and initial rate of actin polymerization (Fig. 6A). In this case, G-actin containing 10% of pyrene-labeled protein was preincubated with HSP25 or its 3D mutant, and polymerization was initiated by the simultaneous addition of actin nuclei and salts. Neither type of HSP25 had any significant effect on the critical concentration of actin (Fig. 6A). Indeed the critical concentration of actin was equal to 0.25 ± 0.07; 0.21 ± 0.05, and 0.15 ± 0.06 l M in the absence of HSP25, and in the presence of the wild type HSP25 and its 3D mutant, respectively. At the same time, both wild type and especially 3D mutant of HSP25 significantly decreased the initial rate of actin polymerization. The extent of polymeri- zation was followed by means of fluorescence spectroscopy and ultracentrifugation. In the first case, after preincubation with HSP25, polymerization of actin was initiated by salt addition. As can be seen from Fig. 6B, during the first 40 min the wild type recombinant HSP25 hardly affects actin polymerization, whereas the 3D mutant decreased the extent of polymerization by 20–25%. Similar results were obtained by ultracentrifugation. In this case, different quantities of actin were preincubated with HSP25 or its mutant, and immediately after salt addition the samples were subjected to ultracentrifugation at 100 000 g for 1 h. The quantity of actin in the pellet was plotted against the total quantity of actin in the probe (Fig. 6C). The wild type HSP25 has little effect on the extent of polymerization of intact actin, whereas both 2D and 3D mutants decreased the quantity of polymerized actin in the pellet. It is worthwhile to mention that the mutants of HSP25 affect the rate but not the maximal extent of intact actin polymerization. If the samples before ultracentrifugation were incubated for 4 h at room temperature, the quantity of polymerized actin in the pellet was almost independent of the presence of HSP25 or its mutants (data not presented). Summing up, we may conclude that the recombinant wild type HSP25 has almost no effect on the polymerization of intact actin. 2D and 3D mutants mimic the phosphorylation of HSP25 by MAPKAP-2 kinase and form oligomers with smaller molecular mass than the wild type HSP25 [20]. These mutants decrease the initial rate of actin polymeriza- tion without affecting its critical concentration (Fig. 6A). Decrease of the initial rate results in decreased extent of polymerisation, measured during the first 40–60 min after initiation of polymerization (Figs 6B,C), but does not affect the final maximal extent of polymerisation, measured 4 h after initiation of polymerization. Effect of HSP25 on polymerization of actin heated at 43 °C As mentioned earlier, mild heating at 43 °Cresultsinsome changes in actin structure, but this treatment does not Fig. 5. Effect of heating on the kinetics of polymerization (A) and salt- induced increase of the light scattering (B) of actin. Actin (15 l M ) containing 10% or pyrene-labeled actin in buffer G were incubated at 4 (1), 43 (2), or 60 °C (3) for 1 h. After cooling and diluting with buffer G, so that the concentration of actin becomes equal to 10 l M ,the reaction was started by the addition of KCl and MgCl 2 up to the final concentrations 50 and 2 m M , respectively. Polymerization was fol- lowed by an increase in fluorescence at 407 nm excited at 366 nm. Light scattering (I/I o ) was followed at 560 nm. 896 O. O. Panasenko et al.(Eur. J. Biochem. 270) Ó FEBS 2003 completely prevent actin polymerization. In addition, this heating regime resembles that occurring in vivo under certain physiological conditions. In preliminary experiments we have shown that the increase of the time of heating at 43 °C is accompanied by exponential decrease of parameter A(I 320 /I 365 ), that tends to a limit equal to 1.8 (data not shown). This value is significantly higher than the corres- ponding parameter obtained after heating at temperatures higher than 55 °C, which is equal to 1.3 (Fig. 1). Actin (containing 10% of pyrene-labeled protein) was heated at 43 °C for different periods of time, and samples of actin having different parameter A (2.2–2.7) were analyzed for their ability to polymerize. In these particular experiments we measured the initial rate of polymerization that was initiated by simultaneous addition of actin nuclei and salts. In good agreement with earlier presented results, we found that the wild type HSP25 has almost no effect on the initial rate of polymerization of intact actin (A > 2.55), whereas the 3D mutant of HSP25 slightly decreased the initial rate of polymerization (15–25%, Fig. 7). The effect of HSP25 became negligible when parameter A of actin was close to 2.4. Increase of the time of heating at 43 °C leading to decrease of parameter A up to 2.2–2.3 was accompanied by further decrease of the initial rate of polymerization. When parameter A was in the range of 2.20–2.35 the wild type HSP25 had no effect on the initial rate of polymeri- zation, whereas its 3D mutant activated the initial rate of polymerization by 25–35% (Fig. 7). This means that depending on the state of actin the 3D mutant of HSP25 can either increase (if A < 2.4) or decrease (if A > 2.4) the initial rate of polymerization. We also analyzed the effect of HSP25 and its mutants on the extent of heated actin polymerization. In this case, actin (containing 10% of pyrene-labeled protein) was heated at 43 °C until parameter A reached 2.2. Polymeri- zation was initiated by salt addition and was followed for 1 h. Under the conditions used, wild type HSP25 did not affect actin polymerization, whereas the 3D mutant of HSP25 increased the rate of polymerization without affecting the maximal extent of polymerization (data not presented). We suppose that this effect of the 3D mutant of HSP25 on the initial rate of actin polymerization can be explained by the prevention of nonspecific aggregation of partially denatured actin that can trap intact actin. The 3D mutant prevents aggregation and by this means increases the quantity of available actin monomers and therefore increa- ses the initial rate of polymerization. Fig. 6. Effect of recombinant wild type HSP25 and its mutants mimicking phosphorylation on polymerization of intact actin. (A) Effect of HSP25 and its 3D mutant on the kinetics of actin polymerization. Samples containing 1–4 l M of actin (10% of pyrene-labeled protein) were incubated for 5 min in the absence (1) or in the presence of 6 l M of HSP25 (2) or its 3D mutant (3). Polymerization was initiated by the simultaneous addition of actin nuclei, KCl and MgCl 2 up to the final concentrations 0.5 l M ,50m M and 2 m M , respectively. The results shown are representative of three experiments with three different purified actin samples, and triplicate measurements of each experi- mental point. If not shown the error bars are smaller than the size of symbol. (B) Influence of HSP25 and its 3D mutant on the extent of actin polymerization. Samples containing 4 l M of actin (10% of pyrene-labeled protein) were preincubated for 5 min at 25 °Cinthe absence (1) or in the presence of wild type HSP25 (2) or its 3D mutant (3). Polymerization was initiated by addition of KCl and MgCl 2 up to 50 and 2 m M , respectively. The results shown are representative of three independent experiments. (C) Effect of HSP25 and its mutants on actin polymerization measured by ultracentrifugation. Samples con- taining 0.12–0.48 nmol of unheated unmodified actin in 60 lLof G-buffer were incubated for 5 min at 25 °C in the absence (1) or in the presence of 10 l M of wild type HSP25 (2) or 10 l M ofits2D(3)or3D (4) mutants. One fifth of the volume of buffer S was added and after mixing the samples were immediately subjected to ultracentrifugation. The quantity of actin in the pellet is plotted against the total quantity of actin in the sample. The results are representative of five experiments with four different preparations of actin. Ó FEBS 2003 Actin and small heat shock protein (Eur. J. Biochem. 270) 897 Effect of HSP25 on the aggregation of partially unfolded thermally inactivated actin As already shown, heating of actin at a temperature higher than 55 °C completely prevents its polymerization (Fig. 5A). After this type of heating actin formed small oligomers (Fig. 2), and an increase in ionic strength induced further aggregation (Fig. 5B). We analyzed the ability of HSP25 and its mutants to prevent salt-induced aggregation of partially unfolded actin heated at 60 °C. In the first series of experiments actin was heated at 60 °C for 1 h, cooled, mixedwithdifferentspeciesofHSP25,andafterthe addition of salt subjected to ultracentrifugation. Under these conditions only aggregated actin was sedimented. Therefore, by measuring the quantity of actin in the pellet we were able to estimate the chaperone activity of HSP25 and its mutants. In the absence of HSP25 about 90% of heated actin was found in the pellet after ultracentrifugation (Fig. 8A). Addition of the wild type recombinant HSP25 reduced the quantity of sedimented actin up to 60%. Point mutants mimicking phosphorylation of HSP25 by MAPKAP-2 kinase were very effective in preventing salt- induced aggregation of partially unfolded actin (Fig. 8A). Only 10–20% of the total quantity of actin presented in the sample was precipitated in the presence of the 1D, 2D or 3D mutant of HSP25. The data presented indicate that HSP25, and especially its mutants mimicking phosphorylation, effectively prevent salt-induced aggregation of partially folded actin. To obtain more detailed information on the interaction of HSP25 with partially unfolded actin, we analyzed the effect of different quantities of HSP25 on the salt-induced aggregation of actin. In this case actin was either kept on ice or heated at 60 °C for 1 h. The samples of actin were incubated with different quantities of HSP25 or its 3D mutant. Polymerization (in the case of intact, unheated Fig. 8. Effect of HSP25 on the salt-induced aggregation of partially unfolded thermally inactivated actin. (A) The influence of different HSP25 species on the salt-induced aggregation of thermally inactivated actin measured by ultracentifugation. Actin (15 l M )inbufferGwas heated for 1 h at 60 °C. After cooling and dilution to 2 l M ,actinwas mixed with different species of HSP25 (final concentration 4 l M )and incubatedfor20minat25°C. Aggregation was initiated by the addition of 1/5 of the sample volume of buffer S. Samples were incu- bated for 1 h at 25 °C and subjected to ultracentrifugation (1 h, 100 000 g). Actin in the pellet was determined by quantitative SDS gel electrophoresis. C, control without HSP25; WT, wild type recombin- ant HSP25; 1D, 2D and 3D, HSP25 mutants with replacement of one, two or three S residues (S15, S77 and S81) by D. The results are representative of four independent experiments with three different preparations of actin and triplicate measurements of each experimental point. (B) Concentration-dependent effect of the wild type HSP25 and its 3D mutant on polymerization of intact actin (1,2) and salt-induced aggregation of actin heated at 60 °C (3,4). 15 l M actininbufferGwas kept at 4 °C (1,2) or at 60 °C (3,4) for 1 h. After cooling and dilution to 2 l M , actin was mixed with different quantities of the wild type recombinant HSP25 (1,3) or its 3D mutant (2,4) and incubated for 20 min. One fifth of the volume of buffer S was added and after incubation for 2 h at 25 °C the samples were subjected to ultracentri- fugation (1 h, 100 000 g). Actin in the pellet (percentage of the total actin in the sample) was determined by quantitative SDS gel electro- phoresis. The results are representative of six independent experiments with three different actin samples. Fig. 7. Dependence of the rate of polymerization upon parameter A of actin. Actin (15 l M , containing 10% pyrene-labeled protein) in buffer G was heated for 0–90 min at 43 °CandparameterA(I 320 /I 365 )was recorded. The samples were cooled, diluted to a final actin concen- tration of 4 l M and incubated in the absence (1) or in the presence of 6 l M of the wild type HSP25 (2) or its 3D mutant (3) for 5 min at 25 °C. Polymerization was initiated by the simultaneous addition of actin nuclei, KCl and MgCl 2 at final concentrations of 0.2 l M ,50m M and 2 m M , respectively. The initial rate of polymerization was deter- mined during the first 2 min of the reaction by increase in fluorescence at 407 nm excited at 366 nm. The results are representative of two independent experiments with two different actin samples, with trip- licate measurements of each experimental point. 898 O. O. Panasenko et al.(Eur. J. Biochem. 270) Ó FEBS 2003 actin) or aggregation (in the case of heated actin) was initiated by salt addition and was allowed to proceed for 2 h at 25 °C. After ultracentrifugation the quantity of actin in the pellet was determined by quantitative SDS gel electro- phoresis. Actin in the pellet (as a percentage of actin in the total sample) was plotted against the HSP25 concentration (Fig. 8B). As described earlier, neither the wild type HSP25 nor its 3D mutant affected the final extent of polymerization of intact unheated actin (Fig. 8B, curves 1,2). Addition of the wild type HSP25 reduced the quantity of aggregated partially folded actin heated at 60 °C (Fig. 8B, curve 3). However, even at a very high concentrations the wild type HSP25 was not able to completely prevent aggregation of partially unfolded actin. In contrast, even much lower quantities of the 3D mutant of HSP25 completely prevented aggregation of actin heated at 60 °C (Fig. 8B, curve 4). We may conclude that HSP25 effectively prevents aggregation of partially folded actin. Phosphorylation (or mutations mimicking phosphorylation) increased the chap- erone effect of HSP25 so that it becomes able to completely prevent salt-induced aggregation of heated actin. Discussion Heating induces significant changes in the structure of G-actin. Bertazzon et al. [4] suggested that upon heating native (N) actin is irreversibly converted to denatured (D) (or partially folded) actin. This first step of unfolding is enthalpic and involves the denaturation of two independent domains of approximately 11 and 31 kDa. Addition of a high concentration of guanidine hydrochloride or urea can reversibly convert D (or partially folded) actin to the completely unfolded (U) state. This second step of unfolding is reversible and purely enthropic. Thus, the mechanism of unfolding of G-actin was described by a simple scheme: N-actin ! D-actin Ð U-actin However, the first transition from N-actin to D-actin was not a one-step process. It has been shown [4] that the calorimetric melting curve of actin was asymmetric, and that the excess heat capacity curve of G-actin can be fitted into two independent intermediate steps with T m of 52 and 57 °C, respectively. Therefore, the scheme of actin unfolding is more complex and can be represented in the form: N-actin Ð D 1 -actin ! D 2 -actin Ð U-actin where D 1 -andD 2 -actins represent two states of denatured actin and the reversibility or irreversibility of transitions between N, D 1 and D 2 are unknown. Although D 1 and D 2 states of actin were postulated, their properties and even their existence was not confirmed experimentally. We propose that heating of actin at 40–45 °Cleadstoa slow transition from N-actin to D 1 -actin. The D 1 state of actin is different from both native actin and from the well- characterized denatured (or D 2 ) state of actin. Heating under these mild physiologically relevant conditions results in the accumulation of the protein with only a moderate change in the W environment (Fig. 1), diminished ability to interact with ANS (Fig. 3) and a small quantity of high molecular mass aggregates (Fig. 2). In addition, actin in this state was more resistant to proteolysis than intact protein and in this respect was completely different from denatured actin (Fig. 4). Moreover, even after prolonged heating at 43 °C actin retained its ability to polymerize (Figs 5 and 7). This property was completely lost by denatured actin (Fig. 5). We may suppose that heating at 43 °C induces unfolding of a small domain proposed by Bertazzon et al. [4]. At present it is difficult to locate this domain exactly in the crystallographic structure of actin. However, it is known that trypsin predominantly cleaves actin at residues 62 and 68, forming fragments with apparent molecular masses of 33 and 9 kDa [26]. Heating at 43 °C partially protects actin from trypsinolysis (Fig. 4). This fact may indicate that the above-mentioned small domain with molecular mass 9–11 kDa may include subdomains 1 and 2 of actin. After heating at 43 °C, actin turns into a state that is different from both the intact and denatured conformations. This intermediate state may be of importance because under physiological conditions the body temperature of warm- blooded animals can rise up to 40–42 °C. The denatured (D or D 2 ) state of actin was analyzed in detail [4–6]. This state is characterized by a very hydrophilic environment of W residues, exposed hydrophobic sites interacting with ANS and increased susceptibility to proteolysis. Exposure of hydrophobic sites increases the probability of self-aggregation and therefore partially folded actin tends to aggregate. Self-aggregation may be the reason for the irreversibility of transition from the native to the partially folded state [5]. Transition of N-actin to the D (or D 2 ) state was observed after heating at temperatures higher than 55 °C, after the removal of calcium or after the addition of low concentrations of urea or guanidine hydrochloride [5,6]. Accumulation of denatured actin can be dangerous for the cell as it tends to form high molecular mass aggregates. Let us analyze the interaction of HSP25 with the N-, D 1 - and D 2 -forms of actin. Miron et al. [11,12] claimed that HSP25 effectively inhibits polymerization of N-actin by increasing its critical concentration. Similar results were obtained by Benndorf et al. [13], but only with the monomeric unphosphorylated form of HSP25. These results were obtained with HSP25 purified from avian or human tissues. In both cases the starting steps of purifica- tion of HSP25 were performed according to Feramisco and Burridge [27] and an initial crude mixture contained a number of different proteins with molecular mass in the range of 20–80 kDa that were able to inhibit polymerization of actin [28]. The ability of HSP25 to inhibit actin polymerization was diminished or completely deteriorated if the protein was purified on a hydroxyapatite column under special conditions [12]; recombinant HSP25 was also ineffective in the inhibition of actin polymerization [13,18]. All of these facts can be explained by the suggestion that HSP25 purified from animal tissues contained trace amounts of a highly effective inhibitor of actin polymeriza- tion having a molecular mass of monomers close to that of HSP25. Cofilin, which has an apparent molecular mass of 20–22 kDa, could be one of the candidates for this role. At substoichiometric concentrations cofilin inhibits actin poly- merization, induces depolymerization of actin and is able to form oligomers with molecular mass in the range 22–100 kDa [29]. As heat shock is accompanied by the simultaneous translocation of both cofilin and HSP25 to the Ó FEBS 2003 Actin and small heat shock protein (Eur. J. Biochem. 270) 899 nucleus [30], we may suppose that these two proteins can interact with each other. Analyzing the interaction of native actin with HSP25 purified from avian tissues and with recombinant protein, we found that the wild type HSP25 has little effect on the rate or extent of actin polymerization. At the same time, the 3D mutant of HSP25 slightly decreased the initial rate of actin polymerization (Fig. 6A) without affecting the maxi- mal extent of polymerization. Mutations mimicking phos- phorylation induce partial dissociation of high molecular mass oligomers of HSP25 and accumulation of dimers and tetramers [20]. Low molecular mass oligomers of HSP25 may interact with G-actin and in this way decrease the initial rate of polymerization. However, this interaction seems to be weak and therefore HSP25 (or its 3D mutant) does not affect the final extent of polymerization. Heating at 40–45 °C leads to transition of native actin to the D 1 form and is accompanied by a decrease in the rate and extent of actin polymerization (Figs 5 and 7). This could be due to the fact that at this heating regime some actin becomes aggregated (Fig. 2) and therefore is excluded from polymerization. The wild type HSP25 has little effect on the polymerization of D 1 -actin (Fig. 7), whereas the 3D mutant of HSP25 increases the rate of polymerization without affecting its maximal extent (Fig. 7). This effect of the 3D mutant may be explained by preventing aggregation of actin leading to an increase in the concentration of G-actin available for polymerization. Conversion of native actin to the D 2 -form completely prevents polymerization (Fig. 5A). Denatured actin con- tains exposed hydrophobic sites (Fig. 3) and tends to aggregate upon addition of salt (Fig. 5B). HSP25 prevents salt-induced aggregation of denatured actin (Fig. 8), and HSP25 mutants mimicking phosphorylation possessed higher chaperone activity than the wild type HSP25. The chaperone activity of HSP25 strongly depends on the nature of the target protein and on the state of HSP25 phosphory- lation. For example, phosphorylation (or mutations mimicking phosphorylation) decreases the chaperone activity of human or murine HSP27 with citrate synthase, insulin [16] and avian HSP25 with a-lactalbumin [20]. At the same time, phoshorylation (or mutations mimicking phos- phorylation) increases the chaperone activity of HSP25 with alcohol dehydrogenase [20]. The same effect was observed in the case of denatured actin. Different types of stress induce phosphorylation of HSP25 [10], thus converting it to the form that effectively prevents aggregation of actin, and in this way protect the cell from accumulation of large quantities of insoluble material. Summing up we may conclude that depending on the conditions HSP25 has multiple effects on polymerization and aggregation of G-actin. Monomers or low molecular mass oligomers of HSP25 weakly interact with G-actin and thereby slightly inhibit the initial rate of polymerization of intact actin. The HSP25 mutants, mimicking phosphoryla- tion, stabilize partially denatured molecules of G-actin, prevent formation of high molecular mass aggregates and in this way increase the initial rate of polymerization of partially denatured actin. Finally HSP25, and especially its mutants, effectively prevent salt-induced aggregation of denatured actin, thereby protecting the cell from the accumulation of insoluble proteins. Acknowledgements The authors are grateful to Dr Alim S. Seit-Nebi (V.A. Engelhardt Institute of Molecular Biology, Russian Academy of Sciences) for the cloning and expression of recombinant forms of HSP25 and its mutants. This investigation was supported by Russian Foundation for Basic Research and by the Wellcome Trust. References 1. Walsh, M.P. (1994) Regulation of vascular smooth muscle tone. Can. J. Physiol. Pharmacol. 72, 919–936. 2. Kabsch, W., Mannherz, H.G., Suck, D., Pai, E.F. & Holmes, K.C. (1990) Atomic structure of the actin: DNase I complex. Nature 347, 37–44. 3. Otterbein, L.R., Graceffa, P. & Dominguez, R. (2001) The crystal structure of uncomplexed actin in the ADP state. Science 293, 708–711. 4. Bertazzon, A., Tian, G.H., Lamblin, A. & Tsong, T.Y. (1990) Enthalpic and entropic contributions to actin stability: calorimetry, circular dichroism, and fluorescence study and effects of calcium. Biochemistry 29, 291–298. 5. Turoverov, K.K., Biktashev, A.G., Khaitlina, S.Y. & Kuznetsova, I.M. (1999) The structure and dynamics of partially folded actin. Biochemistry 38, 6261–6269. 6. Kuznetsova, I.M., Biktashev, A.G., Khaitlina, S.Y., Vassilenko, K.S., Turoverov, K.K. & Uversky, V.N. (1999) Effect of self association on the structural organization of partially folded proteins: inactivated actin. Biophys. J. 77, 2788–2800. 7. Lavoie, J.N., Gingras-Breton, G., Tanguay, R.M. & Landry, J. (1993) Induction of Chinese hamster HSP27 gene expression in mouse cells confers resistance to heat shock. HSP27 stabilization of the microfilament organization. J. Biol. Chem. 268, 3420–3429. 8. Lavoie,J.N.,Lambert,H.,Hickey,E.,Weber,L.A.&Landry,J. (1995) Modulation of cellular thermoresistance and actin filament stability accompanies phosphorylation-induced changes in the oligomeric structure of heat shock protein 27. Mol. Cell Biol. 15, 505–516. 9. Loktionova, S.A., Ilyinskaya, O.P. & Kabakov, A.E. (1998) Early and delayed tolerance to simulated ischemia in heat- preconditioned endothelial cells: a role for HSP27. Am. J. Physiol. 275, H2147–H2158. 10. Gerthoffer, W.T. & Gunst, S.J. (2001) Invited review: focal adhesion and small heat shock proteins in the regulation of actin remodeling and contractility in smooth muscle. J. Appl. Physiol. 91, 963–972. 11. Miron, T., Vancompernolle, K., Vandekerckhove, J., Wilchek, M. & Geiger, B. (1991) A 25-kD inhibitor of actin polymerization is a low molecular mass heat shock protein. J. Cell Biol. 114, 255–261. 12. Miron, T., Wilchek, M. & Geiger, B. (1988) Characterization of an inhibitor of actin polymerization in vinculin-rich fraction of turkey gizzard smooth muscle. Eur. J. Biochem. 178, 543–553. 13. Benndorf,R.,Hayess,K.,Ryazantsev,S.,Wieske,M.,Behlke,J. & Lutsch, G. (1994) Phosphorylation and supramolecular organization of murine small heat shock protein HSP25 abolish its actin polymerization-inhibiting activity. J. Biol. Chem. 269, 20780–20784. 14. Wieske, M., Benndorf, R., Behlke, J., Dolling, R., Grelle, G., Bielka, H. & Lutsch, G. (2001) Defined sequence segments of the small heat shock proteins HSP25 and aB-crystallin inhibit actin polymerization. Eur. J. Biochem. 268, 2083–2090. 15. Ehrnsperger, M., Lilie, H., Gaestel, M. & Buchner, J. (1999) The dynamics of Hsp25 quaternary structure. Structure and function of different oligomeric species. J. Biol. Chem. 274, 14867–14874. 16. Rogalla, T., Ehrnsperger, M., Preville, X., Kotlyarov, A., Lutsch, G., Ducasse, C., Paul, C., Wieske, M., Arrigo, A.P., Buchner, J. & 900 O. O. Panasenko et al.(Eur. J. Biochem. 270) Ó FEBS 2003 Gaestel, M. (1999) Regulation of Hsp27 oligomerization, chaperone function, and protective activity against oxidative stress/tumor necrosis factor a by phosphorylation. J. Biol. Chem. 274, 18947–18956. 17. Schafer, C., Clapp, P., Welsh, M.J., Benndorf, R. & Williams, J.A. (1999) HSP27 expression regulates CCK-induced changes of the actin cytoskeleton in CHO-CCK-A cells. Am.J.Physiol.277, C1032–C1043. 18. Butt, E., Immler, D., Meyer, H.E., Kotlyarov, A., Laass, K. & Gaestel, M. (2001) Heat shock protein 27 is a substrate of cGMP- dependent protein kinase in intact human platelets: phos- phorylation-induced actin polymerization caused by HSP27 mutants. J. Biol. Chem. 276, 7108–7113. 19. Bukach, O.V., SeitNebi, A.S., Panasenko, O.O., Kim, M.V. & Gusev, N.B. (2002) Isolation of tissue and recombinant small heat shock protein with a molecular weight 25 kDa (HSP25) from aviansmoothmuscles.Problems Biol. Med. Pharmaceut. Chem. 1, 50–57. 20. Panasenko, O.O., Seit Nebi, A.S., Bukach, O.V., Marston, S.B. & Gusev, N.B. (2002) Structure and properties of avian small heat shock protein with molecular weight 25 kDa. Biochim. Biophys. Acta 1601, 64–74. 21. Pardee, J.D. & Spudich, J.A. (1982) Purification of muscle actin. Methods Cell Biol. 24, 271–289. 22. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. 23. Turoverov, K.K., Haitlina, S.Y. & Pinaev, G.P. (1976) Ultra- violet fluorescence of actin. Determination of native actin content in actin preparations. FEBS Lett. 62, 4–6. 24. Kouyama, T. & Mihashi, K. (1981) Fluorimetry study of N-(1- pyrenyl) iodoacetamide-labelled F-actin. Local structural change of actin protomer both on polymerization and on binding of heavy meromyosin. Eur. J. Biochem. 114, 33–38. 25. Pollard, T.D. (1983) Measurement of rate constants for actin filament elongation in solution. Anal. Biochem. 134, 406–412. 26. Mornet, D. & Ue, K. (1983) Proteolysis and structure of skeletal muscle actin. Proc. Natl Acad. Sci. USA 81, 3680–3684. 27. Feramisco, J.R. & Burridge, K. (1980) A rapid purification of a-actinin, filamin, and a 130,000-dalton protein from smooth muscle. J. Biol. Chem. 255, 1194–1199. 28. Ruhnau, K., Schroer, E. & Wegner, A. (1988) Characterization of the actin polymerization-inhibiting protein from chicken gizzard smooth muscle. Eur. J. Biochem. 170, 583–587. 29. Pfannstiel, J., Cyrklaff, M., Habermann, A., Stoeva, S., Griffiths, G., Shoeman, R. & Faulstich, H. (2001) Human cofilin forms oligomers exhibiting actin bundling activity. J. Biol. Chem. 276, 49476–49484. 30. Ohta, Y., Nishida, E., Sakai, H. & Miyamoto, E. (1989) Dephosphorylation of cofilin accompanies heat shock-induced nuclear accumulation of cofilin. J. Biol. Chem. 264, 16143–16148. Ó FEBS 2003 Actin and small heat shock protein (Eur. J. Biochem. 270) 901 . Interaction of the small heat shock protein with molecular mass 25 kDa (hsp25) with actin Olesya O. Panasenko 1 , Maria. analyzed the effect of HSP25 on the polymerization of intact actin and aggregation of heated actin. Interaction of HSP25 with intact actin In the first series of