- Email: [email protected]
Identification of bovine brain calcium binding proteins
Cell Calcium Lungman
t
(1987) 8. 229-239 Group UK Ltd 1987
IDENTIFICATION OF BOVINE BRAIN CALCIUM BINDING PROTEINS Masaaki Tokuda, Navin C. Rhanna and David M. Waisman Cell Regulation Group, Department of Medical Biochemistry, The University of Calgary, Calgary, Alberta, Canada T2N 4Nl
(reprint requests to DMW) ABSTRACT Three peaks of calcium binding activity have been identified by the Chelex-106 calcium binding assay of the fractions from DEAE cellulose These chromatography of 100,000 x g supernatant of bovine brain. calcium binding activity peaks have been subjected to extensive purification and three novel calcium binding proteins (Mr 27,000, Mr 48,000 and Mr 63,000) and two previously characterized proteins (calcineurin and calmodulin) have been identified as components of calcium binding activity peaks. Analysis of the calcium binding properties of the novel proteins by equilibrium dialysis suggests these proteins may be intracellular calcium receptors. INTRODUCTION Calcium is believed.to play a crucial role in the process of synaptic transmission. When an action potential arrives at a presynaptic nerve terminal the resultant membrane depolarization opens calcium channels in the presynaptic membrane. The resultant influx of calcium leads to a transient rise in the intracellular calcium concentration;this is believed to be the intracellular signal for neurotransmitter release Electrophysiological and morphological studies have indicated (1). that the calcium induced fusion of synaptic vesicles with the presynaptic membrane is a key event in the release process (2). Changes in cytoplasmic calcium have multiple consequences on synaptic function. Nerve terminal glycogenolysis and respiration are increased in response to the calcium signal (3). Cytoplasmic calcium also stimulates the synthesis of several neurotransmitters by increasing the activity of rate limiting transporters of enzymes. For example, choline transport in cholinergic neurons, and tyrosine hydroxylase in catecholamine neurons are activated by increases in cytoplasmic calcium. Cytoplasmic calcium also reduces the activity of certain calcium channels and activates calcium dependent potassium channels resulting in changes in synaptic efficiency (reviewed in 2).
229
Fundamental to an understanding of the molecular events involved in the calcium dependent regulation of synaptic function is the identification and characterization of the various intracellular calcium binding proteins of the brain since these proteins are ultimately responsible for mediating the actions of cytoplasmic calcium. Calmodulin has been shown to be present in high concentrations in nerve terminals (4) and has been suggested to be a major, but not exclusive calcium binding protein and mediator of calcium transduction in the synapse. Other calcium binding proteins such as parvalbumin (5), calcineurin (6), S100 (7), synaptophysin (8) and synexin (2) have been identified in the brain tissue and elucidation of the function of these proteins in brain tissue is of obvious importance. Considering the fundamental importance of calcium binding proteins to the calcium transduction process, identification of the complete spectrum of intracellular brain calcium binding proteins is essential. Our approach to the identification of the intracellular calcium binding proteins has involved chromatography of the 100,000 x g supernatants of a variety of tissues on DEAE cellulose and analysis of the resultant fractions for calcium binding activity by the Chelex 100 competitive calcium binding assay (9) or by 45Ca2+ au&radiography (10). Results from these studies have suggested the existence in the 100,000 x g supernatants of bovine liver, heart, and brain of several novel calcium binding proteins (10, 11). In the present communication we report on the complete analysis of the 100,000 x g supernatant of bovine brain by the Chelex 100 assay. The three major peaks of calcium binding activity resolved by this procedure have been purified and the calcium binding proteins responsible for these calcium binding activity peaks identified. MATERIALS AND METHODS Chelex-100 and hydroxylapatite were obtained from Bio-Rad. Diisodithiothreitol propylfluorophosphate benzamidine, (DTT), (DFP), phenylmethylsulfonylfluoride (PMSF) and anti-rabbit IgG antibodies peroxidase conjugated were obtained from Sigma. DEAE-cellulose (DE-52) was obtained from Whatman. 45CaC12 was purchased from Amersham. DEAE Cellulose Chromatoeranhv of Bovine Brain 100.000 x e sunernatant The chromatographic procedure for the purification of bovine brain calcium binding proteins is summarized in Fig 1. One kg of frozen bovine brain was chopped and then further minced in a meat grinder. The minced tissue was mixed with 3.0 liters of ice cold buffer containing 40 mM Tris-HCl (pH 8.0), 0.5 mM PMSF, 100 mg/l soybean trypsin inhibitor, 1.0 mM DFP, 50 ml of packed Chelex-100, 5 mg/l leupeptin, 5.0 mM benzamidine and 1.0 mM DTT, and then homogenized in a Waring blender. The resultant extract was centrifuged at 20,000 x g for 30 min and the supernatant was then centrifuged at 100,000 x g for 60 min. The 100,000 x g supernatant was diluted with five volumes of buffer A (40 mM Tris-HCl (pH 8.0), 0.5 mM PMSF, 5.0 mM benzamidine, 1.0 mM DTT) and 600 ml of packed DEAE-cellulose was then added. The mixture was
230
stirred rapidly for 90 min and then filtered through a coarse sintered The resultant slurry was washed with 6.0 liters of glass funnel. buffer A and poured into a 5.0 x 60 cm column. Proteins were eluted with 4.5 liters of a linear NaCl gradient (O-O.45 M) in buffer A. The resulting fractions were analyzed for both calcium binding activity using the Chelex-100 competitive calcium binding assay (9) and calmodulin activity assayed by phosphodiesterase activity (12). As shown in Figure 2, three peaks of calcium binding activity (Peak I-III) were resolved. Each peak was pooled separately, and subjected to further purification.
100,000 x g sup. I
DEAE /
Peak I
Ccl lulose I
Peak11
\
Peak III
Calregulin Figure 1. The scheme of purification and identification of calcium binding proteins from bovine brain.
FRACTION
NUMBER
Figure 2. DEAE cellulose chromatography of 100,000 x g supernatant of bovine brain. Chromatography was performed as described under "Materials and Methods". 231
Purification of CAB-27 from Peak I Peak I was concentrated by ultrafiltration (PM-lo, Amicon) and then dialyzed against buffer B (10 mM sodium-phosphate buffer (pH 6.8), 0.5 mM PMSF, 5.0 mM benzamidine, 1.0 mM DTT) at 4°C for overnight. The dialyzed material was applied to a 5.0 x 20 cm column of hydroxylapatite previously equilibrated with buffer B. The column was washed with two column volumes of buffer B and eluted with 2.5 liters of linear phosphate gradient (lo-200 mM) in buffer B. Fractions were collected and analyzed for calcium binding activity (Figure 3a). The calcium binding activity peak was pooled and heated at 80°C for 2
% ‘.“T
9
x-
4-L. 20
a
B z' 2 d 5 0.28 5
20
40 FRACTION
60 NUMBER
80
0.1% 2 E! OaI 100 FRACTION
NUMBER
Figure 3. Identification of CAB-27 from Peak-I. Peak-I was chromatographed on a hydroxylapatite column as described under "Materials and Methods" (a). Calcium binding activity peak was pooled and heat treated at 80°C for 2 min. Denatured protein was removed by centrifugation at 30,000 x g for 30 min, and the supernatant was applied to HPLC gel permeation chromatography (b). min and the denatured protein was removed by centrifugation at 100,000 x g for 60 min. The supernatant was concentrated to 2.0 ml by ultrafiltration (PM-lo), and applied to a high performance gel permeation liquid chromatography column (TSK-G3000 SW, LKB) and eluted with buffer C (40 mM MOPS, (pH 7.1), 150 mM NaCl, 1.0 mM DTT) at a flow rate of 30 ml/hr. Fractions were collected and assayed for calcium binding The peak of calcium binding activity was activity (Figure 3b). analyzed by SDS-PAGE and the Mr 27,000 protein which corresponded with the calcium binding activity peak was pooled. Purification and identification of CAB-48 and calcineurin from Peak II Peak II (from DEAE cellulose, Figure 2) was chromatographed on a hydroxylapatite column using the identical procedure as described for Peak I except the phosphate gradient was lo-350 mM in buffer B. Fractions were assayed for calcium binding activity, and two major activity peaks (Figure 4a, Peak IIa and IIb) were resolved. Each peak
232
was pooled,
concentrated
(PM-lo) and then applied
to TSK-G3000
% f; 0.4W ‘, E
&
O.l$
2
8 z or
FRACTION
SW
oo” . . .._I*-.__. 20 _
40 FRACTION
60 NUMBER
NUMBER
80
--
100
FRACTION
NUMBER
Figure 4. Purification of CAB-48 and identification of calcineurin from Peak-II. Peak-II was chromatographed on a hydroxylapatite column as described under "Materials and Methods" (a). HPLC gel permeation chromatography was performed for two calcium binding activity peaks, Peak-IIa (b) and Peak-IIb (c).
column (IIa-Figure 4b, IIb-Figure 4~). The peak of calcium binding activity resulting from gel permeation chromatography of Peak 16 (Figure 5) was further purified by chromatography on a calmodulin sepharose affinity column according to Sharma et al. (17). The peak of calcium binding activity resulting from gel permeation chromatography of Peak IIa (Figure 4b) was analyzed by SDS PAGE and a homogeneous
233
protein band of Mr 48,000 was found to correlate with calcium binding
0
20
40 FRACTION
60
60
NUMBER
Identification of CAB-63 and calmodulin from Peak-III. Figure 5. Peak-III was chromatographed on a HPLC gel permeation column as described under "Materials and Methods". Immunoblot analysis using a rabbit anti-CAB-63 antibody were performed (inset). Lane 1: 5 pg of purified bovine liver CAB-63, lane 2: 20 pg protein of fraction 40.
Identification of calregulin and calmodulin from Peak III Peak III was concentrated by ultrafiltration (PM-lo) and applied to TSK-G3000 SW column. Proteins were eluted with buffer C at a flow rate of 30 ml/hr. Fractions were collected and analyzed for both calregulin (CAB-63) concentration by an ELISA using a rabbit polyclonal antibody against calregulin (11) and calmodulin activity (12) (Figure 5). Both peaks were pooled separately and subjected to further purification according to published procedures (peak IIIa (ll), peak IIIb (13)). Other Methods Immunoblot analysis (ll), polyacrylamide gel electrophoresis (14), protein determination (15), calcineurin activity (16), Chelex-100 competitive calcium binding assay (9) and equilibrium dialysis (17) were performed as described. RESULTS AND DISCUSSION Bovine brain 100,000 x g supernatant was chromatographed on DEAEcellulose column and fractions were assayed for calcium binding As shown in Figure 2, three peaks of calcium binding activity. activity were resolved (Peak I-III) and only Peak III coincided with
234
calmodulin activity. In order to identify the calcium binding protein responsible for calcium binding activity peaks an extensive purification procedure was developed. The procedure for the purification of the calcium binding proteins responsible for the calcium binding activity peaks is presented in Figure 1. Peak I was chromatographed on a hydroxylapatite column, and only one The peak calcium binding activity peak was resolved (Figure 3a). fractions were pooled and assayed for heat stability of calcium binding activity. About 85% of the activity was retained after heating at 80°C for 2 min. Therefore pooled peak I was heat treated, denatured protein removed by centrifugation, and the supernatant was concentrated and chromatographed by HPLC gel permeation chromatography (Figure 3b). A single peak of calcium binding activity migrating with an apparent Mr 40,000 was resolved. Analysis of this peak by SDS-PAGE demonstrated a single protein band of Mr 27,000 (Figure 6).
b
t LOG [Co2+]
Figure 6. SDS-polyacrylamide gel electrophoresis of major calcium binding proteins. Five. major calcium binding proteins from bovine brain are analyzed on 12.5% polyacrylamide gel electrophoresis contained 0.1% SDS. a: molecular weight standards, b: CAB-27 (3 pg), c: CAB-48 (3 pg), d: calcinuerin (6 pg), e: CAB-63 (3 pg), f: calmodulin (3 /Jg) Figure 7. Calcium binding saturation curves 'of major calcium binding proteins. Calcium binding assay was performed by equilibrium dialysis in the presence of 3 mM MgC12 and 150 mM KC1 in the MOPS buffer (pH 7.1), for purified CAB-27 (0), CAB-48 (O), and CAB-63 (m). Data for calmodulin (0) was used from ref. 18. This protein was identical to the Mr 27,000 calcium binding protein identified by 45Ca2+ -autoradiography of bovine brain 100,000 x g supernatants (10). These results suggest that the calcium binding
235
activity of peak I is due to CAB-27. Hydroxylapatite chromatography of Peak II resolved two calcium binding activity peaks between 60 and 100 mM phosphate (Peak IIa) and between 120 and 160 mM phosphate (Peak IIb) (Figure 4a). Peak IIa was chromatographed on HPLC gel permeation column and two calcium binding activity peaks migrating with apparent Mr 75,000 at 72,000 (Figure 4b) were detected. The higher molecular weight peak (Peak IIa-1) was identified by SDS-PAGE as a single homogeneous protein band of Mr 48,000 (Figure 6) which we named CAB-48. The lower molecular weight calcium binding activity peak (Peak IIa-2) was not homogeneous by SDS PAGE. Since the SDS-PAGE analysis of the peak fractions revealed prominent bands at Mr 60,000 and Mr 14,500 we suspected that this calcium binding activity peak contained calcineurin. Interestingly, calcineurin activity was found to correlate with calcium binding activity. Chromatography of this calcium binding activity peak on calmodulin-sepharose (19) conserved the calcium binding activity as a single protein peak (data not shown) and SDS PAGE resolved two protein bands of Mr 60,000 and Mr 14,500. HPLC gel permeation chromatography of Peak IIb resolved a single calcium binding activity peak (Peak IIb-1) migrating with an apparent Mr 72,000 which also coincided with calcineurin activity (Figure 4~). Subsequent chromatography of this peak on calmodulin-sepharose (19) resolved a single peak of calcium binding activity (data not shown) and SDS-PAGE analysis revealed two protein bands of Mr 60,000 and Mr 14,500. To confirm that Peak IIa-2 and Peak IIb-1 were calcineurin, immunoblot analysis using mouse monoclonal anti-calcineurin a-subunit IgG was performed and both peaks showed strong immune-reactivity (data After immuno-precipitation using the same antibody, not shown). calcium binding was reduced to lo-20% in both peaks, whereas no change was detected by that using pre-immune IgG. This confirmed that the calcium binding activity in these peaks was due solely to calcineurin. Therefore, we have concluded that DEAE cellulose Peak II (Figure 2) is composed of two major calcium binding proteins, CAB-48 and calcineurin. Peak III was directly chromatographed by HPLC gel permeation column, and two calcium binding activity peaks were resolved, migrating with apparent molecular weight of 78,000 and 30,000 (Figure 5). The calcium binding activity of the higher molecular weight peak (Peak IIIa) was destroyed by heat treatment but the activity of the lower molecular weight peak (Peak IIIb) was stable to heat-treatment (data not shown). Peak IIIb which coincided with the calmodulin activity peak was subjected to purification on phenylsepharose (13). The calcium binding activity, after phenylsepharose chromatography was found to be conserved in a single peak of calcium binding activity of Mr 18,000 by SDS-PAGE (Figure 6). Therefore, on the basis of molecular weight and phosphodiesterase activation, we concluded that the calcium binding activity of peak IIIb was due to calmodulin. Analysis of peak IIIa by SDS PAGE resolved 8-10 protein bands, the major band was of Mr 63,000. This suggested the presence of calregulin (CAB-63) (11) in this calcium binding activity peak. Radioimmunoassay established that the calcium binding activity peak (IIIa, Figure 5) and
236
That the calcium the calregulin immunoreactivity peak coeluted. binding activity of peak IIIa was entirely due to calregulin was Therefore, DEAE confirmed by immunoprecipitation (data not shown). cellulose peak III consists of two major calcium binding proteins, calregulin and calmodulin. The SDS PAGE analysis of the five calcium binding proteins purified from the calcium binding activity peaks is presented in Figure 6. Among these five brain calcium binding proteins, two have known functions. Calcineurin is a calcium-calmodulin dependent phosphatase (20) and calmodulin is a calcium-dependent regulatory protein (21). The physiological functions of CAB-27, CAB-48 and calregulin are unknown. The calcium binding properties of four of the bovine brain calcium binding proteins has been examined by equilibrium dialysis (Figure 7). In the presence of 3.0 mM MgC12 and 150 mM KC1 all four bovine brain calcium binding proteins bind calcium with pKd(Ca2+) 7-5. Calcineurin has been reported to bind calcium under similar conditions with a Kd l5 PM (18). Since free cytosolic calcium may range from 0.05 PM in unstimulated cells to as high as 10 PM in stimulated cells (22), it is reasonable to postulate that all five bovine brain calcium binding proteins are physiological calcium receptors. Our approach to the identification of the major calcium binding proteins of bovine brain has involved the chromatography of the brain 100,000 x g supernatant on DEAE cellulose and analysis of the resultant fractions for calcium binding proteins by two distinct procedures. In a previous communication (10) the 45Ca2+ autoradiographic procedure identified CAB-27, the p subunit of calcineurin and calmodulin as the principal calcium binding proteins of the DEAE cellulose fractions. In the present communication we have used the Chelex-100 competitive calcium binding assay to analyze the DEAE cellulose fractions and in addition to CAB-27, calcineurin, and calmodulin we have identified calregulin (CAB-63) and CAB-48 as major calcium binding proteins of bovine brain. Comparison of the 45Ca2+ autoradiographic and Chelex-100 procedures suggests that the 45Ca2+ autoradiographic procedure detects only heat stable calcium binding proteins, i.e. calcium binding proteins that do not denature after incubation at 100°C in SDS PAGE disruption buffer. Calregulin and CAB-48 are denatured by heattreatment and consequently are not detected by the 45Ca2+ autoradiographic procedure. The failure of the Chelex-100 assay to detect other known brain calcium binding proteins such as parvalbumin (S), lOK-CaBP, and 28 K-CaBP (23) may be due to the low concentrations of these proteins in brain tissues. In contrast the failure of the Chelex-100 assay to detect the S-100 calcium binding protein (7) is probably due to the low affinity of calcium binding exhibited by this protein rather than its concentration in brain tissue. Nevertheless, the Chelex-100 competitive calcium binding assay appears to be an extremely useful assay for screening for tissue calcium binding proteins.
237
ACKNOWLEDGEMENTS This work is supported by a grant from the Alberta Heritage Foundation for Medical Research (MT and NCK) and the Medical Research Council of Canada (DMW). We are grateful to Dr. J.H. Wang (University of Calgary) for the generous gifts of calmodulin, and anti-calcineurin a subunit antibody. REFERENCES 1.
2. 3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Katz, B. (1969) The Release of Neural Transmitter Substances. Thomas Springfield, IL. Reichardt, L., and Kelly, R. (1983) A Molecular Description of Nerve Terminal Function. Ann. Rev. Biochem. 52, 871-912. Lansdowne, D., and Ritchie, J.M. (1971) Optical Studies on the Kinetics of the Sodium Pump in Mammalian Non-myelinated Nerve Fibres. J. Physiol. 212, 483-502. DeLorenzo, R.J. (1980) Role of Calmodulin in Neurotransmitter Release and Synaptic Function. Ann. N.Y. Acad. Sci. 356, 92109. Baron, G., Demaille, J. and Dutruge, E. (1975) The Distribution of Parvalbumins in Muscle and in Other Tissues. Febs Lettr. 56, 156-160. Wallace, R.W., Tallant, E.A., and Cheung, W.Y. (1980) High Levels of Heat-Labile Calmodulin-binding Protein (CaM-BP80) in Bovine Neostriatum. Biochem. 19, 1831-1837. Hidaka, H., Endo, T., and Kato, K. (1983) Assay of S-100 Protein by an Enzyme Immunoassay Method. Methods in Enzymol. 102, 256-261. Rehm, H., Wiedenmann, B., and Betz, H. (1986) Molecular Characterization of Synaptophysin, a Major Calcium-binding Protein of the Synaptic Vesicle Membrane. Embo. J. 5, 535-541. Waisman, D.M., and Rasmussen, H. (1982) A Reexamination of the Chelex Competitive Calcium Binding Assay. Cell Calcium 4, 89105. Waisman, D.M., Tokuda, M., Morys, S.J.M., Buckland, L.T., and Clark, T. (1985a). Identification of Bovine Brain Ca2+-Binding Proteins. Biochem. Biophys. Res. Commun. 128, 1138-1144. Waisman, D.M., Salimath, B.P., and Anderson, M.J. (1985b) Isolation and Characterization of CAB-63, a Novel Calciumbinding Protein. J. Biol. Chem. 260, 1652-1660. Teo, T.S., Wang, T.H., and Wang, J.H. (1973) Purification and Properties of the Protein Activator of Bovine Heart Cyclic Adenosine 3,5'-Monophosphate Phosphodiesterase. J. Biol. Chem. 248, 588-595. Ca2+-induced Gopalakrishna, R. and Anderson, W.B. (1982) Hydrophobic Site on Calmodulin: Application for the Purification of Calmodulin by Phenyl-Sepharose Chromatography. Biochem. Biophys. Res. Commun. 104, 830-836. Laemmli, U.K. (1970) Cleavage of Structural Proteins during the Assembly of the Head of Bacteriophage T4. Nature (London) 227, 680-685. A Rapid and Sensitive Method for tFBradford, M.M. (1976) Quantitation of Microgram Quantities of Protein Utilizing the
238
16
17
18
19
20
21 22
23
Principle of Protein-Dye. Binding. Anal. Biochem. 72, 248-254. Calmodulin-Stimulated Pallen, C.J., and Wang, J.H. (1983) Dephosphorylation of p-nitrophenyl Phosphate and Free Phosphotyrosine by Calcineurin. J. Biol. Chem. 258, 8550-8556. Khanna, N.C., Tokuda, M., and Waisman, D.M. (1986) Conformational Changes Induced by Binding of Divalent Cations to Calregulin. J. Biol. Chem. 261, 8883-8887. Crouch, T.H. and Klee, C.B. (1980) Positive Cooperative Binding of Calcium to Bovine Brain Calmodulin. Biochemistry, 19, 36923698. Sharma, R.K., Taylor, W.A., and Wang, J.H. (1983) Use of Calmodulin Affinity Chromatography for Purification of Specific Calmodulin-dependent Enzymes. Met. In Enzymol. 102, 210-219. Stewart, A.A., Ingebritsen, T.S., Manalan, A., Klee, C.B., and Cohen, P. (1982) Disorder of a Ca2+ and Calmodulin-dependent Protein Phosphatase. FEBS Lett. 137, 80-85. Klee, C.B., Crouch, T.H., and Richman, P.G. (1980) Calmodulin. Ann. Rev. Biochem. 49, 489-515. Snowdowne, K.W., and Borle, A.B. (1984). Measurement of Cytosolic Free Calcium in Mammalian Cells with Aequorin. Am. J. Physiol. 247, C396-C408. Thomasset, M., Parkes, C.O. and Cuisinier, P. (1982) Rat Calcium Binding Proteins: Distribution, Development and Vitamin D Dependence. Am. J. Physiol. 243, E483-E488.
received 1.3.87 revised version received 1.4.87
accepted 14.4.5’
239
t
(1987) 8. 229-239 Group UK Ltd 1987
IDENTIFICATION OF BOVINE BRAIN CALCIUM BINDING PROTEINS Masaaki Tokuda, Navin C. Rhanna and David M. Waisman Cell Regulation Group, Department of Medical Biochemistry, The University of Calgary, Calgary, Alberta, Canada T2N 4Nl
(reprint requests to DMW) ABSTRACT Three peaks of calcium binding activity have been identified by the Chelex-106 calcium binding assay of the fractions from DEAE cellulose These chromatography of 100,000 x g supernatant of bovine brain. calcium binding activity peaks have been subjected to extensive purification and three novel calcium binding proteins (Mr 27,000, Mr 48,000 and Mr 63,000) and two previously characterized proteins (calcineurin and calmodulin) have been identified as components of calcium binding activity peaks. Analysis of the calcium binding properties of the novel proteins by equilibrium dialysis suggests these proteins may be intracellular calcium receptors. INTRODUCTION Calcium is believed.to play a crucial role in the process of synaptic transmission. When an action potential arrives at a presynaptic nerve terminal the resultant membrane depolarization opens calcium channels in the presynaptic membrane. The resultant influx of calcium leads to a transient rise in the intracellular calcium concentration;this is believed to be the intracellular signal for neurotransmitter release Electrophysiological and morphological studies have indicated (1). that the calcium induced fusion of synaptic vesicles with the presynaptic membrane is a key event in the release process (2). Changes in cytoplasmic calcium have multiple consequences on synaptic function. Nerve terminal glycogenolysis and respiration are increased in response to the calcium signal (3). Cytoplasmic calcium also stimulates the synthesis of several neurotransmitters by increasing the activity of rate limiting transporters of enzymes. For example, choline transport in cholinergic neurons, and tyrosine hydroxylase in catecholamine neurons are activated by increases in cytoplasmic calcium. Cytoplasmic calcium also reduces the activity of certain calcium channels and activates calcium dependent potassium channels resulting in changes in synaptic efficiency (reviewed in 2).
229
Fundamental to an understanding of the molecular events involved in the calcium dependent regulation of synaptic function is the identification and characterization of the various intracellular calcium binding proteins of the brain since these proteins are ultimately responsible for mediating the actions of cytoplasmic calcium. Calmodulin has been shown to be present in high concentrations in nerve terminals (4) and has been suggested to be a major, but not exclusive calcium binding protein and mediator of calcium transduction in the synapse. Other calcium binding proteins such as parvalbumin (5), calcineurin (6), S100 (7), synaptophysin (8) and synexin (2) have been identified in the brain tissue and elucidation of the function of these proteins in brain tissue is of obvious importance. Considering the fundamental importance of calcium binding proteins to the calcium transduction process, identification of the complete spectrum of intracellular brain calcium binding proteins is essential. Our approach to the identification of the intracellular calcium binding proteins has involved chromatography of the 100,000 x g supernatants of a variety of tissues on DEAE cellulose and analysis of the resultant fractions for calcium binding activity by the Chelex 100 competitive calcium binding assay (9) or by 45Ca2+ au&radiography (10). Results from these studies have suggested the existence in the 100,000 x g supernatants of bovine liver, heart, and brain of several novel calcium binding proteins (10, 11). In the present communication we report on the complete analysis of the 100,000 x g supernatant of bovine brain by the Chelex 100 assay. The three major peaks of calcium binding activity resolved by this procedure have been purified and the calcium binding proteins responsible for these calcium binding activity peaks identified. MATERIALS AND METHODS Chelex-100 and hydroxylapatite were obtained from Bio-Rad. Diisodithiothreitol propylfluorophosphate benzamidine, (DTT), (DFP), phenylmethylsulfonylfluoride (PMSF) and anti-rabbit IgG antibodies peroxidase conjugated were obtained from Sigma. DEAE-cellulose (DE-52) was obtained from Whatman. 45CaC12 was purchased from Amersham. DEAE Cellulose Chromatoeranhv of Bovine Brain 100.000 x e sunernatant The chromatographic procedure for the purification of bovine brain calcium binding proteins is summarized in Fig 1. One kg of frozen bovine brain was chopped and then further minced in a meat grinder. The minced tissue was mixed with 3.0 liters of ice cold buffer containing 40 mM Tris-HCl (pH 8.0), 0.5 mM PMSF, 100 mg/l soybean trypsin inhibitor, 1.0 mM DFP, 50 ml of packed Chelex-100, 5 mg/l leupeptin, 5.0 mM benzamidine and 1.0 mM DTT, and then homogenized in a Waring blender. The resultant extract was centrifuged at 20,000 x g for 30 min and the supernatant was then centrifuged at 100,000 x g for 60 min. The 100,000 x g supernatant was diluted with five volumes of buffer A (40 mM Tris-HCl (pH 8.0), 0.5 mM PMSF, 5.0 mM benzamidine, 1.0 mM DTT) and 600 ml of packed DEAE-cellulose was then added. The mixture was
230
stirred rapidly for 90 min and then filtered through a coarse sintered The resultant slurry was washed with 6.0 liters of glass funnel. buffer A and poured into a 5.0 x 60 cm column. Proteins were eluted with 4.5 liters of a linear NaCl gradient (O-O.45 M) in buffer A. The resulting fractions were analyzed for both calcium binding activity using the Chelex-100 competitive calcium binding assay (9) and calmodulin activity assayed by phosphodiesterase activity (12). As shown in Figure 2, three peaks of calcium binding activity (Peak I-III) were resolved. Each peak was pooled separately, and subjected to further purification.
100,000 x g sup. I
DEAE /
Peak I
Ccl lulose I
Peak11
\
Peak III
Calregulin Figure 1. The scheme of purification and identification of calcium binding proteins from bovine brain.
FRACTION
NUMBER
Figure 2. DEAE cellulose chromatography of 100,000 x g supernatant of bovine brain. Chromatography was performed as described under "Materials and Methods". 231
Purification of CAB-27 from Peak I Peak I was concentrated by ultrafiltration (PM-lo, Amicon) and then dialyzed against buffer B (10 mM sodium-phosphate buffer (pH 6.8), 0.5 mM PMSF, 5.0 mM benzamidine, 1.0 mM DTT) at 4°C for overnight. The dialyzed material was applied to a 5.0 x 20 cm column of hydroxylapatite previously equilibrated with buffer B. The column was washed with two column volumes of buffer B and eluted with 2.5 liters of linear phosphate gradient (lo-200 mM) in buffer B. Fractions were collected and analyzed for calcium binding activity (Figure 3a). The calcium binding activity peak was pooled and heated at 80°C for 2
% ‘.“T
9
x-
4-L. 20
a
B z' 2 d 5 0.28 5
20
40 FRACTION
60 NUMBER
80
0.1% 2 E! OaI 100 FRACTION
NUMBER
Figure 3. Identification of CAB-27 from Peak-I. Peak-I was chromatographed on a hydroxylapatite column as described under "Materials and Methods" (a). Calcium binding activity peak was pooled and heat treated at 80°C for 2 min. Denatured protein was removed by centrifugation at 30,000 x g for 30 min, and the supernatant was applied to HPLC gel permeation chromatography (b). min and the denatured protein was removed by centrifugation at 100,000 x g for 60 min. The supernatant was concentrated to 2.0 ml by ultrafiltration (PM-lo), and applied to a high performance gel permeation liquid chromatography column (TSK-G3000 SW, LKB) and eluted with buffer C (40 mM MOPS, (pH 7.1), 150 mM NaCl, 1.0 mM DTT) at a flow rate of 30 ml/hr. Fractions were collected and assayed for calcium binding The peak of calcium binding activity was activity (Figure 3b). analyzed by SDS-PAGE and the Mr 27,000 protein which corresponded with the calcium binding activity peak was pooled. Purification and identification of CAB-48 and calcineurin from Peak II Peak II (from DEAE cellulose, Figure 2) was chromatographed on a hydroxylapatite column using the identical procedure as described for Peak I except the phosphate gradient was lo-350 mM in buffer B. Fractions were assayed for calcium binding activity, and two major activity peaks (Figure 4a, Peak IIa and IIb) were resolved. Each peak
232
was pooled,
concentrated
(PM-lo) and then applied
to TSK-G3000
% f; 0.4W ‘, E
&
O.l$
2
8 z or
FRACTION
SW
oo” . . .._I*-.__. 20 _
40 FRACTION
60 NUMBER
NUMBER
80
--
100
FRACTION
NUMBER
Figure 4. Purification of CAB-48 and identification of calcineurin from Peak-II. Peak-II was chromatographed on a hydroxylapatite column as described under "Materials and Methods" (a). HPLC gel permeation chromatography was performed for two calcium binding activity peaks, Peak-IIa (b) and Peak-IIb (c).
column (IIa-Figure 4b, IIb-Figure 4~). The peak of calcium binding activity resulting from gel permeation chromatography of Peak 16 (Figure 5) was further purified by chromatography on a calmodulin sepharose affinity column according to Sharma et al. (17). The peak of calcium binding activity resulting from gel permeation chromatography of Peak IIa (Figure 4b) was analyzed by SDS PAGE and a homogeneous
233
protein band of Mr 48,000 was found to correlate with calcium binding
0
20
40 FRACTION
60
60
NUMBER
Identification of CAB-63 and calmodulin from Peak-III. Figure 5. Peak-III was chromatographed on a HPLC gel permeation column as described under "Materials and Methods". Immunoblot analysis using a rabbit anti-CAB-63 antibody were performed (inset). Lane 1: 5 pg of purified bovine liver CAB-63, lane 2: 20 pg protein of fraction 40.
Identification of calregulin and calmodulin from Peak III Peak III was concentrated by ultrafiltration (PM-lo) and applied to TSK-G3000 SW column. Proteins were eluted with buffer C at a flow rate of 30 ml/hr. Fractions were collected and analyzed for both calregulin (CAB-63) concentration by an ELISA using a rabbit polyclonal antibody against calregulin (11) and calmodulin activity (12) (Figure 5). Both peaks were pooled separately and subjected to further purification according to published procedures (peak IIIa (ll), peak IIIb (13)). Other Methods Immunoblot analysis (ll), polyacrylamide gel electrophoresis (14), protein determination (15), calcineurin activity (16), Chelex-100 competitive calcium binding assay (9) and equilibrium dialysis (17) were performed as described. RESULTS AND DISCUSSION Bovine brain 100,000 x g supernatant was chromatographed on DEAEcellulose column and fractions were assayed for calcium binding As shown in Figure 2, three peaks of calcium binding activity. activity were resolved (Peak I-III) and only Peak III coincided with
234
calmodulin activity. In order to identify the calcium binding protein responsible for calcium binding activity peaks an extensive purification procedure was developed. The procedure for the purification of the calcium binding proteins responsible for the calcium binding activity peaks is presented in Figure 1. Peak I was chromatographed on a hydroxylapatite column, and only one The peak calcium binding activity peak was resolved (Figure 3a). fractions were pooled and assayed for heat stability of calcium binding activity. About 85% of the activity was retained after heating at 80°C for 2 min. Therefore pooled peak I was heat treated, denatured protein removed by centrifugation, and the supernatant was concentrated and chromatographed by HPLC gel permeation chromatography (Figure 3b). A single peak of calcium binding activity migrating with an apparent Mr 40,000 was resolved. Analysis of this peak by SDS-PAGE demonstrated a single protein band of Mr 27,000 (Figure 6).
b
t LOG [Co2+]
Figure 6. SDS-polyacrylamide gel electrophoresis of major calcium binding proteins. Five. major calcium binding proteins from bovine brain are analyzed on 12.5% polyacrylamide gel electrophoresis contained 0.1% SDS. a: molecular weight standards, b: CAB-27 (3 pg), c: CAB-48 (3 pg), d: calcinuerin (6 pg), e: CAB-63 (3 pg), f: calmodulin (3 /Jg) Figure 7. Calcium binding saturation curves 'of major calcium binding proteins. Calcium binding assay was performed by equilibrium dialysis in the presence of 3 mM MgC12 and 150 mM KC1 in the MOPS buffer (pH 7.1), for purified CAB-27 (0), CAB-48 (O), and CAB-63 (m). Data for calmodulin (0) was used from ref. 18. This protein was identical to the Mr 27,000 calcium binding protein identified by 45Ca2+ -autoradiography of bovine brain 100,000 x g supernatants (10). These results suggest that the calcium binding
235
activity of peak I is due to CAB-27. Hydroxylapatite chromatography of Peak II resolved two calcium binding activity peaks between 60 and 100 mM phosphate (Peak IIa) and between 120 and 160 mM phosphate (Peak IIb) (Figure 4a). Peak IIa was chromatographed on HPLC gel permeation column and two calcium binding activity peaks migrating with apparent Mr 75,000 at 72,000 (Figure 4b) were detected. The higher molecular weight peak (Peak IIa-1) was identified by SDS-PAGE as a single homogeneous protein band of Mr 48,000 (Figure 6) which we named CAB-48. The lower molecular weight calcium binding activity peak (Peak IIa-2) was not homogeneous by SDS PAGE. Since the SDS-PAGE analysis of the peak fractions revealed prominent bands at Mr 60,000 and Mr 14,500 we suspected that this calcium binding activity peak contained calcineurin. Interestingly, calcineurin activity was found to correlate with calcium binding activity. Chromatography of this calcium binding activity peak on calmodulin-sepharose (19) conserved the calcium binding activity as a single protein peak (data not shown) and SDS PAGE resolved two protein bands of Mr 60,000 and Mr 14,500. HPLC gel permeation chromatography of Peak IIb resolved a single calcium binding activity peak (Peak IIb-1) migrating with an apparent Mr 72,000 which also coincided with calcineurin activity (Figure 4~). Subsequent chromatography of this peak on calmodulin-sepharose (19) resolved a single peak of calcium binding activity (data not shown) and SDS-PAGE analysis revealed two protein bands of Mr 60,000 and Mr 14,500. To confirm that Peak IIa-2 and Peak IIb-1 were calcineurin, immunoblot analysis using mouse monoclonal anti-calcineurin a-subunit IgG was performed and both peaks showed strong immune-reactivity (data After immuno-precipitation using the same antibody, not shown). calcium binding was reduced to lo-20% in both peaks, whereas no change was detected by that using pre-immune IgG. This confirmed that the calcium binding activity in these peaks was due solely to calcineurin. Therefore, we have concluded that DEAE cellulose Peak II (Figure 2) is composed of two major calcium binding proteins, CAB-48 and calcineurin. Peak III was directly chromatographed by HPLC gel permeation column, and two calcium binding activity peaks were resolved, migrating with apparent molecular weight of 78,000 and 30,000 (Figure 5). The calcium binding activity of the higher molecular weight peak (Peak IIIa) was destroyed by heat treatment but the activity of the lower molecular weight peak (Peak IIIb) was stable to heat-treatment (data not shown). Peak IIIb which coincided with the calmodulin activity peak was subjected to purification on phenylsepharose (13). The calcium binding activity, after phenylsepharose chromatography was found to be conserved in a single peak of calcium binding activity of Mr 18,000 by SDS-PAGE (Figure 6). Therefore, on the basis of molecular weight and phosphodiesterase activation, we concluded that the calcium binding activity of peak IIIb was due to calmodulin. Analysis of peak IIIa by SDS PAGE resolved 8-10 protein bands, the major band was of Mr 63,000. This suggested the presence of calregulin (CAB-63) (11) in this calcium binding activity peak. Radioimmunoassay established that the calcium binding activity peak (IIIa, Figure 5) and
236
That the calcium the calregulin immunoreactivity peak coeluted. binding activity of peak IIIa was entirely due to calregulin was Therefore, DEAE confirmed by immunoprecipitation (data not shown). cellulose peak III consists of two major calcium binding proteins, calregulin and calmodulin. The SDS PAGE analysis of the five calcium binding proteins purified from the calcium binding activity peaks is presented in Figure 6. Among these five brain calcium binding proteins, two have known functions. Calcineurin is a calcium-calmodulin dependent phosphatase (20) and calmodulin is a calcium-dependent regulatory protein (21). The physiological functions of CAB-27, CAB-48 and calregulin are unknown. The calcium binding properties of four of the bovine brain calcium binding proteins has been examined by equilibrium dialysis (Figure 7). In the presence of 3.0 mM MgC12 and 150 mM KC1 all four bovine brain calcium binding proteins bind calcium with pKd(Ca2+) 7-5. Calcineurin has been reported to bind calcium under similar conditions with a Kd l5 PM (18). Since free cytosolic calcium may range from 0.05 PM in unstimulated cells to as high as 10 PM in stimulated cells (22), it is reasonable to postulate that all five bovine brain calcium binding proteins are physiological calcium receptors. Our approach to the identification of the major calcium binding proteins of bovine brain has involved the chromatography of the brain 100,000 x g supernatant on DEAE cellulose and analysis of the resultant fractions for calcium binding proteins by two distinct procedures. In a previous communication (10) the 45Ca2+ autoradiographic procedure identified CAB-27, the p subunit of calcineurin and calmodulin as the principal calcium binding proteins of the DEAE cellulose fractions. In the present communication we have used the Chelex-100 competitive calcium binding assay to analyze the DEAE cellulose fractions and in addition to CAB-27, calcineurin, and calmodulin we have identified calregulin (CAB-63) and CAB-48 as major calcium binding proteins of bovine brain. Comparison of the 45Ca2+ autoradiographic and Chelex-100 procedures suggests that the 45Ca2+ autoradiographic procedure detects only heat stable calcium binding proteins, i.e. calcium binding proteins that do not denature after incubation at 100°C in SDS PAGE disruption buffer. Calregulin and CAB-48 are denatured by heattreatment and consequently are not detected by the 45Ca2+ autoradiographic procedure. The failure of the Chelex-100 assay to detect other known brain calcium binding proteins such as parvalbumin (S), lOK-CaBP, and 28 K-CaBP (23) may be due to the low concentrations of these proteins in brain tissues. In contrast the failure of the Chelex-100 assay to detect the S-100 calcium binding protein (7) is probably due to the low affinity of calcium binding exhibited by this protein rather than its concentration in brain tissue. Nevertheless, the Chelex-100 competitive calcium binding assay appears to be an extremely useful assay for screening for tissue calcium binding proteins.
237
ACKNOWLEDGEMENTS This work is supported by a grant from the Alberta Heritage Foundation for Medical Research (MT and NCK) and the Medical Research Council of Canada (DMW). We are grateful to Dr. J.H. Wang (University of Calgary) for the generous gifts of calmodulin, and anti-calcineurin a subunit antibody. REFERENCES 1.
2. 3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Katz, B. (1969) The Release of Neural Transmitter Substances. Thomas Springfield, IL. Reichardt, L., and Kelly, R. (1983) A Molecular Description of Nerve Terminal Function. Ann. Rev. Biochem. 52, 871-912. Lansdowne, D., and Ritchie, J.M. (1971) Optical Studies on the Kinetics of the Sodium Pump in Mammalian Non-myelinated Nerve Fibres. J. Physiol. 212, 483-502. DeLorenzo, R.J. (1980) Role of Calmodulin in Neurotransmitter Release and Synaptic Function. Ann. N.Y. Acad. Sci. 356, 92109. Baron, G., Demaille, J. and Dutruge, E. (1975) The Distribution of Parvalbumins in Muscle and in Other Tissues. Febs Lettr. 56, 156-160. Wallace, R.W., Tallant, E.A., and Cheung, W.Y. (1980) High Levels of Heat-Labile Calmodulin-binding Protein (CaM-BP80) in Bovine Neostriatum. Biochem. 19, 1831-1837. Hidaka, H., Endo, T., and Kato, K. (1983) Assay of S-100 Protein by an Enzyme Immunoassay Method. Methods in Enzymol. 102, 256-261. Rehm, H., Wiedenmann, B., and Betz, H. (1986) Molecular Characterization of Synaptophysin, a Major Calcium-binding Protein of the Synaptic Vesicle Membrane. Embo. J. 5, 535-541. Waisman, D.M., and Rasmussen, H. (1982) A Reexamination of the Chelex Competitive Calcium Binding Assay. Cell Calcium 4, 89105. Waisman, D.M., Tokuda, M., Morys, S.J.M., Buckland, L.T., and Clark, T. (1985a). Identification of Bovine Brain Ca2+-Binding Proteins. Biochem. Biophys. Res. Commun. 128, 1138-1144. Waisman, D.M., Salimath, B.P., and Anderson, M.J. (1985b) Isolation and Characterization of CAB-63, a Novel Calciumbinding Protein. J. Biol. Chem. 260, 1652-1660. Teo, T.S., Wang, T.H., and Wang, J.H. (1973) Purification and Properties of the Protein Activator of Bovine Heart Cyclic Adenosine 3,5'-Monophosphate Phosphodiesterase. J. Biol. Chem. 248, 588-595. Ca2+-induced Gopalakrishna, R. and Anderson, W.B. (1982) Hydrophobic Site on Calmodulin: Application for the Purification of Calmodulin by Phenyl-Sepharose Chromatography. Biochem. Biophys. Res. Commun. 104, 830-836. Laemmli, U.K. (1970) Cleavage of Structural Proteins during the Assembly of the Head of Bacteriophage T4. Nature (London) 227, 680-685. A Rapid and Sensitive Method for tFBradford, M.M. (1976) Quantitation of Microgram Quantities of Protein Utilizing the
238
16
17
18
19
20
21 22
23
Principle of Protein-Dye. Binding. Anal. Biochem. 72, 248-254. Calmodulin-Stimulated Pallen, C.J., and Wang, J.H. (1983) Dephosphorylation of p-nitrophenyl Phosphate and Free Phosphotyrosine by Calcineurin. J. Biol. Chem. 258, 8550-8556. Khanna, N.C., Tokuda, M., and Waisman, D.M. (1986) Conformational Changes Induced by Binding of Divalent Cations to Calregulin. J. Biol. Chem. 261, 8883-8887. Crouch, T.H. and Klee, C.B. (1980) Positive Cooperative Binding of Calcium to Bovine Brain Calmodulin. Biochemistry, 19, 36923698. Sharma, R.K., Taylor, W.A., and Wang, J.H. (1983) Use of Calmodulin Affinity Chromatography for Purification of Specific Calmodulin-dependent Enzymes. Met. In Enzymol. 102, 210-219. Stewart, A.A., Ingebritsen, T.S., Manalan, A., Klee, C.B., and Cohen, P. (1982) Disorder of a Ca2+ and Calmodulin-dependent Protein Phosphatase. FEBS Lett. 137, 80-85. Klee, C.B., Crouch, T.H., and Richman, P.G. (1980) Calmodulin. Ann. Rev. Biochem. 49, 489-515. Snowdowne, K.W., and Borle, A.B. (1984). Measurement of Cytosolic Free Calcium in Mammalian Cells with Aequorin. Am. J. Physiol. 247, C396-C408. Thomasset, M., Parkes, C.O. and Cuisinier, P. (1982) Rat Calcium Binding Proteins: Distribution, Development and Vitamin D Dependence. Am. J. Physiol. 243, E483-E488.
received 1.3.87 revised version received 1.4.87
accepted 14.4.5’
239