Purification and properties of a 23 kDa Ca2(+)-binding protein from Drosophila melanogaster.

Recent reports have shown that there exists in mammalian brain a number of heat-stable Ca2(+)-binding proteins that are distinct from calmodulin [McDonald & Walsh (1985) Biochem. J. 232, 559-567]. We have attempted to characterize equivalent Ca2(+)-binding proteins from Drosophila. Affigel-phenothiazine chromatography, which can be used to purify calmodulin and other Ca2(+)-binding proteins, allowed the identification of a possible heat-stable 23 kDa Ca2(+)-binding protein. A purification procedure for this protein has been devised. Purified 23 kDa protein shows characteristics typical of a Ca2(+)-binding protein; there is a mobility shift on SDS/polyacrylamide gels in the presence of EGTA, and Western blotting, followed by the use of the 45Ca2+ overlay technique, confirms that the 23 kDa protein does bind Ca2+. 45Ca2+ binding studies indicate that this protein binds 1 mol of Ca2+/mol of protein, with Kd 1.9 microM. A single band with pI 5.2 is obtained on isoelectric focusing. Analysis of Western blots of Drosophila tissues probed with antibodies to the Ca2(+)-binding protein indicates that it has a widespread distribution, but is absent from muscle tissue. The antibodies also cross-react with a protein of identical molecular mass in extracts of sheep brain. The possible similarity between this Drosophila Ca2(+)-binding protein and mammalian proteins is discussed, and comparison is made between this Drosophila protein and other Ca2(+)-binding proteins purified from vertebrates.     

Spectroscopic, thermodynamic and kinetic properties of Candida nitratophila nitrate reductase.

HL-60 cells possess a 60 kDa Ca2(+)-binding protein that is contained in a discrete subcellular compartment, referred to as calciosomes. Subcellular fractionation studies have suggested that, in HL-60 cells, this intracellular compartment is an Ins(1,4,5)P3-sensitive Ca2+ store. In order to investigate the structural relationship of the 60 kDa Ca2(+)-binding protein of HL-60 cells to other Ca2(+)-binding proteins, we have purified the protein by ammonium sulphate extraction, acid precipitation, and DEAE-cellulose and phenyl-Sepharose column chromatography. The N-terminal sequence of the protein shows 93% identity with rabbit muscle calreticulin, a recently cloned sarcoplasmic reticulum Ca2(+)-binding protein. No amino acid sequence similarity with calsequestrin was found, although the purified protein cross-reacted with anti-calsequestrin antibodies. The calreticulin-related protein of HL-60 cells might play a role as an intravesicular Ca2(+)-binding protein of an Ins(1,4,5)P3-sensitive Ca2+ store.     

Calreticulin is a candidate for a calsequestrin-like function in Ca2(+)-storage compartments (calciosomes) of liver and brain.

In a search for the non-muscle equivalent of calsequestrin (the low-affinity high-capacity Ca2(+)-binding protein responsible for Ca2+ storage within the terminal cisternae of the sarcoplasmic reticulum), acidic proteins were extracted from rat liver and brain microsomal preparations and purified by column chromatography. No calsequestrin was observed in these extracts, but the N-terminal amino acid sequence of the major Ca2(+)-binding protein of the liver microsomal fraction was determined and found to correspond to that of calreticulin. This protein was found to bind approx. 50 mol of Ca2+/mol of protein, with low affinity (average Kd approx. 1.0 mM). A monoclonal antibody, C6, raised against skeletal-muscle calsequestrin cross-reacted with calreticulin in SDS/PAGE immunoblots, but polyclonal antibodies reacted with native calreticulin only weakly, or not at all, after SDS denaturation. Immuno-gold decoration of liver ultrathin cryosections with affinity-purified antibodies against liver calreticulin revealed luminal labelling of vacuolar profiles indistinguishable from calciosomes, the subcellular structures previously identified by the use of anti-calsequestrin antibodies. We conclude that calreticulin is the Ca2(+)-binding protein segregated within the calciosome lumen, previously described as being calsequestrin-like. Because of its properties and intraluminal location, calreticulin might play a critical role in Ca2+ storage and release in non-muscle cells, similar to that played by calsequestrin in the muscle sarcoplasmic reticulum.     

Trifluoperazine binding to porcine brain calmodulin and skeletal muscle troponin C

Binding of trifluoperazine (TFP), a phenothiazine tranquilizer, to porcine brain calmodulin (CaM) and rabbit skeletal muscle troponin C (Tn C) was measured by an automated high-performance liquid chromatography binding assay using a molecular sieving column; 10 micrograms of either protein per injection is sufficient for determining TFP binding, and results are comparable to those obtained by equilibrium dialysis. Very little binding was observed to either protein in the absence of Ca2+ while in the presence of Ca2+ both proteins bind 4 equiv of TFP. Other characteristics of TFP binding however are different for each protein. For CaM, half-maximal binding occurs at 5.8 microM TFP, the Hill coefficient is 0.82, and the fit of the data to the Scatchard equation is consistent with four independent TFP-binding sites. Binding of one melittin displaces two TFP from CaM. Thus, there are two recognizable classes of TFP-binding sites: those that are displaced by melittin and those that are not. TFP causes an increase in the Ca2+ affinity of CaM, and three Ca2+ must be bound to CaM for TFP binding to occur. The studies also yielded a measure of the intrinsic affinity of three of CaM's Ca2(+)-binding sites that is in agreement with previous reports. For troponin C, half-maximal binding occurs at 16 microM TFP, the Hill coefficient is 1.7, and the data best fit the Adair equation for four binding sites. The measured constants K1, K2, K3, and K4 were 2.5 X 10(4), 6.6 X 10(3), 5.8 X 10(5), and 2.0 X 10(5) M-1, respectively, in 1 mM Ca2+ and were similar when Mg2+ was additionally included. TFP also increases troponin C's Ca2+ affinity, and it is the low-affinity, Ca2(+)-specific binding sites that are affected. These studies yielded a measure of the intrinsic affinity of these Ca2(+)-binding sites that is in agreement with previous measurements.     

Calcium-regulated interactions of human alpha-lactalbumin with bee venom melittin

Affinity chromatography, fluorescence and circular dichroism spectroscopy methods have been used to study the interaction of melittin, a 26-residue peptide from bee venom, with Ca2(+)-binding alpha-lactalbumin from human milk. It has been revealed that melittin binds to the apo- and acidic states of alpha-lactalbumin while the presence of Ca2+ makes the interaction essentially weaker. The association constant for the complex of melittin with apo-alpha-lactalbumin determined from spectropolarimetric melittin-titration data is 2 X 10(7) M-1. The complexation of alpha-lactalbumin with melittin decreases its affinity to Ca2+ by three orders of magnitude. The interaction of apo-alpha-lactalbumin with melittin causes some changes in the environment of its aromatic amino acid residues and drastically alters the conformation of melittin, increasing its alpha-helical content but leaving its single tryptophan residue accessible to water. In the case of the acidic state of alpha-lactalbumin the interaction does not induce an increase in alpha-helical content of melittin.     

Characterization of functionally important domains in human vitamin K-dependent protein S using monoclonal antibodies.

Vitamin K-dependent protein S is an anticoagulant plasma protein functioning as a cofactor to activated protein C in the degradation of coagulation factors Va and VIIIa. To determine which regions in protein S are important for its cofactor activity, we have raised and characterized a large panel of monoclonal antibodies against human protein S. Several of the antibodies were directed against Ca2(+)-dependent epitopes, and they were found to be located either in the domain containing gamma-carboxyglutamic acid (Gla), the thrombin-sensitive region, or in the first epidermal growth factor (EGF)-like domain. The first two types of epitopes were exposed at approximately 1 mM Ca2+, whereas the epitope(s) in the EGF-like domains required less than 1 microM Ca2+, suggesting the presence of one or more high affinity Ca2(+)-binding site(s). The antibodies, as well as their Fab' fragments, against all three types of Ca2(+)-dependent epitopes efficiently inhibited the activated protein C cofactor function of protein S, but through different mechanisms. The antibodies against the Gla domain exerted their effects through inhibition of protein S binding to negatively charged phospholipid. Fab'-fragments of antibodies against the thrombin-sensitive region and the first EGF-like domain were the most potent inhibitors of the activated protein C cofactor function but did not inhibit phospholipid binding of protein S. In conclusion, we have identified the domains in protein S that are important for the activated protein C cofactor activity. The Gla domain is instrumental in the binding of protein S to phospholipid, whereas the thrombin-sensitive region and the first EGF-like domain may be directly involved in protein-protein interactions on the phospholipid surface.     

Evidence for a calsequestrin-like calcium-binding protein in human spermatozoa

Human spermatozoa were investigated for the presence of protein(s) recognized by antibodies against calsequestrin, the high capacity, moderate affinity Ca2(+)-binding protein, originally described in striated muscle fibers. Western immunoblots of detergent-soluble sperm extracts probed with polyclonal antibodies raised against human skeletal muscle calsequestrin identified a strongly cross-reactive protein. This protein resembles muscle calsequestrin in many respects. In fact, its migration in sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) is pH dependent, its apparent molecular mass being 64 kDa in alkaline SDS-PAGE and 44 kDa in neutral SDS-PAGE; its isoelectric point is acidic (4.6); it is metachromatically stained blue by the carboxycyanine dye, Stains-All; it is a Ca2(+)-binding protein (45Ca blot overlay). Indirect immunofluorescence experiments showed that the immunoreactive protein has an intracellular localization confined to the tail mid-piece. From these findings we conclude that human sperm cells express a protein structurally and antigenically related to skeletal muscle calsequestrin; a basis for a novel interpretation of Ca2(+)-mediated events in spermatozoa is thus provided.     

Solubilization of guanylyl cyclase from bovine rod outer segments and effects of lowering Ca2+ and nitro compounds.

Guanylyl cyclase from bovine rod outer segments was solubilized using Triton X-100 and a high concentration of KCl, and its regulation was studied. The efficiency of solubilization was about 50-90% of total activity. When the Ca2+ content was lowered (less than 80 nM), guanylyl cyclase was activated about 2-fold. In the presence of higher concentrations of Ca2+ (greater than 140 nM), the activity was decreased. The regulation by Ca2+ was also demonstrated with solubilized preparations. In the presence of 186 nM Ca2+ which inhibited guanylyl cyclase, La3+ activated the enzyme about 2-fold, suggesting that the Ca2(+)-binding protein similar to other Ca2(+)-binding proteins associates with guanylyl cyclase regulation. Sodium nitroprusside and nitric oxide which are activators of soluble guanylyl cyclase in other tissues also activated the retinal guanylyl cyclase. Maximum activation by sodium nitroprusside was 20-fold using Mg2+ as a cofactor. Activation by nitric oxide and related compounds suggests that retinal guanylyl cyclase contains a heme prosthetic group that may participate in a novel regulatory mechanism for this enzyme.     

Terbium as a luminescent probe of metal-binding sites in protein kinase C

In the present report, we demonstrate that Tb3+ binds to protein kinase C and serves as a luminescent reporter of certain cationic metal-binding sites. Tb3+ titration of 50 nM protein kinase C results in a 20-fold enhancement of Tb3+ luminescence which is half-maximal at 12 microM Tb3+. A Kd of approximately 145 nM was determined for Tb3+ binding to the enzyme. The excitation spectrum of bound Tb3+ exhibits a peak at 280 nm characteristic of energy transfer from protein tryptophan or tyrosine residues. The luminescence of this complex can be markedly decreased by other metals, including Pb2+ (IC50 = 25 microM), La3+ (IC50 = 50 microM), Hg2+ (IC50 = 300 microM), Ca2+ (IC50 = 6 mM), and Zn2+ (IC50 greater than 10 mM), and chelation of Tb3+ by 2 mM EGTA. Tb3+ binding to protein kinase C is correlated with its inhibition of protein kinase activity (IC50 = 8 microM), r = 0.99) and phorbol ester binding (IC50 = 15 microM, r = 0.98). Tb3+ inhibition of protein kinase C activity cannot be overcome by excess Ca2+, but can be partially overcome with excess phosphatidylserine or by chelation of Tb3+ with EGTA. Tb3+ noncompetitively inhibits phorbol ester binding by decreasing the maximal extent of binding without significantly altering binding affinity. The results suggest that the Tb3(+)-binding site is at or allosterically related to the enzyme's phosphatidylserine-binding site, but is distinct from the phorbol ester-binding domain and the Ca2(+)-binding site that regulates enzyme activity.     

Three-dimensional structure of a sarcoplasmic calcium-binding protein from Nereis diversicolor

The three-dimensional structure of a sarcoplasmic Ca2(+)-binding protein from the sandworm Nereis diversicolor has been determined at 3.0 A resolution using multiple isomorphous replacement techniques. The NH2-terminal half of the molecule contains one variant Ca2(+)-binding domain with a novel helix-loop-helix conformation and one Ca2(+)-binding domain that is no longer functional because of amino acid changes. The overall conformation of this pair of domains is different from any previously described Ca2(+)-binding protein. The COOH-terminal half of the protein contains two Ca2(+)-binding domains with the usual helix-loop-helix configuration and is similar to calmodulin and troponin C. Unlike calmodulin or troponin C, there is no exposed alpha-helix connecting the two halves of the molecule, so the overall structure is much more compact.     

Calbindin-D28K, a 1 alpha,25-dihydroxyvitamin D3-induced calcium-binding protein, binds five or six Ca2+ ions with high affinity

Calbindin-D28K is a 1 alpha,25-dihydroxyvitamin D3-dependent protein that belongs to the superfamily of high affinity calcium-binding proteins which includes parvalbumin, calmodulin, and troponin C. All of these proteins bind Ca2+ ligands by an alpha-helix-loop-alpha-helix domain that is termed an EF-hand. Calbindin-D28K has been reported previously to have four high affinity Ca2(+)-binding sites (KD less than 10(-7)) as quantitated by equilibrium dialysis. With the determination of the amino acid sequence, it was clear that there are in fact six apparent EF-hand domains, although the Ca2(+)-binding functionality of the two additional domains was unclear. It was of interest to quantitate the Ca2(+)-binding ability of chick intestinal calbindin-D28K utilizing several different Ca2+ titration methods that cover a range of macroscopic binding constants for weak or strong Ca2+ sites. Titrations with the Ca2+ chelator dibromo-1,2-bis(2-aminophenoxy)-ethane-N,N,N',N'-tetraacetic acid (5,5'-Br2BAPTA), a Ca2+ selective electrode, and as followed by 1H NMR, which measure KD values of 10(-6)-10(-8) M, 10(-4)-10(-7) and 10(-3)-10(-5) M, respectively, gave no evidence for the presence of weak Ca2(+)-binding sites. However, Ca2+ titration of the fluorescent Ca2+ chelator Quin 2 in the presence of calbindin-D28K yielded a least squares fit optimal for 5.7 +/- 0.8 Ca2(+)-binding sites with macroscopic dissociation constants around 10(-8) M. The binding of Ca2+ by calbindin was found to be cooperative with at least two of the sites exhibiting positive cooperativity.     

Calcium-binding affinity and calcium-enhanced activity of Clostridium thermocellum endoglucanase D.

Clostridium thermocellum endoglucanase D (EC 3.2.1.4: EGD), which is encoded by the celD gene, was found to bind Ca2+ with an association constant of 2.03 x 10(6) M-1. Ca2+ stimulated the activity of EGD towards swollen Avicel by 2-fold. In the presence of Ca2+, the Kd of the enzyme towards p-nitrophenyl-beta-D-cellobioside and carboxymethylcellulose was decreased by 4-fold. Furthermore, Ca2+ increased the half-life of the enzyme at 75 degrees C from 13 to 47 min. Since the 3' sequence of celD encodes a duplicated region sharing similarities with the Ca2+-binding site of several Ca2+-binding proteins, a deleted clone was constructed and used to purify a truncated form of the enzyme which no longer contained the duplicated region. The truncated enzyme was very similar to EGD expressed from the intact gene with respect to activity, Ca2(+)-binding kinetics and Ca2+ effects on substrate binding and thermostability. Thus the latter parameters do not appear to be mediated through the duplicated conserved region.     

Altered synthesis of the 26-kDa heat stress protein family and thermotolerance in cell lines with elevated levels of calcium-binding proteins

Using a bovine papilloma virus-based vector, mouse mammary adenocarcinoma cells have been transformed to express elevated amounts of functional calmodulin (CaM) (Rasmussen and Means, 1987) and another Ca2(+)-binding protein, parvalbumin (PV) (Rasmussen and Means, 1989) that is not normally synthesized in these cells. Parental cells (C127) and cells transformed by the vector alone (BPV-1), the vector containing a CaM gene (CM-1), or the vector containing parvalbumin (PV-1) were used to study the effect of increased synthesis of Ca2(+)-binding proteins on heat-stress protein (HSP) synthesis and cell survival following heating at 43 degrees C. The induction, stability, and repression of the synthesis of most HSPs after 43 degrees C heating was not significantly affected by increased amounts of Ca2(+)-binding proteins, but the rate of synthesis of all three isoforms of the 26-kDa HSP (HSP26) was greatly reduced. C127 cells, which have about one half as much CaM as do BPV-1 cells, synthesized the most HSP26. CM-1 cells, which have more than fourfold higher levels of CaM than do BPV-1 cells, had a rate of synthesis of HSP26 approaching that of unheated cells. BPV-1 cells, with a two-fold increase in CaM, were intermediate in HSP26 synthesis. This effect on HSP26 synthesis may be largely related to the Ca2(+)-binding capacity of CaM rather than to a specific CaM-regulated function, since PV-1 cells also showed reduced rates of HSP26 synthesis. Survival experiments showed that reduced HSP26 synthesis in cells with increased amounts of Ca2(+)-binding proteins did not significantly alter intrinsic resistance to continuous 43 degrees C heating. Thermotolerance was not reduced and appeared to develop more rapidly in CM-1 and PV-1 cells. These results suggest that (1) the signal for HSP26 synthesis can be largely abrogated by elevated Ca2+ binding protein levels, and (2) if these HSPs are involved in thermotolerance development, that function may be associated with intracellular Ca2+ homeostasis.     

Inhibition by melittin of phospholipid-sensitive and calmodulin-sensitive Ca2+-dependent protein kinases.

Effects of melittin, an amphipathic polypeptide, on various species of protein kinases were investigated. It was found that melittin inhibited the newly identified phospholipid-sensitive Ca2+-dependent protein kinase (from heart, brain, spleen and neutrophils) and the cardiac myosin light-chain kinase, a calmodulin-sensitive Ca2+-dependent enzyme. In contrast, melittin had little or no effect on either the holoenzymes of the cardiac cyclic AMP-dependent and cyclic GMP-dependent protein kinases or the catalytic subunit of the former. Kinetic analysis indicated that melittin inhibited phospholipid-sensitive Ca2+-dependent protein kinase non-competitively with respect to ATP (Ki = 1.3 microM); although exhibiting complex kinetics, its inhibition of the enzyme was overcome by phosphatidylserine (a phospholipid cofactor), but not by protein substrate (histone H1) or Ca2+. On the other hand, melittin inhibited myosin light-chain kinase non-competitively with respect to ATP (Ki = 1.4 microM) or Ca2+ (Ki = 1.9 microM), and competitively with respect to calmodulin (Ki = 0.08 microM); although exhibiting complex kinetics, its inhibition of the enzyme was reversed by myosin light chains (substrate protein). The present findings indicate the presence of functionally important hydrophobic or hydrophilic loci on the Ca2+-dependent protein kinases, but not on the cyclic nucleotide-dependent class of protein kinase, with which melittin can interact. Moreover, the kinetic data suggest that melittin inhibited myosin light-chain kinase by interacting with a site on the enzyme the same as, or proximal to, the calmodulin-binding site, thus interfering with the formation of active enzyme-calmodulin-Ca2+ complex.     

Purification and expression of gCap39. An intracellular and secreted Ca2(+)-dependent actin-binding protein enriched in mononuclear phagocytes

A protein of approximately 40 kDa was the major Ca2(+)-binding protein purified by Ca2(+)-dependent hydrophobic affinity chromatography from the cell lysates and conditioned media of RAW macrophages. Other Ca2(+)-binding proteins, including several annexins (calelectrins), S100-like proteins, and calmodulin, were less abundant and preferentially found in the cell lysates. Amino acid sequences of tryptic fragments from the purified 40-kDa protein revealed its identity to gCap39, an actin-binding protein encoded by a cDNA isolated on the basis of its homology with gelsolin. When an expression vector containing the gCap39 coding region was transfected into COS cells, high levels of gCap39 were found in both the cells and conditioned media, whereas annexins were only present in the cells. gCap39 could also be purified from human plasma where it appeared to be a minor component. No signal sequence was detected in the primary structure of gCap39 and the secreted and intracellular forms of gCap39 are of identical size, suggesting that unlike gelsolin, the mechanism of gCap39 secretion may not depend on a signal sequence. The high concentration of gCap39 in macrophages and its constitutive secretion as well as intracellular retention suggest that this protein may have a dual role in macrophage function, namely that of a Ca2(+)- and polyphosphoinositide-regulated intracellular modulator of the cytoskeleton as well as that of a secreted protein involved in the clearance of actin from the extracellular environment.     

Interaction of ruthenium red with Ca2(+)-binding proteins.

The interaction of ruthenium red, [(NH3)5Ru-O-Ru(NH3)4-O-Ru(NH3)5]Cl6.4H2O, with various Ca2(+)-binding proteins was studied. Ruthenium red inhibited Ca2+ binding to the sarcoplasmic reticulum protein, calsequestrin, immobilized on Sepharose 4B. Furthermore, ruthenium red bound to calsequestrin with high affinity (Kd = 0.7 microM; Bmax = 218 nmol/mg protein). The dye stained calsequestrin in sodium dodecyl sulfate-polyacrylamide gels or on nitrocellulose paper and was displaced by Ca2+ (Ki = 1.4 mM). The specificity of ruthenium red staining of several Ca2(+)-binding proteins was investigated by comparison with two other detection methods, 45Ca2+ autoradiography and the Stains-all reaction. Ruthenium red bound to the same proteins detected by the 45Ca2+ overlay technique. Ruthenium red stained both the erythrocyte Band 3 anion transporter and the Ca2(+)-ATPase of skeletal muscle sarcoplasmic reticulum. Ruthenium red also stained the EF hand conformation Ca2(+)-binding proteins, calmodulin, troponin C, and S-100. This inorganic dye provides a simple, rapid method for detecting various types of Ca2(+)-binding proteins following electrophoresis.     

Effect of myosin cross-bridge interaction with actin on the Ca2(+)-binding properties of troponin C in fast skeletal myofibrils

Ca2+ binding to fast skeletal muscle troponin C reincorporated into troponin C-depleted (CDTA-treated) myofibrils has been measured directly by using 45Ca and indirectly by using a fluorescent probe. Direct Ca2(+)-binding measurements have shown that the Ca2+ affinity of the low-affinity sites is enhanced in the absence of ATP and conversely reduced when myosin is selectively extracted from myofibrils, compared to the Ca2+ affinity in the presence of ATP. Fluorescence intensity changes of a dansylaziridine label at the Met-25 residue of troponin C have shown the same Ca2(+)-sensitivity whether or not ATP is present, while much lower Ca2(+)-sensitivity is seen in the myosin-extracted myofibrils. Since the Met-25 residue is in the amino terminal side alpha-helix of Ca2(+)-binding site I and far from Ca2(+)-binding site II in the primary structure, Ca2+ binding to site II has been evaluated by assuming that the fluorescence change monitors Ca2+ binding to site I alone. Ca2+ binding to site II thus estimated has shown high positive cooperativity only in the presence of ATP and has been found to be nearly proportional to the activation of myofibrillar ATPase, suggesting that Ca2(+)-binding site II is directly involved in the activation of myofibrillar ATPase activity. On the other hand, Ca2(+)-binding site I has been suggested to regulate the interaction of weakly binding cross-bridges with the thin filament, since the fluorescence change in the presence of ATP is saturated at the free Ca2+ concentration required for the activation of myofibrillar ATPase.     

Use of site-directed mutations in the individual Ca2(+)-binding sites of calmodulin to examine Ca2(+)-induced conformational changes

Mutant versions of the calmodulin of Drosophila melanogaster have been prepared for use in the study of Ca2+ binding and Ca2(+)-induced conformational changes. In each mutant, a conserved glutamic acid residue indicated to play a critical role in Ca2+ binding has been mutated to glutamine in one of the Ca2(+)-binding sites. Thus a series of four proteins, each with an analogous mutation in one of the four binding sites, has been generated. Here the Ca2(+)-induced conformational changes in these proteins have been examined by use of the fluorescent hydrophobic reporter molecule, 9-anthroyl choline. These studies confirm earlier work which indicates that the carboxyl-terminal pair of Ca2(+)-binding sites shows cooperative Ca2+ binding to produce a major conformational change in the protein. However, these studies provide evidence that the sites of the amino-terminal pair are more independent in their Ca2+ binding properties and contribute individually to the conformational changes associated with Ca2+ binding in the amino-terminal half of the protein. This work also indicates that mutation of either of the amino-terminal Ca2(+)-binding sites can influence the conformational change produced by Ca2+ binding to the carboxyl-terminal sites.     

Phospholipase A2-independent Ca2+ entry and subsequent apoptosis induced by melittin in human MG63 osteosarcoma cells

Melittin, a peptide from bee venom, is thought to be a phospholipase A(2) activator and Ca(2+) influx inducer that can evoke cell death in different cell types. However, the effect of melittin on cytosolic free Ca(2+) concentration ([Ca(2+)](i)) and viability has not been explored in human osteoblast-like cells. This study examined whether melittin altered [Ca(2+)](i) and killed cells in MG63 human osteosarcoma cells. [Ca(2+)](i) changes and cell viability were measured by using the fluorescent dyes fura-2 and WST-1, respectively. Melittin at concentrations above 0.075 microM increased [Ca(2+)](i) in a concentration-dependent manner. The Ca(2+) signal was abolished by removing extracellular Ca(2+). Melittin-induced Ca(2+) entry was confirmed by Mn(2+) quenching of fura-2 fluorescence at 360 nm excitation wavelength which was Ca(2+)-insensitive. The melittin-induced Ca(2+) influx was unchanged by modulation of protein kinase-C activity with phorbol 12-myristate 13-acetate (PMA) and GF 109203X, or inhibition of phospholipase A(2) with AACOCF(3) and aristolochic acid; but was substantially inhibited by blocking L-type Ca(2+) channels. At concentrations of 0.5 microM and 1 microM, melittin killed 33% and 45% of cells, respectively, via inducing apoptosis. Lower concentrations of melittin failed to kill cells. The cytotoxic effect of 1 microM melittin was completely reversed by pre-chelating cytosolic Ca(2+) with BAPTA. Taken together, these data showed that in MG63 cells, melittin induced a [Ca(2+)](i) increase by causing Ca(2+) entry through L-type Ca(2+) channels in a manner independent of protein kinase-C and phospholipase A(2) activity; and this [Ca(2+)](i) increase subsequently caused apoptosis.