- Email: [email protected]
Characterization of the palytoxin effect on Ca2+-ATPase from sarcoplasmic reticulum (SERCA)
Archives of Biochemistry and Biophysics 478 (2008) 36–42
Contents lists available at ScienceDirect
Archives of Biochemistry and Biophysics journal homepage: www.elsevier.com/locate/yabbi
Characterization of the palytoxin effect on Ca2+-ATPase from sarcoplasmic reticulum (SERCA) Ramón Coca, Fernando Soler, Francisco Fernández-Belda * Departamento de Bioquímica y Biología Molecular A, Universidad de Murcia, Campus de Espinardo, 30071 Espinardo, Murcia, Spain
a r t i c l e
i n f o
Article history: Received 9 May 2008 and in revised form 16 July 2008 Available online 24 July 2008 Keywords: Ca2+-ATPase Palytoxin Inhibition mechanism Hydrolytic and transport cycle Sarcoplasmic reticulum
a b s t r a c t The effect of palytoxin was studied in a microsomal fraction enriched in longitudinal tubules of the sarcoplasmic reticulum membrane. Half-maximal effect of palytoxin on Ca2+-ATPase activity yielded an apparent inhibition constant of approx. 0.4 lM. The inhibition process exhibited the following characteristics: (i) the degree of inhibition was dependent on membrane protein concentration; (ii) no protection was observed when the ATP concentration was raised; (iii) dependence on Ca2+ concentration with a decreased maximum catalytic rate; (iv) it occurred in the absence of Ca2+ ionophoric activity. Likewise, the inhibition mechanism was linked to: (i) rapid enzyme phosphorylation from ATP in the presence of Ca2+ but lower steady-state levels of phosphoenzyme; (ii) more drastic effect on phosphoenzyme levels when the toxin was added to the enzyme in the absence of Ca2+; (iii) decreased phosphoenzyme levels at saturating Ca2+ concentrations; (iv) no effect on kinetics of phosphoenzyme decomposition. The palytoxin effect is related with lock of the enzyme in the Ca2+-free conformation so that progression of the catalytic cycle is impeded. Ó 2008 Elsevier Inc. All rights reserved.
Palytoxin (PTX)1 is one of the most potent marine toxins showing extreme toxicity in mammals [1]. It was first isolated from the zoanthid Palythoa toxica [2] and later from other species of the genera Palythoa and Zoanthus. It can be found in other organisms living near or feeding on toxic zoanthid colonies [3] and the presence of PTX in several fish species has also been reported [4,5]. The relatively high resistance of marine animals toward secondary metabolites toxins may favors the entry of PTX into food chains and therefore it can be potentially harmful for humans. PTX is a non-peptide molecule that has a partially unsaturated continuous chain of 115-carbon atoms. It is also noticeable the presence of cyclic ethers, 42 hydroxyl groups and 64 chiral centers [6]. There are minor variations of the toxin structure related with the specific source and the location of the source. Early studies associated the effect of PTX to membrane depolarization of excitable cells due to an increased permeability for Na+ and K+ [7,8]. Other results suggested the appearance of a low conductance (10 picosiemens) and a non-selective cation channel in cardiac cells [9]. It was also recognized that Na+, K+-ATPase was the * Corresponding author. Fax: +34 968 364 147. E-mail address: [email protected] (F. Fernández-Belda). 1 Abbreviations used: SR, sarcoplasmic reticulum; EGTA, ethylene glycol-bis(baminoethyl ether)-N,N,N0 ,N0 -tetraacetic acid; PTX, palytoxin; Mops, 4-morpholinepropanesulfonic acid; pCa, negative logarithm of free Ca2+ expressed as molar concentration; E1Ca2, enzyme conformation when sarcoplasmic reticulum vesicles are in a free Ca2+-containing medium; E2, enzyme conformation when sarcoplasmic reticulum vesicles are in a Ca2+-free medium; EP, phosphoenzyme. 0003-9861/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2008.07.017
molecular target for PTX [10] and the effect on membrane permeability could be antagonized by cardiac glycosides [11]. The heterologous expression of the Na+-pump in yeast confirmed that the transformed cells were highly sensitive to PTX and the toxin-induced K+ efflux could be blocked by ouabain [12]. Furthermore, PTX was shown to inhibit the Na+, K+-ATPase activity from heart and cerebral tissue with half-maximal effect at 3.1 and 0.9 lM, respectively [13]. The effect of PTX was described as concentration-dependent, thus a concentration 1 pM increased the ouabain-sensitive K+ efflux from erythrocytes whereas higher concentrations such as 100 nM and above inhibited the Na+, K+ATPase [14]. The remarkable and highly effective toxic principle attributed to PTX was the conversion of the Na+-pump into a non-selective cation channel [12,15]. It was described that PTX interferes with the alternating access mechanism of the Na+-pump thereby promoting a permanently channel-like open state [16]. Cysteine-scanning mutagenesis of the Na+-pump a1 subunit and the subsequent mutant accessibility study with sulfhydryl reagents indicated that transmembrane segments were implicated in the PTX-induced channel [17–19]. It was also observed that the extracellular loop L7/8 of the Na+-pump a subunit interacts with the b subunit and plays an important role in ion transport [20,21]. Moreover, it was proved that the presence of the b subunit is essential for the PTX-induced channel formation [12]. In this regard, the pore-forming effect of PTX was not restricted to the Na+, K+-ATPase and the effect was also observed on the H+,K+-ATPase [22].
R. Coca et al. / Archives of Biochemistry and Biophysics 478 (2008) 36–42
The most sensitive target at the physiological level is the cardiovascular system. Indeed, the PTX effect on the vascular smooth muscle is well documented [23] whereas a direct effect of PTX on the cardiac tissue is less established. Studies on atrial myocytes described a PTX-induced depolarization as expected for the effect on Na+-pump but inhibition of the SR Ca2+-pump was also reported [24]. The aim of the present study was the ‘‘in vitro” characterization of the PTX effect on SR Ca2+-ATPase to uncover potential similarities with the closely related sarcolemmal Na+, K+-ATPase.
37
Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.1 mM EGTA, 0.154 mM 45 CaCl2 at 100,000 cpm/nmol (pCa 4.3), 3 lg SR/ml, 2 mM potassium oxalate and 1 mM ATP. The addition of 5 mM EGTA took place when the time elapsed was 2 min and 30 s and the addition of either 1 lM A23187 or 1 lM PTX occurred at time 6 min and 30 s. Aliquots taken at different time intervals were rapidly filtered through HAWP Millipore filters (0.45 lm pore size) placed in a FH 225V filtration module from Hoefer/Amersham Pharmacia Biotech. Thereafter, filters under vacuum were rinsed with 10 ml of ice-cold medium containing 20 mM Mops, pH 7.0 and 1 mM LaCl3 and then subjected to liquid scintillation counting.
Materials and methods Radioactive EP Reagents The radioactive compound 45CaCl2 was obtained from Amersham Radiochemicals (GE Healthcare) and [c-32P]ATP was from Perkin Elmer Life Sciences. PTX from P. caribaeorum was purchased from either Sigma or Alexis Biochemicals. A23187 from Streptomyces chartreusensis was provided by Calbiochem. The Ca2+ standard solution TitrisolÒ was from Merck. All other reagents of analytical grade were supplied by Sigma. Samples of SR membrane White muscle was extracted from the hind legs of female New Zealand rabbit (body weight 2–2.5 kg). A microsomal fraction enriched in longitudinal tubules of the SR membrane was obtained according to Eletr and Inesi [25]. This is a suitable experimental model due to the purity of the preparation and the high content in Ca2+ATPase protein. Native SR vesicles at 15–20 mg/ml were aliquoted, quick-frozen in liquid nitrogen and stored at 80 °C for further use. The concentration of the SR membrane was evaluated by the Lowry et al. method [26] and expressed as mg of total protein/ml. Free Ca2+ Concentration of ionic species and complexes at equilibrium, including free Ca2+, were calculated as previously described [27]. The computer program took into account the absolute stability constant for the complex Ca2+-EGTA [28], the EGTA protonation equilibria [29], the presence of Ca2+ ligands and the pH in the medium. A nominally Ca2+-free medium was prepared by including 2 mM EGTA and no Ca2+ added. Under these conditions, the free Ca2+ concentration was below the activation threshold of the enzyme. When indicated, free Ca2+ was given as pCa. ATPase activity The rate of ATP hydrolysis was evaluated at 25 °C by measuring the release of inorganic phosphate with a malachite green reagent [30]. The standard reaction mixture contained 20 mM Mops, pH 7.0, 80 KC1, 5 mM MgCl2, 0.5 mM EGTA, 0.55 mM CaC12, equivalent to pCa 4.3, 1 lM A23187, 1 lg SR/ml, 1 mM ATP and a certain concentration of PTX. When indicated, the reaction medium was supplemented with 2 mM phosphoenolpyruvate and 6 U/ml pyruvate kinase as an ATP-regenerating system. Data in the figures correspond to Ca2+-dependent activities and were calculated by subtracting the hydrolytic activity measured in a Ca2+-free medium. Full details of specific assay conditions are given in the corresponding figure caption. Permeability of Ca2+-loaded vesicles Native SR vesicles were initially loaded with Ca2+ using 45Ca2+ as a radioactive tracer [31]. The assay medium consisted of 20 mM
The evaluation of EP accumulated after addition of [c-32P]ATP was based in the method described by Sarkadi et al. [32] for measurements of the plasma membrane Ca2+-ATPase. The phosphorylating substrate was 5 lM [c-32P]ATP and the quenching solution was 7% trichloroacetic acid plus 10 mM sodium phosphate. All solutions were pre-cooled and the experiments were conducted at the ice-water temperature according to the following protocols. Addition of [c-32P]ATP to E1Ca2 samples The initial reaction medium was 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.5 mM EGTA, 0.55 mM CaC12, 1 lM A23187 and 1 lg SR/ml. Aliquots of 60 ll reaction medium in the absence or presence of 1 lM PTX were mixed with 6 ll of 55 lM [c-32P]ATP at 2 107 cpm/nmol. The reaction was stopped at different time periods by adding 0.5 ml quenching solution. The blank assay was performed by adding the quenching solution before radioactive ATP. Quenched samples were supplemented with 80 ll of 1 mg/ ml bovine serum albumin and maintained for 5 min in an ice bath. Afterwards, samples were centrifuged at 14,000g for 15 min and 4 °C and the resulting pellets were reserved. EP levels at different pCa The experimental design was similar to that described in the preceding paragraph, i.e., addition of [c-32P]ATP to SR vesicles equilibrated in the presence of Ca2+. In this case, the CaCl2 concentration was varied to yield different free Ca2+ values. The reaction medium included 1 lM PTX when indicated. Samples were processed as described above. Addition of [c-32P]ATP plus Ca2+ to E2 samples The initial reaction medium containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.5 mM EGTA, 1 lM A23187 and 1 lg SR/ml was distributed in 60 ll aliquots. When indicated, 1 lM PTX was also included. The phosphorylation reaction was initiated by adding 6 ll of medium containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 6.06 mM CaC12 and 55 lM [c-32P]ATP at 2 107 cpm/nmol. The reaction was stopped at timed intervals by adding 0.5 ml quenching solution. The blank assay was performed by changing the order of the ATP and quenching solution additions. The denatured protein was processed as described for the E1Ca2 samples. EP decomposition Samples were distributed in 60 ll aliquots containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.5 mM EGTA, 0.55 mM CaC12, 1 lM A23187 and 1 lg SR/ml in the absence or presence of 1 lM PTX. The phosphorylation reaction was started by adding 6 ll of 55 lM [c-32P]ATP at 2 107 cpm/nmol and arrested 4 s later by the addition of 5 mM EGTA. The time-dependent EP decomposition was studied by adding 0.5 ml aliquots of quenching solution at different time intervals. Samples were processed as described above.
38
R. Coca et al. / Archives of Biochemistry and Biophysics 478 (2008) 36–42
Proteins resolution and EP detection Radioactive pellets containing denatured protein were dissolved in 15 ll of electrophoresis sample buffer consisting of 7 mM monosodium phosphate, 3 mM disodium phosphate, 2.5% lithium dodecyl sulfate, 0.5% b-mercaptoethanol, 0.005% bromophenol blue and 20% glycerol. Remaining traces of acid were neutralized by adding 1 ll of 0.35 M Tris. Sample aliquots containing 0.06 lg of SR protein were loaded and electrophoresed on 7.5% polyacrylamide gel slabs at pH 6.3 [33]. Gels were stained and dried after electrophoresis. Imaging and quantitation of radioactive bands was performed using a Kodak Storage Phosphor Screen model SD230 and a Personal Molecular Imager FX with Quantity One v 4.4.0 software from Bio Rad. Data presentation Each value of enzyme activity or Ca2+ loading is an average of at least three independent measurements and standard deviations of mean values (plus or minus) are given. Autoradiographic data on EP are representative for the corresponding experiment. Curve fitting and calculation of first-order rate constants was performed with version 8.0 of SigmaPlot Graph System. EP assays for PTXtreated and the corresponding control samples were carried out simultaneously and under the same conditions to allow comparison. Quantitation of radioactive bands was expressed as percentage.
Results PTX inhibits Ca2+-ATPase without membrane permeabilization The Ca2+-ATPase activity associated to ‘‘in vitro” preparations of the SR membrane was sensitive to the presence of PTX. Initial experiments were performed at 25 °C, in a buffered medium at neutral pH containing 50 lM free Ca2+. Isolated SR vesicles at 1 lg/ml were made leaky to Ca2+ by the presence of A23187 and the reaction was started by adding 1 mM ATP. The ATP hydrolysis rate measured under these conditions was 1.22 lmol Pi/min/mg protein. When the experiment was repeated but a certain PTX concentration was included the specific activity was inhibited by PTX in a concentration-dependent manner (Fig. 1A). The inhibition data points could be fitted to a monophasic curve and half-maximal effect was observed at approx. 0.4 lM PTX. The dependence of Ca2+-ATPase activity on PTX concentration was also measured at different membrane protein concentrations using the assay conditions described above. The degree of inhibition was higher when the SR concentration was 1 lg/ml and a progressive decrease in inhibition was observed when the SR concentration was raised to 3, 10 or 50 lg/ml (Fig. 1B). These data revealed that the inhibition induced by PTX was dependent on membrane protein concentration. Measurements of enzyme activity in the presence of different PTX concentrations were compared when the ATP concentration was 1 mM ATP or 50 lM ATP supplemented with an ATP-regenerating system. No protection was observed by 20-fold increase of the ATP concentration (data not shown). The Ca2+-ATPase activity of leaky SR vesicles at 3 lg/ml was measured at different free Ca2+ concentrations and the reaction was started by adding 1 mM ATP. As expected, the ATP hydrolysis rate displayed a sigmoidal dependence with respect to Ca2+ concentration (Fig. 2). The Ca2+-dependent activation curve was affected when the experiments were repeated in the presence of 1 lM PTX. It is significant that the maximum catalytic rate was reduced by 45%.
Fig. 1. PTX inhibited the Ca2+-ATPase activity from SR and the inhibition was dependent on enzyme concentration. (A) The reaction medium was 20 mM Mops, pH 7.0, 80 mM KC1, 5 mM MgCl2, 0.5 mM EGTA, 0.55 mM CaC12 (pCa 4.3), 1 lM A23187, 1 lg SR/ml and a defined PTX concentration. The enzyme activity was studied at 25 °C by adding 1 mM ATP (d). The release of inorganic phosphate was evaluated by a colorimetric procedure [30]. (B) The dependence on enzyme concentration was analyzed by studying the inhibitory effect of PTX at the following SR concentrations: 1 (d), 3 (s), 10 (j) or 50 lg/ml (h). Experimental conditions were similar to those described for panel A. The A23187 concentration was raised to 5 lM when 50 lg SR/ml was used.
Fig. 2. PTX affected the dependence of Ca2+-ATPase activity on pCa. The hydrolytic activity was measured at 25 °C in a medium containing 20 mM Mops, pH 7.0, 80 mM KC1, 5 mM MgCl2, 0.5 mM EGTA, a CaCl2 concentration to give a defined pCa, 1 lM A23187, 3 lg SR/ml and 1 mM ATP. Experiments were carried out in the absence (s) or presence of 1 lM PTX (d).
The described effect of PTX as inducer of a pump to channel transformation in the Na+, K+-ATPase [12,15] prompted us to check the ability of the toxin to collapse a pre-formed Ca2+ gradient using the SR membrane preparation. The positive control assay was carried out by adding 1 mM ATP to sealed SR vesicles (3 lg/ml) equil-
R. Coca et al. / Archives of Biochemistry and Biophysics 478 (2008) 36–42
39
ibrated in a medium containing 50 lM 45Ca2+ and 2 mM oxalate. The time course of active Ca2+ transport inside the vesicles showed a lineal dependence during the early time points (Fig. 3A). Moreover, the addition of 5 mM EGTA at time 2 min and 30 s prevented any further Ca2+ entry and gave rise to a gradient equivalent to 1.2 lmol Ca2+/mg protein. Under these conditions, complete discharge of the vesicles was observed when 1 lM A23187 was added at time 6 min and 30 s. In a parallel experiment, sealed SR vesicles were actively loaded with 45Ca2+ and a Ca2+ gradient was established by adding excess EGTA at 2 min and 30 s, as before. In this case, the subsequent addition of 1 lM PTX, instead of A23187, was unable to induce release of Ca2+ (Fig. 3B). This indicates that PTX does not promote any Ca2+ ionophoric activity. The inhibitory effect of PTX can be pinpointed The hydrolytic and transport cycle of the SR Ca2+-ATPase is triggered by enzyme phosphorylation from ATP once the high affinity Ca2+ transport sites are occupied giving rise to the rapid formation and accumulation of EP. This partial reaction was studied at the ice-water temperature by mixing the membrane suspension containing 1 lg/ml of leaky vesicles equilibrated in the presence of 50 lM free Ca2+ with the phosphorylation medium to give a final concentration of 5 lM [c-32P]ATP. The reaction was studied during the first 10 s after acid quenching, electrophoretic protein resolution and quantitation of radioactive 32P. As can be seen, the EP level was maximal at 1 s when samples were phosphorylated in the absence of PTX (Fig. 4). The steady-state level of EP was maintained during the studied time frame. When the phosphorylation process took place in the presence of 1 lM PTX, a rapid EP accumulation was also observed, although the steady-state level was lower than that obtained in the absence of toxin. The Ca2+-dependent activation of the enzyme, previously shown through ATP hydrolysis experiments, was also analyzed measuring the accumulation of EP. The reaction medium composition was as indicated in the above paragraph but free Ca2+ concentration was varied. The accumulated EP was initially measured in the absence of PTX. When the pCa selected was 7.0, a low level of EP was accumulated, however higher EP values were observed as the Ca2+ concentration was raised. The EP accumulation reached a saturation level (Fig. 5). When the phosphorylation experiment was performed in the presence of 1 lM PTX a Ca2+ dependence was observed even though, the maximal EP level was considerable lower than that accumulated when the experiment was performed in the absence of PTX. Incubation of SR vesicles in a medium with excess EGTA allowed exposure of the E2 enzyme conformation to the toxin. The
Fig. 4. PTX decreased the EP level when [c-32P]ATP was added to SR vesicles in the presence of Ca2+. The initial reaction medium was 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.5 mM EGTA, 0.55 mM CaC12, 1 lM A23187 and 1 lg SR/ml in the absence (s) or presence of 1 lM PTX (d). Aliquots of 60 ll placed in an ice bath were phosphorylated by adding 6 ll of 55 lM [c-32P]ATP at 2 107 cpm/nmol. The reaction was stopped at different time intervals by adding 0.5 ml of ice-cold quenching solution. The assay at time zero was performed by adding the quenching solution before radioactive ATP. The evaluation of radioactive EP was carried by autoradiographic procedure once the SR proteins were resolved by electrophoresis at pH 6.3.
subsequent phosphorylation process required the simultaneous addition of [c-32P]ATP and Ca2+ to give final concentrations of 5 lM ATP and 50 lM free Ca2+. Data obtained in the absence of PTX showed rapid accumulation of EP that was patent at 1 s phosphorylation (Fig. 6). The inclusion of 1 lM PTX in the reaction medium also revealed the rapid phosphorylation process although a strong reduction of the EP level was observed when compared with the control experiment. EP decomposition is usually a rate limiting step allowing accumulation of the phosphorylated enzyme under steady-state conditions. The study of this partial reaction was accomplished by a pulse and chase experiment. Thus, leaky SR vesicles at 1 lg/ml and in the presence of 50 lM free Ca2+ were first phosphorylated with 5 lM [c-32P]ATP and excess EGTA was added after 4 s to
Fig. 3. PTX did not permeabilize the SR membrane to Ca2+. Sealed SR vesicles were actively loaded with Ca2+ at 25 °C in a medium containing 20 mM Mops, pH 7.0, 80 mM KC1, 5 mM MgCl2, 0.1 mM EGTA, 0.154 mM 45CaCl2 at 100,000 cpm/nmol (pCa 4.3), 3 lg SR/ml, 2 mM potassium oxalate and 1 mM ATP. Ca2+ loading was arrested at 2 min and 30 s by adding 5 mM EGTA. The subsequent addition of 1 lM A23187 (A) or 1 lM PTX (B) was performed at time 6 min and 30 s. Aliquots of the reaction medium were filtered at different time intervals and processed for evaluation of the 45Ca2+ retained by the vesicles.
40
R. Coca et al. / Archives of Biochemistry and Biophysics 478 (2008) 36–42
Fig. 5. PTX affected the EP dependence on pCa. The initial reaction medium containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.5 mM EGTA, different CaC12 concentrations to yield defined pCa, 1 lM A23187 and 1 lg SR/ml was distributed in 60 ll aliquots. The phosphorylation of samples placed in an ice bath was started by adding 6 ll of 55 lM [c-32P]ATP at 2 107 cpm/nmol. The reaction was stopped after 4 s by adding 0.5 ml of ice-cold quenching solution (s). The experiment was repeated by including 1 lM PTX in the initial reaction medium (d). Quenched samples were processed for quantitation of radioactive EP.
Fig. 7. PTX did not modify the EP decomposition kinetics. Aliquots of 60 ll containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.5 mM EGTA, 0.554 mM CaC12, 1 lM A23187 and 1 lg SR/ml in the absence (s) or presence of 1 lM PTX (d) were placed in an ice bath. Samples were phosphorylated by adding 6 ll of 55 lM [c-32P]ATP at 2 107 cpm/nmol. After 4 s, 5 mM EGTA was added and EP decomposition was followed at different time intervals by adding 0.5 ml ice-cold quenching solution. The EGTA addition corresponded to time zero for dephosphorylation.
interrupt the progression of the reaction cycle. The induced decomposition of radioactive EP was monitored at different time intervals after acid quenching of the samples. The experiment was repeated by including 1 lM PTX in the initial reaction medium. The steady-state level of radioactive EP was higher in the absence than in the presence of PTX (Fig. 7). However, the kinetics of EP decomposition followed monoexponential decay and was very similar whether PTX was present or absent. Curve fitting provided an apparent rate constant of 0.35 s1 for EP decomposition in the presence of PTX and 0.33 s1 when the experiment was performed in the absence of toxin. Discussion
Fig. 6. PTX decreased the EP level when [c-32P]ATP plus Ca2+ were added to SR vesicles in the absence of Ca2+. SR vesicles at 1 lg/ml were initially equilibrated in a medium containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.5 mM EGTA and 1 lM A23187 in the absence (s) or presence of 1 lM PTX (d). Aliquots of 60 ll placed in an ice bath were phosphorylated by adding 6 ll of medium containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 6.06 mM CaC12 and 55 lM [c-32P]ATP at 2 107 cpm/nmol. The reaction was stopped at different time intervals by adding 0.5 ml of ice-cold quenching solution. The quenching solution was added before the phosphorylation medium in the assay at time zero. Quenched samples were subjected to electrophoresis and autoradiography for EP determination.
The dependence of the Ca2+-ATPase activity on PTX concentration was consistent with a single enzyme population displaying an apparent inhibition constant of approx. 0.4 lM (Fig. 1A). In terms of molar ratio, a PTX concentration of 2 lM produced practically full effect when the SR concentration was 1 lg/ml. This was equivalent to a PTX/enzyme molar ratio of 500, assuming from the maximal EP level that 1 mg of SR protein contains 4 nmol of Ca2+-ATPase active sites. Therefore, PTX can be considered a strong inhibitor of the enzyme. The effect of Ca2+ on enzyme activity is related with the equilibrium: E2 + 2Ca2+ ? E1Ca2. In fact, the abundance of Ca2+-bound or Ca2+-free enzyme conformations is dependent on pCa, and the accumulation of active conformation E1Ca2 is favored when the Ca2+ concentration is raised. The PTX effect on Ca2+ concentration dependence (Fig. 2) indicated that the Ca2+-dependent activation process was incomplete. This can be explained by an alteration of the Ca2+ binding equilibrium causing stabilization of the enzyme in the inactive E2 conformation. This finding is consistent with the existence of a fraction of fully active enzyme molecules and another fraction of totally inhibited molecules.
R. Coca et al. / Archives of Biochemistry and Biophysics 478 (2008) 36–42
Our data indicated that addition of PTX to Ca2+ loaded vesicles after arrest of the Ca2+ transport activity by EGTA did not alter the permeability of the SR membrane to Ca2+ (Fig. 3). Therefore, the SR Ca2+-pump cannot be transformed into a permanently open Ca2+ channel by the presence of PTX. The described effect of PTX on the Na+-pump did not require a catalytically active enzyme as demonstrated when the Asp369 ? Ala mutation was introduced in the a subunit [34]. When SR vesicles are equilibrated in a saturating Ca2+ medium all the enzyme molecules will be in the E1Ca2 conformation. This allows the study of the phosphorylation reaction, i.e., E1Ca2 + ATP ? E1PCa2 + ADP. The results obtained at the ice-water temperature confirmed that phosphorylation in the absence or presence of PTX was a very fast process (Fig. 4). This was an indication that PTX did not affect enzyme phosphorylation by ATP and was consistent with no dependence on ATP concentration when the enzyme activity was measured as a function of PTX (data not shown). Nevertheless, the accumulated EP level was somewhat lower in the presence than in the absence of PTX. This can be attributed to the effect of PTX on the Ca2+ binding process and is consistent with the lower maximal rate observed when ATP hydrolysis was measured as a function of pCa (Fig. 2). The effect of PTX on EP was also studied by equilibrating SR vesicles in the presence of different free Ca2+ concentrations before phosphorylation. This assay confirmed that the accumulation of EP was related with the free Ca2+ concentration used whether PTX was absent or present in the reaction medium (Fig. 5). It was also shown that phosphorylation by ATP was diminished when the experiment was performed in the presence of 1 lM PTX. When SR vesicles were placed in a Ca2+-free medium to induce the E2 conformation, the phosphorylation experiments required the addition of Ca2+ and ATP. This allowed the study of the following process: E2 + 2Ca2+ + ATP ? E1Ca2 ATP ? E1PCa2 + ADP. Ca2+ binding and phosphorylation by ATP are very rapid partial reactions leading to the steady-state accumulation of EP in less than 1 s when control experiments were carried out (Fig. 6). The addition of 1 lM PTX to the Ca2+-free samples before phosphorylation by ATP in the presence of Ca2+ provoked a clear decrease of the EP levels. Comparison of these results with those reported in Fig. 4 confirms preferential interaction of PTX with the E2 conformation of the enzyme. The effect of PTX on the enzyme turnover was marked by a decreased steady-state EP level without alteration of the EP decomposition kinetics (Fig. 7). These data indicated that formation and decomposition of EP were not affected by PTX and the inhibition can be attributed to a limitation of the enzyme population that can be phosphorylated in the presence of toxin. Stabilization of the E2 conformation and interference with the Ca2+ binding process seems to be the functional effect of PTX on this Ca2+-dependent ATPase. Many hydrophobic compounds with different structural characteristics including thapsigargin [35] and 2,5-di(tert-butyl)-1,4dihydroxibenzene [36] showed inhibitory action of the SR Ca2+ATPase by interfering with the Ca2+ binding process. Crystallographic data have revealed that the common functional effect of thapsigargin [37] and 2,5-di(tert-butyl)-1,4-dihydroxibenzene [38] was to restraint the movement of the enzyme transmembrane helices although the interaction of these two inhibitors occurred at different binding sites. This suggests that binding of different hydrophobic molecules to the transmembrane region that locks the enzyme into the E2 conformation is a prevalent inhibitory mechanism. Despite the close structural analogy among members of the Ptype ATPases, PTX did not transform the SR Ca2+-ATPase into a channel as occurs with the heterodimeric Na+, K+-ATPase. It is significant that the Na+-pump was only sensitive to PTX, and also to
41
ouabain, when both a and b subunits were co-expressed and not when either subunits was expressed alone [12]. Moreover, H+,K+ATPase from colon epithelial cells that was affected by PTX in the same way [22] has also a two-subunit composition. Notably, the Na+-pump a subunit can form functional pump complexes with the b subunit of the H+-pump [39] and a hybrid between Na+,K+ATPase and H+,K+-ATPase is sensitive to PTX [40]. It seems that the PTX effect as a channel inducer is related with the a/b-oligomeric composition of the pump rather than with the primary structure of the protein. It can be concluded that the PTX action on sarcolemmal Na+,K+-ATPase and SR Ca2+-ATPase [24] occurs through different molecular mechanisms. Pilot assays have revealed that PTX from P. tuberculosa does not reproduce the inhibitory profile displayed by PTX from P. caribaeorum on SR Ca2+-ATPase (data not shown). This is consistent with reported differences in the chemical structure of PTX depending on sources and locations and with observed differences in the functional effect on Na+,K+-ATPase [13]. This fact may contribute to the variety of effects that have been attributed to the toxin. Acknowledgment This study was supported by Grant BMC2002-02474 from Ministerio de Educación y Cultura, Spain. References [1] [2] [3] [4] [5] [6]
[7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32]
J.S. Wiles, J.A. Vick, M.K. Christensen, Toxicon 12 (1974) 427–433. R.E. Moore, P.J. Scheuer, Science 172 (1971) 495–498. S. Gleibs, D. Mebs, Toxicon 37 (1999) 1521–1527. M. Fukui, M. Murata, A. Inoue, M. Gawel, T. Yasumoto, Toxicon 25 (1987) 1121–1124. A.M. Kodama, Y. Hokama, T. Yasumoto, M. Fukui, S.J. Manea, N. Sutherland, Toxicon 27 (1989) 1051–1053. J.K. Cha, W.J. Christ, J.M. Finan, H. Fujioka, Y. Kishi, L.L. Klein, S.S. Ko, J. Leder, W.W. McWhorter Jr., K.-P. Pfaff, M. Yonaga, J. Am. Chem. Soc. 104 (1982) 7369– 7371. J.M. Dubois, J.B. Cohen, J. Pharmacol. Exp. Ther. 201 (1977) 148–155. H. Ozaki, J. Tomono, H. Nagase, N. Urakawa, Jpn. J. Pharmacol. 33 (1983) 1155– 1162. M. Ikeda, K. Mitani, K. Ito, Naunyn Schmiedebergs Arch. Pharmacol. 337 (1988) 591–593. H. Bottinger, L. Beress, E. Habermann, Biochim. Biophys. Acta 861 (1986) 165– 176. N.A. Castle, G.R. Strichartz, Toxicon 26 (1988) 941–951. G. Scheiner-Bobis, D. Meyer zu Heringdorf, M. Christ, E. Habermann, Mol. Pharmacol. 45 (1994) 1132–1136. Y. Ishida, K. Takagi, M. Takahashi, N. Satake, S. Shibata, J. Biol. Chem. 258 (1983) 7900–7902. E. Habermann, Toxicon 27 (1989) 1171–1187. P. Artigas, D.C. Gadsby, J. Gen. Physiol. 123 (2004) 357–376. P. Artigas, D.C. Gadsby, Proc. Natl. Acad. Sci. USA 100 (2003) 501–505. S. Guennoun, J.-D. Horisberger, FEBS Lett. 482 (2000) 144–148. S. Guennoun, J.D. Horisberger, FEBS Lett. 513 (2002) 277–281. J.-D. Horisberger, S. Kharoubi-Hess, S. Guennoun, O. Michielin, J. Biol. Chem. 279 (2004) 29542–29550. H. Schneider, G. Scheiner-Bobis, J. Biol. Chem. 272 (1997) 16158–16165. S. Becker, H. Schneider, G. Schneider-Bobis, Eur. J. Biochem. 271 (2004) 3821– 3831. G. Scheiner-Bobis, T. Hübschle, M. Diener, Eur. J. Biochem. 269 (2002) 3905– 3911. K. Ito, N. Urakawa, H. Koike, Arch. Int. Pharmacodyn. Ther. 258 (1982) 146– 154. J. Kockskämper, G.U. Ahmmed, A.V. Zima, K.A. Sheehan, H.G. Glitsch, L.A. Blatter, Am. J. Physiol. Cell Physiol. 287 (2004) C527–C538. S. Eletr, G. Inesi, Biochim. Biophys. Acta 282 (1972) 174–179. O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, J. Biol. Chem. 193 (1951) 265–275. A. Fabiato, Methods Enzymol. 157 (1988) 378–417. G. Schwartzenbach, H. Senn, G. Anderegg, Helv. Chim. Acta 40 (1957) 1886– 1900. J.R. Blinks, W.G. Wier, P. Hess, F. Prendergast, Prog. Biophys. Mol. Biol. 40 (1982) 1–114. P.A. Lanzetta, L.J. Alvarez, P.S. Reinach, O.A. Candia, Anal. Biochem. 100 (1979) 95–97. A. Martonosi, R. Feretos, J. Biol. Chem. 239 (1964) 648–658. B. Sarkadi, A. Enyedi, Z. Földes-Papp, G. Gárdos, J. Biol. Chem. 261 (1986) 9552– 9557.
42 [33] [34] [35] [36]
R. Coca et al. / Archives of Biochemistry and Biophysics 478 (2008) 36–42
K. Weber, M. Osborn, J. Biol. Chem. 244 (1969) 4406–4412. G. Scheiner-Bobis, H. Schneider, Eur. J. Biochem. 248 (1997) 717–723. Y. Sagara, G. Inesi, J. Biol. Chem. 266 (1991) 13503–13506. Y.M. Khan, M. Wictome, J.M. East, A.G. Lee, Biochemistry 34 (1995) 14385– 14393. [37] C. Toyoshima, H. Nomura, Nature 418 (2002) 605–611.
[38] K. Obara, N. Miyashita, C. Xu, I. Toyoshima, Y. Sugita, G. Inesi, C. Toyoshima, Proc. Natl. Acad. Sci. USA 102 (2005) 14489–14496. [39] J.-D. Horisberger, P. Jaunin, M.A. Reuben, L.S. Lasater, D.C. Chow, J.G. Forte, G. Sachs, B.C. Rossier, K. Geering, J. Biol. Chem. 266 (1991) 19131–19134. [40] R.A. Farley, S. Schreiber, S.-G. Wang, G. Scheiner-Bobis, J. Biol. Chem. 276 (2001) 2608–2615.
Contents lists available at ScienceDirect
Archives of Biochemistry and Biophysics journal homepage: www.elsevier.com/locate/yabbi
Characterization of the palytoxin effect on Ca2+-ATPase from sarcoplasmic reticulum (SERCA) Ramón Coca, Fernando Soler, Francisco Fernández-Belda * Departamento de Bioquímica y Biología Molecular A, Universidad de Murcia, Campus de Espinardo, 30071 Espinardo, Murcia, Spain
a r t i c l e
i n f o
Article history: Received 9 May 2008 and in revised form 16 July 2008 Available online 24 July 2008 Keywords: Ca2+-ATPase Palytoxin Inhibition mechanism Hydrolytic and transport cycle Sarcoplasmic reticulum
a b s t r a c t The effect of palytoxin was studied in a microsomal fraction enriched in longitudinal tubules of the sarcoplasmic reticulum membrane. Half-maximal effect of palytoxin on Ca2+-ATPase activity yielded an apparent inhibition constant of approx. 0.4 lM. The inhibition process exhibited the following characteristics: (i) the degree of inhibition was dependent on membrane protein concentration; (ii) no protection was observed when the ATP concentration was raised; (iii) dependence on Ca2+ concentration with a decreased maximum catalytic rate; (iv) it occurred in the absence of Ca2+ ionophoric activity. Likewise, the inhibition mechanism was linked to: (i) rapid enzyme phosphorylation from ATP in the presence of Ca2+ but lower steady-state levels of phosphoenzyme; (ii) more drastic effect on phosphoenzyme levels when the toxin was added to the enzyme in the absence of Ca2+; (iii) decreased phosphoenzyme levels at saturating Ca2+ concentrations; (iv) no effect on kinetics of phosphoenzyme decomposition. The palytoxin effect is related with lock of the enzyme in the Ca2+-free conformation so that progression of the catalytic cycle is impeded. Ó 2008 Elsevier Inc. All rights reserved.
Palytoxin (PTX)1 is one of the most potent marine toxins showing extreme toxicity in mammals [1]. It was first isolated from the zoanthid Palythoa toxica [2] and later from other species of the genera Palythoa and Zoanthus. It can be found in other organisms living near or feeding on toxic zoanthid colonies [3] and the presence of PTX in several fish species has also been reported [4,5]. The relatively high resistance of marine animals toward secondary metabolites toxins may favors the entry of PTX into food chains and therefore it can be potentially harmful for humans. PTX is a non-peptide molecule that has a partially unsaturated continuous chain of 115-carbon atoms. It is also noticeable the presence of cyclic ethers, 42 hydroxyl groups and 64 chiral centers [6]. There are minor variations of the toxin structure related with the specific source and the location of the source. Early studies associated the effect of PTX to membrane depolarization of excitable cells due to an increased permeability for Na+ and K+ [7,8]. Other results suggested the appearance of a low conductance (10 picosiemens) and a non-selective cation channel in cardiac cells [9]. It was also recognized that Na+, K+-ATPase was the * Corresponding author. Fax: +34 968 364 147. E-mail address: [email protected] (F. Fernández-Belda). 1 Abbreviations used: SR, sarcoplasmic reticulum; EGTA, ethylene glycol-bis(baminoethyl ether)-N,N,N0 ,N0 -tetraacetic acid; PTX, palytoxin; Mops, 4-morpholinepropanesulfonic acid; pCa, negative logarithm of free Ca2+ expressed as molar concentration; E1Ca2, enzyme conformation when sarcoplasmic reticulum vesicles are in a free Ca2+-containing medium; E2, enzyme conformation when sarcoplasmic reticulum vesicles are in a Ca2+-free medium; EP, phosphoenzyme. 0003-9861/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2008.07.017
molecular target for PTX [10] and the effect on membrane permeability could be antagonized by cardiac glycosides [11]. The heterologous expression of the Na+-pump in yeast confirmed that the transformed cells were highly sensitive to PTX and the toxin-induced K+ efflux could be blocked by ouabain [12]. Furthermore, PTX was shown to inhibit the Na+, K+-ATPase activity from heart and cerebral tissue with half-maximal effect at 3.1 and 0.9 lM, respectively [13]. The effect of PTX was described as concentration-dependent, thus a concentration 1 pM increased the ouabain-sensitive K+ efflux from erythrocytes whereas higher concentrations such as 100 nM and above inhibited the Na+, K+ATPase [14]. The remarkable and highly effective toxic principle attributed to PTX was the conversion of the Na+-pump into a non-selective cation channel [12,15]. It was described that PTX interferes with the alternating access mechanism of the Na+-pump thereby promoting a permanently channel-like open state [16]. Cysteine-scanning mutagenesis of the Na+-pump a1 subunit and the subsequent mutant accessibility study with sulfhydryl reagents indicated that transmembrane segments were implicated in the PTX-induced channel [17–19]. It was also observed that the extracellular loop L7/8 of the Na+-pump a subunit interacts with the b subunit and plays an important role in ion transport [20,21]. Moreover, it was proved that the presence of the b subunit is essential for the PTX-induced channel formation [12]. In this regard, the pore-forming effect of PTX was not restricted to the Na+, K+-ATPase and the effect was also observed on the H+,K+-ATPase [22].
R. Coca et al. / Archives of Biochemistry and Biophysics 478 (2008) 36–42
The most sensitive target at the physiological level is the cardiovascular system. Indeed, the PTX effect on the vascular smooth muscle is well documented [23] whereas a direct effect of PTX on the cardiac tissue is less established. Studies on atrial myocytes described a PTX-induced depolarization as expected for the effect on Na+-pump but inhibition of the SR Ca2+-pump was also reported [24]. The aim of the present study was the ‘‘in vitro” characterization of the PTX effect on SR Ca2+-ATPase to uncover potential similarities with the closely related sarcolemmal Na+, K+-ATPase.
37
Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.1 mM EGTA, 0.154 mM 45 CaCl2 at 100,000 cpm/nmol (pCa 4.3), 3 lg SR/ml, 2 mM potassium oxalate and 1 mM ATP. The addition of 5 mM EGTA took place when the time elapsed was 2 min and 30 s and the addition of either 1 lM A23187 or 1 lM PTX occurred at time 6 min and 30 s. Aliquots taken at different time intervals were rapidly filtered through HAWP Millipore filters (0.45 lm pore size) placed in a FH 225V filtration module from Hoefer/Amersham Pharmacia Biotech. Thereafter, filters under vacuum were rinsed with 10 ml of ice-cold medium containing 20 mM Mops, pH 7.0 and 1 mM LaCl3 and then subjected to liquid scintillation counting.
Materials and methods Radioactive EP Reagents The radioactive compound 45CaCl2 was obtained from Amersham Radiochemicals (GE Healthcare) and [c-32P]ATP was from Perkin Elmer Life Sciences. PTX from P. caribaeorum was purchased from either Sigma or Alexis Biochemicals. A23187 from Streptomyces chartreusensis was provided by Calbiochem. The Ca2+ standard solution TitrisolÒ was from Merck. All other reagents of analytical grade were supplied by Sigma. Samples of SR membrane White muscle was extracted from the hind legs of female New Zealand rabbit (body weight 2–2.5 kg). A microsomal fraction enriched in longitudinal tubules of the SR membrane was obtained according to Eletr and Inesi [25]. This is a suitable experimental model due to the purity of the preparation and the high content in Ca2+ATPase protein. Native SR vesicles at 15–20 mg/ml were aliquoted, quick-frozen in liquid nitrogen and stored at 80 °C for further use. The concentration of the SR membrane was evaluated by the Lowry et al. method [26] and expressed as mg of total protein/ml. Free Ca2+ Concentration of ionic species and complexes at equilibrium, including free Ca2+, were calculated as previously described [27]. The computer program took into account the absolute stability constant for the complex Ca2+-EGTA [28], the EGTA protonation equilibria [29], the presence of Ca2+ ligands and the pH in the medium. A nominally Ca2+-free medium was prepared by including 2 mM EGTA and no Ca2+ added. Under these conditions, the free Ca2+ concentration was below the activation threshold of the enzyme. When indicated, free Ca2+ was given as pCa. ATPase activity The rate of ATP hydrolysis was evaluated at 25 °C by measuring the release of inorganic phosphate with a malachite green reagent [30]. The standard reaction mixture contained 20 mM Mops, pH 7.0, 80 KC1, 5 mM MgCl2, 0.5 mM EGTA, 0.55 mM CaC12, equivalent to pCa 4.3, 1 lM A23187, 1 lg SR/ml, 1 mM ATP and a certain concentration of PTX. When indicated, the reaction medium was supplemented with 2 mM phosphoenolpyruvate and 6 U/ml pyruvate kinase as an ATP-regenerating system. Data in the figures correspond to Ca2+-dependent activities and were calculated by subtracting the hydrolytic activity measured in a Ca2+-free medium. Full details of specific assay conditions are given in the corresponding figure caption. Permeability of Ca2+-loaded vesicles Native SR vesicles were initially loaded with Ca2+ using 45Ca2+ as a radioactive tracer [31]. The assay medium consisted of 20 mM
The evaluation of EP accumulated after addition of [c-32P]ATP was based in the method described by Sarkadi et al. [32] for measurements of the plasma membrane Ca2+-ATPase. The phosphorylating substrate was 5 lM [c-32P]ATP and the quenching solution was 7% trichloroacetic acid plus 10 mM sodium phosphate. All solutions were pre-cooled and the experiments were conducted at the ice-water temperature according to the following protocols. Addition of [c-32P]ATP to E1Ca2 samples The initial reaction medium was 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.5 mM EGTA, 0.55 mM CaC12, 1 lM A23187 and 1 lg SR/ml. Aliquots of 60 ll reaction medium in the absence or presence of 1 lM PTX were mixed with 6 ll of 55 lM [c-32P]ATP at 2 107 cpm/nmol. The reaction was stopped at different time periods by adding 0.5 ml quenching solution. The blank assay was performed by adding the quenching solution before radioactive ATP. Quenched samples were supplemented with 80 ll of 1 mg/ ml bovine serum albumin and maintained for 5 min in an ice bath. Afterwards, samples were centrifuged at 14,000g for 15 min and 4 °C and the resulting pellets were reserved. EP levels at different pCa The experimental design was similar to that described in the preceding paragraph, i.e., addition of [c-32P]ATP to SR vesicles equilibrated in the presence of Ca2+. In this case, the CaCl2 concentration was varied to yield different free Ca2+ values. The reaction medium included 1 lM PTX when indicated. Samples were processed as described above. Addition of [c-32P]ATP plus Ca2+ to E2 samples The initial reaction medium containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.5 mM EGTA, 1 lM A23187 and 1 lg SR/ml was distributed in 60 ll aliquots. When indicated, 1 lM PTX was also included. The phosphorylation reaction was initiated by adding 6 ll of medium containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 6.06 mM CaC12 and 55 lM [c-32P]ATP at 2 107 cpm/nmol. The reaction was stopped at timed intervals by adding 0.5 ml quenching solution. The blank assay was performed by changing the order of the ATP and quenching solution additions. The denatured protein was processed as described for the E1Ca2 samples. EP decomposition Samples were distributed in 60 ll aliquots containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.5 mM EGTA, 0.55 mM CaC12, 1 lM A23187 and 1 lg SR/ml in the absence or presence of 1 lM PTX. The phosphorylation reaction was started by adding 6 ll of 55 lM [c-32P]ATP at 2 107 cpm/nmol and arrested 4 s later by the addition of 5 mM EGTA. The time-dependent EP decomposition was studied by adding 0.5 ml aliquots of quenching solution at different time intervals. Samples were processed as described above.
38
R. Coca et al. / Archives of Biochemistry and Biophysics 478 (2008) 36–42
Proteins resolution and EP detection Radioactive pellets containing denatured protein were dissolved in 15 ll of electrophoresis sample buffer consisting of 7 mM monosodium phosphate, 3 mM disodium phosphate, 2.5% lithium dodecyl sulfate, 0.5% b-mercaptoethanol, 0.005% bromophenol blue and 20% glycerol. Remaining traces of acid were neutralized by adding 1 ll of 0.35 M Tris. Sample aliquots containing 0.06 lg of SR protein were loaded and electrophoresed on 7.5% polyacrylamide gel slabs at pH 6.3 [33]. Gels were stained and dried after electrophoresis. Imaging and quantitation of radioactive bands was performed using a Kodak Storage Phosphor Screen model SD230 and a Personal Molecular Imager FX with Quantity One v 4.4.0 software from Bio Rad. Data presentation Each value of enzyme activity or Ca2+ loading is an average of at least three independent measurements and standard deviations of mean values (plus or minus) are given. Autoradiographic data on EP are representative for the corresponding experiment. Curve fitting and calculation of first-order rate constants was performed with version 8.0 of SigmaPlot Graph System. EP assays for PTXtreated and the corresponding control samples were carried out simultaneously and under the same conditions to allow comparison. Quantitation of radioactive bands was expressed as percentage.
Results PTX inhibits Ca2+-ATPase without membrane permeabilization The Ca2+-ATPase activity associated to ‘‘in vitro” preparations of the SR membrane was sensitive to the presence of PTX. Initial experiments were performed at 25 °C, in a buffered medium at neutral pH containing 50 lM free Ca2+. Isolated SR vesicles at 1 lg/ml were made leaky to Ca2+ by the presence of A23187 and the reaction was started by adding 1 mM ATP. The ATP hydrolysis rate measured under these conditions was 1.22 lmol Pi/min/mg protein. When the experiment was repeated but a certain PTX concentration was included the specific activity was inhibited by PTX in a concentration-dependent manner (Fig. 1A). The inhibition data points could be fitted to a monophasic curve and half-maximal effect was observed at approx. 0.4 lM PTX. The dependence of Ca2+-ATPase activity on PTX concentration was also measured at different membrane protein concentrations using the assay conditions described above. The degree of inhibition was higher when the SR concentration was 1 lg/ml and a progressive decrease in inhibition was observed when the SR concentration was raised to 3, 10 or 50 lg/ml (Fig. 1B). These data revealed that the inhibition induced by PTX was dependent on membrane protein concentration. Measurements of enzyme activity in the presence of different PTX concentrations were compared when the ATP concentration was 1 mM ATP or 50 lM ATP supplemented with an ATP-regenerating system. No protection was observed by 20-fold increase of the ATP concentration (data not shown). The Ca2+-ATPase activity of leaky SR vesicles at 3 lg/ml was measured at different free Ca2+ concentrations and the reaction was started by adding 1 mM ATP. As expected, the ATP hydrolysis rate displayed a sigmoidal dependence with respect to Ca2+ concentration (Fig. 2). The Ca2+-dependent activation curve was affected when the experiments were repeated in the presence of 1 lM PTX. It is significant that the maximum catalytic rate was reduced by 45%.
Fig. 1. PTX inhibited the Ca2+-ATPase activity from SR and the inhibition was dependent on enzyme concentration. (A) The reaction medium was 20 mM Mops, pH 7.0, 80 mM KC1, 5 mM MgCl2, 0.5 mM EGTA, 0.55 mM CaC12 (pCa 4.3), 1 lM A23187, 1 lg SR/ml and a defined PTX concentration. The enzyme activity was studied at 25 °C by adding 1 mM ATP (d). The release of inorganic phosphate was evaluated by a colorimetric procedure [30]. (B) The dependence on enzyme concentration was analyzed by studying the inhibitory effect of PTX at the following SR concentrations: 1 (d), 3 (s), 10 (j) or 50 lg/ml (h). Experimental conditions were similar to those described for panel A. The A23187 concentration was raised to 5 lM when 50 lg SR/ml was used.
Fig. 2. PTX affected the dependence of Ca2+-ATPase activity on pCa. The hydrolytic activity was measured at 25 °C in a medium containing 20 mM Mops, pH 7.0, 80 mM KC1, 5 mM MgCl2, 0.5 mM EGTA, a CaCl2 concentration to give a defined pCa, 1 lM A23187, 3 lg SR/ml and 1 mM ATP. Experiments were carried out in the absence (s) or presence of 1 lM PTX (d).
The described effect of PTX as inducer of a pump to channel transformation in the Na+, K+-ATPase [12,15] prompted us to check the ability of the toxin to collapse a pre-formed Ca2+ gradient using the SR membrane preparation. The positive control assay was carried out by adding 1 mM ATP to sealed SR vesicles (3 lg/ml) equil-
R. Coca et al. / Archives of Biochemistry and Biophysics 478 (2008) 36–42
39
ibrated in a medium containing 50 lM 45Ca2+ and 2 mM oxalate. The time course of active Ca2+ transport inside the vesicles showed a lineal dependence during the early time points (Fig. 3A). Moreover, the addition of 5 mM EGTA at time 2 min and 30 s prevented any further Ca2+ entry and gave rise to a gradient equivalent to 1.2 lmol Ca2+/mg protein. Under these conditions, complete discharge of the vesicles was observed when 1 lM A23187 was added at time 6 min and 30 s. In a parallel experiment, sealed SR vesicles were actively loaded with 45Ca2+ and a Ca2+ gradient was established by adding excess EGTA at 2 min and 30 s, as before. In this case, the subsequent addition of 1 lM PTX, instead of A23187, was unable to induce release of Ca2+ (Fig. 3B). This indicates that PTX does not promote any Ca2+ ionophoric activity. The inhibitory effect of PTX can be pinpointed The hydrolytic and transport cycle of the SR Ca2+-ATPase is triggered by enzyme phosphorylation from ATP once the high affinity Ca2+ transport sites are occupied giving rise to the rapid formation and accumulation of EP. This partial reaction was studied at the ice-water temperature by mixing the membrane suspension containing 1 lg/ml of leaky vesicles equilibrated in the presence of 50 lM free Ca2+ with the phosphorylation medium to give a final concentration of 5 lM [c-32P]ATP. The reaction was studied during the first 10 s after acid quenching, electrophoretic protein resolution and quantitation of radioactive 32P. As can be seen, the EP level was maximal at 1 s when samples were phosphorylated in the absence of PTX (Fig. 4). The steady-state level of EP was maintained during the studied time frame. When the phosphorylation process took place in the presence of 1 lM PTX, a rapid EP accumulation was also observed, although the steady-state level was lower than that obtained in the absence of toxin. The Ca2+-dependent activation of the enzyme, previously shown through ATP hydrolysis experiments, was also analyzed measuring the accumulation of EP. The reaction medium composition was as indicated in the above paragraph but free Ca2+ concentration was varied. The accumulated EP was initially measured in the absence of PTX. When the pCa selected was 7.0, a low level of EP was accumulated, however higher EP values were observed as the Ca2+ concentration was raised. The EP accumulation reached a saturation level (Fig. 5). When the phosphorylation experiment was performed in the presence of 1 lM PTX a Ca2+ dependence was observed even though, the maximal EP level was considerable lower than that accumulated when the experiment was performed in the absence of PTX. Incubation of SR vesicles in a medium with excess EGTA allowed exposure of the E2 enzyme conformation to the toxin. The
Fig. 4. PTX decreased the EP level when [c-32P]ATP was added to SR vesicles in the presence of Ca2+. The initial reaction medium was 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.5 mM EGTA, 0.55 mM CaC12, 1 lM A23187 and 1 lg SR/ml in the absence (s) or presence of 1 lM PTX (d). Aliquots of 60 ll placed in an ice bath were phosphorylated by adding 6 ll of 55 lM [c-32P]ATP at 2 107 cpm/nmol. The reaction was stopped at different time intervals by adding 0.5 ml of ice-cold quenching solution. The assay at time zero was performed by adding the quenching solution before radioactive ATP. The evaluation of radioactive EP was carried by autoradiographic procedure once the SR proteins were resolved by electrophoresis at pH 6.3.
subsequent phosphorylation process required the simultaneous addition of [c-32P]ATP and Ca2+ to give final concentrations of 5 lM ATP and 50 lM free Ca2+. Data obtained in the absence of PTX showed rapid accumulation of EP that was patent at 1 s phosphorylation (Fig. 6). The inclusion of 1 lM PTX in the reaction medium also revealed the rapid phosphorylation process although a strong reduction of the EP level was observed when compared with the control experiment. EP decomposition is usually a rate limiting step allowing accumulation of the phosphorylated enzyme under steady-state conditions. The study of this partial reaction was accomplished by a pulse and chase experiment. Thus, leaky SR vesicles at 1 lg/ml and in the presence of 50 lM free Ca2+ were first phosphorylated with 5 lM [c-32P]ATP and excess EGTA was added after 4 s to
Fig. 3. PTX did not permeabilize the SR membrane to Ca2+. Sealed SR vesicles were actively loaded with Ca2+ at 25 °C in a medium containing 20 mM Mops, pH 7.0, 80 mM KC1, 5 mM MgCl2, 0.1 mM EGTA, 0.154 mM 45CaCl2 at 100,000 cpm/nmol (pCa 4.3), 3 lg SR/ml, 2 mM potassium oxalate and 1 mM ATP. Ca2+ loading was arrested at 2 min and 30 s by adding 5 mM EGTA. The subsequent addition of 1 lM A23187 (A) or 1 lM PTX (B) was performed at time 6 min and 30 s. Aliquots of the reaction medium were filtered at different time intervals and processed for evaluation of the 45Ca2+ retained by the vesicles.
40
R. Coca et al. / Archives of Biochemistry and Biophysics 478 (2008) 36–42
Fig. 5. PTX affected the EP dependence on pCa. The initial reaction medium containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.5 mM EGTA, different CaC12 concentrations to yield defined pCa, 1 lM A23187 and 1 lg SR/ml was distributed in 60 ll aliquots. The phosphorylation of samples placed in an ice bath was started by adding 6 ll of 55 lM [c-32P]ATP at 2 107 cpm/nmol. The reaction was stopped after 4 s by adding 0.5 ml of ice-cold quenching solution (s). The experiment was repeated by including 1 lM PTX in the initial reaction medium (d). Quenched samples were processed for quantitation of radioactive EP.
Fig. 7. PTX did not modify the EP decomposition kinetics. Aliquots of 60 ll containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.5 mM EGTA, 0.554 mM CaC12, 1 lM A23187 and 1 lg SR/ml in the absence (s) or presence of 1 lM PTX (d) were placed in an ice bath. Samples were phosphorylated by adding 6 ll of 55 lM [c-32P]ATP at 2 107 cpm/nmol. After 4 s, 5 mM EGTA was added and EP decomposition was followed at different time intervals by adding 0.5 ml ice-cold quenching solution. The EGTA addition corresponded to time zero for dephosphorylation.
interrupt the progression of the reaction cycle. The induced decomposition of radioactive EP was monitored at different time intervals after acid quenching of the samples. The experiment was repeated by including 1 lM PTX in the initial reaction medium. The steady-state level of radioactive EP was higher in the absence than in the presence of PTX (Fig. 7). However, the kinetics of EP decomposition followed monoexponential decay and was very similar whether PTX was present or absent. Curve fitting provided an apparent rate constant of 0.35 s1 for EP decomposition in the presence of PTX and 0.33 s1 when the experiment was performed in the absence of toxin. Discussion
Fig. 6. PTX decreased the EP level when [c-32P]ATP plus Ca2+ were added to SR vesicles in the absence of Ca2+. SR vesicles at 1 lg/ml were initially equilibrated in a medium containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 0.5 mM EGTA and 1 lM A23187 in the absence (s) or presence of 1 lM PTX (d). Aliquots of 60 ll placed in an ice bath were phosphorylated by adding 6 ll of medium containing 20 mM Mops, pH 7.0, 80 mM KCl, 5 mM MgCl2, 6.06 mM CaC12 and 55 lM [c-32P]ATP at 2 107 cpm/nmol. The reaction was stopped at different time intervals by adding 0.5 ml of ice-cold quenching solution. The quenching solution was added before the phosphorylation medium in the assay at time zero. Quenched samples were subjected to electrophoresis and autoradiography for EP determination.
The dependence of the Ca2+-ATPase activity on PTX concentration was consistent with a single enzyme population displaying an apparent inhibition constant of approx. 0.4 lM (Fig. 1A). In terms of molar ratio, a PTX concentration of 2 lM produced practically full effect when the SR concentration was 1 lg/ml. This was equivalent to a PTX/enzyme molar ratio of 500, assuming from the maximal EP level that 1 mg of SR protein contains 4 nmol of Ca2+-ATPase active sites. Therefore, PTX can be considered a strong inhibitor of the enzyme. The effect of Ca2+ on enzyme activity is related with the equilibrium: E2 + 2Ca2+ ? E1Ca2. In fact, the abundance of Ca2+-bound or Ca2+-free enzyme conformations is dependent on pCa, and the accumulation of active conformation E1Ca2 is favored when the Ca2+ concentration is raised. The PTX effect on Ca2+ concentration dependence (Fig. 2) indicated that the Ca2+-dependent activation process was incomplete. This can be explained by an alteration of the Ca2+ binding equilibrium causing stabilization of the enzyme in the inactive E2 conformation. This finding is consistent with the existence of a fraction of fully active enzyme molecules and another fraction of totally inhibited molecules.
R. Coca et al. / Archives of Biochemistry and Biophysics 478 (2008) 36–42
Our data indicated that addition of PTX to Ca2+ loaded vesicles after arrest of the Ca2+ transport activity by EGTA did not alter the permeability of the SR membrane to Ca2+ (Fig. 3). Therefore, the SR Ca2+-pump cannot be transformed into a permanently open Ca2+ channel by the presence of PTX. The described effect of PTX on the Na+-pump did not require a catalytically active enzyme as demonstrated when the Asp369 ? Ala mutation was introduced in the a subunit [34]. When SR vesicles are equilibrated in a saturating Ca2+ medium all the enzyme molecules will be in the E1Ca2 conformation. This allows the study of the phosphorylation reaction, i.e., E1Ca2 + ATP ? E1PCa2 + ADP. The results obtained at the ice-water temperature confirmed that phosphorylation in the absence or presence of PTX was a very fast process (Fig. 4). This was an indication that PTX did not affect enzyme phosphorylation by ATP and was consistent with no dependence on ATP concentration when the enzyme activity was measured as a function of PTX (data not shown). Nevertheless, the accumulated EP level was somewhat lower in the presence than in the absence of PTX. This can be attributed to the effect of PTX on the Ca2+ binding process and is consistent with the lower maximal rate observed when ATP hydrolysis was measured as a function of pCa (Fig. 2). The effect of PTX on EP was also studied by equilibrating SR vesicles in the presence of different free Ca2+ concentrations before phosphorylation. This assay confirmed that the accumulation of EP was related with the free Ca2+ concentration used whether PTX was absent or present in the reaction medium (Fig. 5). It was also shown that phosphorylation by ATP was diminished when the experiment was performed in the presence of 1 lM PTX. When SR vesicles were placed in a Ca2+-free medium to induce the E2 conformation, the phosphorylation experiments required the addition of Ca2+ and ATP. This allowed the study of the following process: E2 + 2Ca2+ + ATP ? E1Ca2 ATP ? E1PCa2 + ADP. Ca2+ binding and phosphorylation by ATP are very rapid partial reactions leading to the steady-state accumulation of EP in less than 1 s when control experiments were carried out (Fig. 6). The addition of 1 lM PTX to the Ca2+-free samples before phosphorylation by ATP in the presence of Ca2+ provoked a clear decrease of the EP levels. Comparison of these results with those reported in Fig. 4 confirms preferential interaction of PTX with the E2 conformation of the enzyme. The effect of PTX on the enzyme turnover was marked by a decreased steady-state EP level without alteration of the EP decomposition kinetics (Fig. 7). These data indicated that formation and decomposition of EP were not affected by PTX and the inhibition can be attributed to a limitation of the enzyme population that can be phosphorylated in the presence of toxin. Stabilization of the E2 conformation and interference with the Ca2+ binding process seems to be the functional effect of PTX on this Ca2+-dependent ATPase. Many hydrophobic compounds with different structural characteristics including thapsigargin [35] and 2,5-di(tert-butyl)-1,4dihydroxibenzene [36] showed inhibitory action of the SR Ca2+ATPase by interfering with the Ca2+ binding process. Crystallographic data have revealed that the common functional effect of thapsigargin [37] and 2,5-di(tert-butyl)-1,4-dihydroxibenzene [38] was to restraint the movement of the enzyme transmembrane helices although the interaction of these two inhibitors occurred at different binding sites. This suggests that binding of different hydrophobic molecules to the transmembrane region that locks the enzyme into the E2 conformation is a prevalent inhibitory mechanism. Despite the close structural analogy among members of the Ptype ATPases, PTX did not transform the SR Ca2+-ATPase into a channel as occurs with the heterodimeric Na+, K+-ATPase. It is significant that the Na+-pump was only sensitive to PTX, and also to
41
ouabain, when both a and b subunits were co-expressed and not when either subunits was expressed alone [12]. Moreover, H+,K+ATPase from colon epithelial cells that was affected by PTX in the same way [22] has also a two-subunit composition. Notably, the Na+-pump a subunit can form functional pump complexes with the b subunit of the H+-pump [39] and a hybrid between Na+,K+ATPase and H+,K+-ATPase is sensitive to PTX [40]. It seems that the PTX effect as a channel inducer is related with the a/b-oligomeric composition of the pump rather than with the primary structure of the protein. It can be concluded that the PTX action on sarcolemmal Na+,K+-ATPase and SR Ca2+-ATPase [24] occurs through different molecular mechanisms. Pilot assays have revealed that PTX from P. tuberculosa does not reproduce the inhibitory profile displayed by PTX from P. caribaeorum on SR Ca2+-ATPase (data not shown). This is consistent with reported differences in the chemical structure of PTX depending on sources and locations and with observed differences in the functional effect on Na+,K+-ATPase [13]. This fact may contribute to the variety of effects that have been attributed to the toxin. Acknowledgment This study was supported by Grant BMC2002-02474 from Ministerio de Educación y Cultura, Spain. References [1] [2] [3] [4] [5] [6]
[7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32]
J.S. Wiles, J.A. Vick, M.K. Christensen, Toxicon 12 (1974) 427–433. R.E. Moore, P.J. Scheuer, Science 172 (1971) 495–498. S. Gleibs, D. Mebs, Toxicon 37 (1999) 1521–1527. M. Fukui, M. Murata, A. Inoue, M. Gawel, T. Yasumoto, Toxicon 25 (1987) 1121–1124. A.M. Kodama, Y. Hokama, T. Yasumoto, M. Fukui, S.J. Manea, N. Sutherland, Toxicon 27 (1989) 1051–1053. J.K. Cha, W.J. Christ, J.M. Finan, H. Fujioka, Y. Kishi, L.L. Klein, S.S. Ko, J. Leder, W.W. McWhorter Jr., K.-P. Pfaff, M. Yonaga, J. Am. Chem. Soc. 104 (1982) 7369– 7371. J.M. Dubois, J.B. Cohen, J. Pharmacol. Exp. Ther. 201 (1977) 148–155. H. Ozaki, J. Tomono, H. Nagase, N. Urakawa, Jpn. J. Pharmacol. 33 (1983) 1155– 1162. M. Ikeda, K. Mitani, K. Ito, Naunyn Schmiedebergs Arch. Pharmacol. 337 (1988) 591–593. H. Bottinger, L. Beress, E. Habermann, Biochim. Biophys. Acta 861 (1986) 165– 176. N.A. Castle, G.R. Strichartz, Toxicon 26 (1988) 941–951. G. Scheiner-Bobis, D. Meyer zu Heringdorf, M. Christ, E. Habermann, Mol. Pharmacol. 45 (1994) 1132–1136. Y. Ishida, K. Takagi, M. Takahashi, N. Satake, S. Shibata, J. Biol. Chem. 258 (1983) 7900–7902. E. Habermann, Toxicon 27 (1989) 1171–1187. P. Artigas, D.C. Gadsby, J. Gen. Physiol. 123 (2004) 357–376. P. Artigas, D.C. Gadsby, Proc. Natl. Acad. Sci. USA 100 (2003) 501–505. S. Guennoun, J.-D. Horisberger, FEBS Lett. 482 (2000) 144–148. S. Guennoun, J.D. Horisberger, FEBS Lett. 513 (2002) 277–281. J.-D. Horisberger, S. Kharoubi-Hess, S. Guennoun, O. Michielin, J. Biol. Chem. 279 (2004) 29542–29550. H. Schneider, G. Scheiner-Bobis, J. Biol. Chem. 272 (1997) 16158–16165. S. Becker, H. Schneider, G. Schneider-Bobis, Eur. J. Biochem. 271 (2004) 3821– 3831. G. Scheiner-Bobis, T. Hübschle, M. Diener, Eur. J. Biochem. 269 (2002) 3905– 3911. K. Ito, N. Urakawa, H. Koike, Arch. Int. Pharmacodyn. Ther. 258 (1982) 146– 154. J. Kockskämper, G.U. Ahmmed, A.V. Zima, K.A. Sheehan, H.G. Glitsch, L.A. Blatter, Am. J. Physiol. Cell Physiol. 287 (2004) C527–C538. S. Eletr, G. Inesi, Biochim. Biophys. Acta 282 (1972) 174–179. O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, J. Biol. Chem. 193 (1951) 265–275. A. Fabiato, Methods Enzymol. 157 (1988) 378–417. G. Schwartzenbach, H. Senn, G. Anderegg, Helv. Chim. Acta 40 (1957) 1886– 1900. J.R. Blinks, W.G. Wier, P. Hess, F. Prendergast, Prog. Biophys. Mol. Biol. 40 (1982) 1–114. P.A. Lanzetta, L.J. Alvarez, P.S. Reinach, O.A. Candia, Anal. Biochem. 100 (1979) 95–97. A. Martonosi, R. Feretos, J. Biol. Chem. 239 (1964) 648–658. B. Sarkadi, A. Enyedi, Z. Földes-Papp, G. Gárdos, J. Biol. Chem. 261 (1986) 9552– 9557.
42 [33] [34] [35] [36]
R. Coca et al. / Archives of Biochemistry and Biophysics 478 (2008) 36–42
K. Weber, M. Osborn, J. Biol. Chem. 244 (1969) 4406–4412. G. Scheiner-Bobis, H. Schneider, Eur. J. Biochem. 248 (1997) 717–723. Y. Sagara, G. Inesi, J. Biol. Chem. 266 (1991) 13503–13506. Y.M. Khan, M. Wictome, J.M. East, A.G. Lee, Biochemistry 34 (1995) 14385– 14393. [37] C. Toyoshima, H. Nomura, Nature 418 (2002) 605–611.
[38] K. Obara, N. Miyashita, C. Xu, I. Toyoshima, Y. Sugita, G. Inesi, C. Toyoshima, Proc. Natl. Acad. Sci. USA 102 (2005) 14489–14496. [39] J.-D. Horisberger, P. Jaunin, M.A. Reuben, L.S. Lasater, D.C. Chow, J.G. Forte, G. Sachs, B.C. Rossier, K. Geering, J. Biol. Chem. 266 (1991) 19131–19134. [40] R.A. Farley, S. Schreiber, S.-G. Wang, G. Scheiner-Bobis, J. Biol. Chem. 276 (2001) 2608–2615.