THE JOURNAL OF BIOLOGICAL CHEMISTRY Q 1993 by The American Society for Biachemistry and Molecular Biology, Inc

Vol. 268, No. 10, Iasue of April 5, pp. 7422-7426,1993 Printed in U.S.A.

In Situ Capture of p-Calpain Activation in Platelets* (Received for publication, October 5, 1992)

Takaomi C. SaidoSfj, Hidenori Suzukiq, Hiroh Yamazakiq, Kenjiro Tanouev, and Koichi SuzukiS From the $Department of Molecular Biology and (Department of Cardiovascular Research, Tokyo Metropolitan Institute of Medical Science, Honkomagome, Bunkyo-ku, Tokyo 113, Japan

In situ detection of calpain activation in intact cells has not been possible to date. Here we present the first direct evidence, employing a novel approach, that p- calpain is rapidly activated at cell membranes in plate- lets upon a rise in intracellular calcium concentration. Immunoelectron microscopy using antibodies capable of distinguishing between the pre- and postautolysis forms of r-calpain revealed that treatment of platelets with calcium ionophore causes the preautolysis form of the protease to translocate from cytosol to mem- branes, where it becomes activated by autolysis. This indicates that proteins associated with membranes serve as primary substrates for calpain in cells.

Calpain (Suzuki et al., 1987; Go11 et al., 1990), calcium- activated neutral protease ubiquitously present in animal cells, stands as a unique receptor for intracellular calcium signals since activation of calpain leads to irreversible prote- olytic processing of substrate proteins. Calpain-catalyzed pro- teolysis proceeds in a limited manner, resulting in alteration of the biochemical and structural parameters of the substrate rather than simple destruction. A typical example is the case with protein kinase C isozymes, which are converted by cal- pain to active forms that do not require cofactors such as phospholipids and phorbol ester (Kishimoto et al., 1983; Saido et al., 1992a). The physiological consequences of calpain ac- tivation, therefore, are defined by the ways its particular substrate proteins are modified by proteolysis. This suggests that to identify the physiological sites of calpain action in cells is of primary importance in approaching its functions.

The physiological behavior of calpain in cells, however, has remained unclear due to difficulties in detecting its intracel- lular action despite the accumulated knowledge concerning its enzymatic properties and structures (Go11 et al., 1990; Suzuki and Ohno, 1990). One approach to the problem is to resolve the transitional activation of the enzyme upon stim- ulation of cells; the activation process of calpain accompanies subunit autolysis and thus can be monitored by probing autolysis (Saido et al., 1992b). Nevertheless, the autolytic transition could only be observed as a change in electropho- retic mobility in sodium dodecyl sulfate-polyacrylamide gel

*This work was supported in part by research grants from the Ministry of Education, Science and Culture of Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduer- tisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

3 To whom correspondence should be addressed Dept. of Molecu- lar Biology, Tokyo Metropolitan Institute of Medical Science, 3-18- 22 Honkomagome, Bunkyo-ku, Tokyo 113, Japan. Tel.: 81-3-3823- 2101 (ext. 5321); Fax: 81-3-5685-6609.

electrophoresis (SDS-PAGE),’ a method which disposes of spatial information as to where in cells the process takes place; the direct observation of calpain activation in cells has not been possible to date. To overcome this technical drawback, we have recently developed a set of antibodies capable of distinguishing between the pre- and postautolysis forms of p- calpain and showed that autolysis precedes substrate prote- olysis both in test tubes and in cells (Saido et al., 1992b). Our efforts now have been directed to acquisition of this spatial information as to where in cells calpain undergoes activation and acts on its targets.

In the present study, we have employed human platelets to examine the intracellular behavior of p-calpain. Platelets have presented a model case for calpain research (Colman and Hoffman, 1990) since Fox et al. (1983,1985) reported calpain- catalyzed proteolysis in activated platelets. Calpain has been suggested to mediate thrombin- and collagen-induced platelet activation (Tsujinaka et al., 1988; Toyo-oka et al., 1989), although its role in platelet activation remains controversial (Elce et al., 1989; Anagoli et al., 1991). The enzyme may rather be involved in postaggregation events such as procoagulation (Verhallen et al., 1987). Here, we show by immunoelectron microscopy using our newly developed antibodies that p- calpain is rapidly activated at cell membranes upon a rise in intracellular calcium, giving clues as to the ways the protease interacts with its physiological target proteins.

EXPERIMENTAL PROCEDURES

Reagents”A23187 and essentially fatty acid-free bovine serum albumin were from Sigma. All the other reagents were of analytical grade, purchased from Wako Pure Chemicals, Nacalai Tesque, or Sigma. Pre- and postautolysis forms of purified p-calpain were pro- duced as described (Saido et al., 1992b). Antibodies specific to the pre- and postautolysis forms of p-calpain were developed and affinity purified as described (Saido et al., 1992b).

Platelets-Washed human platelets were prepared from fresh hu- man blood as previously described (Suzuki et al., 1992). Platelets suspended at concentrations of 4-6 X 106/pl in Tyrode’s solution (pH 7.4) consisting of 137 mM NaCl, 2.7 mM KCl, 0.4 mM NaH2P04, 12 mM NaHC03, 1 mM MgC12, 2 mM Ca&, 22 p M sodium citrate, 0.35% bovine serum albumin, and 0.1% glucose were used for the experi- ments within 5 h after blood collection. All the platelet samples used in the study retained the ability of aggregating in response to stimu- lation by A23187 or thrombin under the stirring conditions. Platelet aggregation was observed using an NBS 8-channel aggregometer model PAC-8S (Niko Bioscience Inc., Tokyo).

Immunoelectron Microscopy-Immunoelectron microscopic obser- vation was performed essentially as described (Suzuki et al., 1992). Briefly, washed platelets stimulated with 2 p M A23187 or solvent in the presence of 2 mM calcium under the nonstirring conditions were fixed by adding an equal volume of 0.1 M sodium phosphate (pH 7.4) containing 4% paraformaldehyde and 2 mM EGTA. After 10-20 min of incubation at room temperature, the platelets collected by centrif- ugation were rinsed three times with 0.1 M phosphate buffer and two

The abbreviations used are: PAGE, polyacrylamide gel electro- phoresis; PBS, phosphate-buffered saline.

7422

Intracellular Calpain Activation 7423

times with phosphate-buffered saline (PBS). The fixed platelets were sequentially immersed in 1 M sucrose in PBS for 60 min, 1.84 M sucrose in PBS for 2 h, and then 1.84 M sucrose containing 20% polyvinylpyrrolidone (M, 10,000; Sigma) in PBS for 16 h a t 4 "C. After freezing in liquid nitrogen, ultra-thin frozen sections were prepared by a Reichert Ultracut N ultramicrosome (Wien, Austria) with cryo-attachment FC-4E at -110 "C and mounted on nickel grids. The grids were floated onto a droplet of PBS for rinsing and then transferred to a droplet of 0.1% bovine serum albumin in PBS for blocking. The sections were incubated with either of the anti-p- calpain antibodies (10 pg/ml) in PBS for 16 h at room temperature. After rinsing five times with PBS, the sections were incubated with goat anti-rabbit I g G coupled to colloidal gold (15 nm immuno-gold; 1:30 dilution with PBS; Jassen Life Science Product, Olen, Belgium) for 3 h at room temperature. Nonimmune rabbit IgG and the immuno- gold were used for the control experiments. After rinsing three times with PBS and five times with distilled water, the sections were stained with uranyl acetate followed by washing and the adsorption ion- staining with a mixture of polyvinyl alcohol (M, 10,000; Sigma) and uranyl acetate. The sections were examined with a JEM lOOC electron microscope (JEOL Co. Ltd., Tokyo) at an accelerating voltage of 80 kV.

Other Procedures-SDS-PAGE and Western blot analysis were performed as described (Saido et al., 1992~).

RESULTS AND DISCUSSION

Autolytic Actiuation of p-Calpain-Fig. 1 outlines the strat- egy that we have taken to resolve the activation process of p- calpain in platelets. As schematically shown in Fig. L4, mam- malian p-calpain undergoes autolysis in both the large 80,000 and small 30,000 subunits upon activation and proceeds to substrate proteolysis (Saido et al., 199%). Activation of cal- pain, therefore, can be monitored by probing the conversion of the enzyme from the preautolysis form to the postautolysis form. In the present study, we have employed a set of anti- bodies capable of distinguishing between these two forms of p-calpain as demonstrated in Fig. 1B; the anti-preautolysis

Autolytic Actlvrtlon Ca"

4 new N termnus

form antibody specifically reacts with the preautolysis form of the enzyme while the anti-postautolysis form antibody recognizes only the enzyme's postautolysis form.

In resting human platelets, p-calpain exists exclusively as the preautolysis form with relative molecular weight (M,) of 80,000 as shown by Western blotting (compare the first lanes in Fig. 2, A and B ) . Stimulation of platelets with calcium ionophore (A23187) under stirring conditions caused a rapid conversion of the enzyme to the postautolysis form of 76,000 within 1 min along with the platelet aggregation (Fig. 2, A-C and E ) . Note that these antibodies are specific enough to recognize the corresponding antigens in crude platelet lysates without any nonspecific binding. Autolysis was followed by proteolysis of calpain substrates such as filamin and talin (Fig. 2 0 ) as reported (Fox et al., 1985; Toyo-oka et al., 1989). The activated calpain also seems to be further metabolized along with time (Fig. 2B).

The activation process of p-calpain in platelets under the nonstirring conditions was also examined (Fig. 3). Under these conditions, platelets do not coaggregate on one another and, therefore, are more suitable for the electron microscopic observation. Here, p-calpain underwent autolysis more slowly and about half of p-calpain was activated by 2 p~ A23187 in 5 min. Although "stirring" has been known to augment plate- let reaction in general (Elce et d., 1989). the reason for the slower autolysis of p-calpain under the nonstirring conditions than under the stirring conditions is not yet understood.

Irnrnunoefectron Microscopic Obseruatwn-To elucidate the intracellular sites for calpain activation, resting and activated platelets were examined by immunoelectron microscopy using the antibodies specific to preactivation calpain (Fig. 4) and to postactivation calpain (Fig. 5). The gold particles in the figures indicate the presence of the corresponding antigens.

CBB stain immunostain

anti--autolysis antl-peSt-aulolysls

1 2 1 2 1 2

';?lb5!rE!t9 ~VJ!WJlY2i2

FIG. 1. Autolytic activation of mammalian p-calpain as resolved by preautolyeb- and poetautolysis-specific antibodies. A. scheme for the autolytic activation process. Both of the p-calpain subunits undergo autolysis upon binding of calcium to the calmodulin-like domains and proceeds to substrate proteolysis (Suzuki et of., 1987; Hathaway and McClelland, 1991; Saido et al.. 1992~). The activation process of p-calpain thus can be monitored by probing autolysis. B, activation of p-calpain by calcium causes reduction of the relative molecular weight of the large subunit from 80,000 to 76,000 in SDS-PAGE as visualized by Coomassie Brilliant Blue ( C R R ) staining ([pit p a n e l ) . The Western blot analysis (right p a n e l ) shows that the antibodies designed to distinguish between these pre- and postautoly~ia forms of the large subunits recognize only the corresponding antigens but not the others. These antibodies were uned to examine the activation process of p-calpain in human platelets. I, preautolysis p-calpain; 2, postautolysis p-calpain.

7424 Intracellular Calpain Activation

FIG. 2. Autolytic activation of platelet p-calpain upon a rise in in- tracellular calcium. A-C, Western blot analysis of p-calpain activation in hu- man platelets induced by 2 p~ A23187 under the stirring conditions, using anti- preautolysis calpain antibody (A ), anti- postautolysis antibody (E), and both of the two ( C ) . Platelets in Tyrode's buffer were solubilized at indicated times by addition of solubilization buffer for SDS- PAGE under the reducing conditions and subsequently subjected to SDS- PAGE. The arrows indicate the relative molecular weights. D, Coomassie Bril- liant Blue staining patterns of the rest- ing and activated platelets. The fifth lane represents the background stains of the Tyrode's buffer used to suspend platelets (mainly ascribed to bovine serum albu- min, see "Experimental Procedures"). The last lane shows the molecular weight markers: from the top, 200,000 (myosin), 116,000 (@-galactosidase), 67,000 (bovine serum albumin), 42,000 (aldolase), 30,000 (carbonic anhydrase), and 17,000 (myoglobin). The arrows indicate fi- lamin, talin, and myosin, respectively, from the top. E, aggregation pattern of platelets stimulated as in A-D.

(A) t h 0 5'

(B) 0 5'

. .

tlnm 0 1' 2 5' 0 1' 2' 5' 0 1' 2' 5' 0' 1' 2 5' e i Y

80K + . 76K +

BOK+ - 76K +

FIG. 3. Western blot analysis of p-calpain activation under the nonstirring conditions. Platelets under the nonstirring condi- tions were examined as described in the legend to Fig. 2 with anti- preautolysis antibody ( A ) and with anti-postautolysis antibody ( E ) .

Since strong amino-modifying reagents significantly reduce the reactivity of the antigens to the antibodies (Saido et al., 1992b), the platelets were rather loosely fixed only with 2% paraformaldehyde. For this reason, the ultrastructure of plate- lets here is not as well conserved as in the case using the stronger fixer glutaraldehyde (Suzuki et al., 1991). and yet the number of the gold particles is less than expected from the amount of p-calpain in platelets ("rugha and Stracher, 1981; McGowan et al., 1989).

In resting discoid-shaped platelets, p-calpain exists domi- nantly as the preautolysis form diffusely in cytoplasm as

0 Time (mln)

5

shown in Fig. 4A. Little of the postautolysis form can be observed (Fig. 5A) in accordance with the results given by Western blotting (Figs. 2 and 3).

Stimulation of the platelets with calcium ionophore under the nonstirring conditions identical to those for Fig. 3 induced the morphological changes characterized by the spheroid shape with pseudopodia along with excretion of granules as shown in Figs. 4B and 5B. Here, the number of the preauto- lysis forms is diminished (Fig. 4R) in accordance with the Western blot results (Fig. 3). and the remaining antigens appear to have moved to the vicinity of membranes as indi- cated by the arrows. In contrast, the number of the postau- tolysis p-calpain antigens increased drastically upon stimu- lation (Fig. 5B) and most of them are observed associated closely to the cytoplasmic surfaces of either the plasma mem- brane (double arrows) or the membranes covering the open canalicular systems and the granules (single arrows). A few other gold particles appear to reside on the outer surface of the plasma membrane as indicated by the arrowhead.9, imply- ing that activated p-calpain may even be transferred to the outside of platelets. This observation needs to be further examined for confirmation in a more quantitative manner using such methods as flow cytometry that would clearly distinguish the outside surface from the inside of cells. The activation process of calpain at the membranes may partly involve the cytoskeletal structures since some of the calpain antigens reside somewhat apart from the membranes in acti- vated platelets (Figs. 4B and 5B). These profiles of calpain translocation upon activation are very similar to those of protein kinase C in platelets activated by phorbol ester (Hag- iwara et al., 1990).

Intracellular Calpain Activation 7425

FIG. 4. Immunoelectron microscopic observation of preac- tivation p-calpain in resting and activated platelets. Resting platelets ( p a n e l A ) and platelets activated by 1 ski A23187 for 5 min under nonstirring conditions as described in the legend to Fig. 3 (panel E ) were examined by immunoelectron microscopy, using the antibody specific to the preautolysis form of p-calpain. The dense 15- nm gold particles indicate the presence of the corresponding antigens. See the text for definition of the arrows. aC, a-granule: OCS, open canalicular system.

This is the first direct in situ evidence to demonstrate that p-calpain is activated at the cell membranes. Statistical analy- sis of the immunoelectron microscopy data, showing activa- tion-induced loss of the preautolysis form from cytoplasm and the concomitant appearance of the postautolysis form a t membrane (Table I), further substantiates the observation and indicates that the preautolysis form of p-calpain first translocates from the cytosol to membranes upon a rise in cytoplasmic calcium and is subsequently activated via auto- lysis. The activated calpain is likely to stay associated with the membranes as long as the intracellular calcium level remains elevated since in oitro experiments have shown that calpain becomes drastically hydrophobic upon activation by calcium (Go11 et al., 1990). These results agree with the previous observation that acidic phospholipids, particularly polyphosphoinositides, promote calpain activation and action (Hathaway and McClelland, 1991; Saido et al., 1992~). The results also suggest that physiological substrate proteins for calpain are likely to be proteolyzed in the vicinity of or at cell membranes, supporting the previous speculation that the effect of calpain on the platelet functions may be exerted through alteration of the properties of the cytoskeleton-mem-

FIG. 5. Immunoelectron microscopic observation of poatac- tivation p-calpain in platelets. Human platelets were examined as described in the legend to Fig. 4 with the antibody specific to the postautolysis form of p-calpain. A, resting platelet. R, activated platelet. See the text for description of the gin,& arrows. doubk arrows. and arrowheads.

TABLE I Quuntitation ofpre- and postautolysis calpain antigens in resting and

activated platelets Immunoelectron microscopic data such as those in Figs. 4 and 5

were statistically analyzed by counting the number of cold particles representing antigens. Cold particles within 40 nm from membranes were counted as "membrane-associated" while those farther away inside platelets were counted as "c-ytosolic." Each figure is represented by the means f standard deviation. n indicates the number of individual platelets examined in each case.

.~ __ -~

Platelets Antibody Number of guld particles

Cytmolic Membrane-aMaciated

Resting Preautolysis 15.3 f 3.2 1.8 f 1.2 ( n = 16) Postautolysis 1.3 f 0.9 1.0 f 0.6 ( n = 20)

Activated Preautolynis 4.4 f 1.4 5.6 f 1.3 ( n = 16) Postautolysis 2.5 1.2 13.2 3.3 ( n = 17)

~~ "

-~ - - -

brane interaction (Fox et al., 1985; Verhallen et al., 1987). Because activation of calpain in platelet9 seems to be largely aggregation-dependent (Elce et al., 1989; Anagoli et ol., 1991). we speculate that calpain action may influence such postag- gregation events as procoagulation and clot retraction under the physiological conditions as both seem to involve drastic changes in structures associated with membranes and cyto- skeletons (Verhallen et al., 1987; Lanza et af., 1992).

7426 Intracellular Calpain Activation

Acknowledgments-We are grateful to Drs. S. Kawashima and H. Kawasaki of our Department and to Dr. Y. Katagiri, Y. Hayashi, and I. Baba, the Department of Cardiovascular Research of our Institute, for technical advice and discussions. We also thank Dr. N. Forsberg, Department of Animal Science, Oregon State University, for critical reading of the manuscript.

REFERENCES Anagli, J., Hagmann, J. & Shaw, E. (1991) Biochem. J. 274,497-502 Colman, R. W. & Hoffman, I. (1990) in Intracellular Calcium-dependent Pro-

teolvsis (Mellaren, R. L. & Murachi. T.. &) pp. 211-224, CRC Press, Boca . . "

Raton, FL - . Elce, J. S., Sigmund, L. & Fox, M. J. (1989) Biochem. J. 261,1039-1042 Fox, J. E. B., Reynolds, C. C. & Phillips, D. R. (1983) J. Biol. Chem. 258, 9973-99111

Fox, J. E. B., Goll, D. E., Reynolds, C. C. & Phillips, D. R. (1985) J. Biol. Chem.

Goll, E. D., Kleese, W. C., Okitani, A,, Kumamoto, T., Cong, J. & Kapprell, H.- 260,1060-1066

P. (1990) in Intracellular Calcium-dependent Proteolysis (Mellgren, R. L. & Murachi, T., eds) pp. 103-114, CRC Press, Boca Raton, FL

Hagiwara, M., Sumi, M., Usuda, N., Nagata, T. & Hidaka, H. (1990) Arch. Bwchem. Biophys. 280, 201-205

Hathaway, D. R. & McClelland, P. (1990) in Intracellular Calcium-dependent

. . . - . . - -

Proteolysis (Mellgren, R. L. & Murachi, T., eds) pp. 91-102, CRC Press, Boca

Kishimoto, A,, Kajikawa, N., Shiota, M. & Nishizuka, Y. (1983) J. Biol. Chem. Raton FL

Lanza. F.. 258,ll

McGowan. h. 1 Cazknave J.

56-1164 Stjerle, A., Founier, D., Marales, M., Andre, G., Nurden, A. T. &

E, BeckezrE. & Detwiler, $. C. (1989) Biochem. Biophys. Res. P. (1992) J. Clin. Inuest. 89 1995-2004

IX-435 3, K., Konno, Y., Osada, S., Ohno, S. & Suzuki, K. (1992a)

., Takenawa, T., Murofushi, H. & Suzuki, K. (1992~) 111 81-86

J. diol. C&m. 267: 24585-24590

FEBk dtt. 220: 231-277

(1995) Astoehem. ~.'9+' 7 ~ 1 1 7 ~ ;

Saido ?. C. Shibata M

Suzukj, K. & Ohno, S. (1990) Cell Stru t . Funct. 16, l+ Suzukl K. Ima~oh S. Emon, Y., Kawasakl, H., Mlnaml, Y. & Ohno, S. (1987)

Suzuki, H., Tanoue, K. & Yamazaki, H. (1991) Histochem. 96,31-39 Suzuki H Nakamura S Itoh Y Tanaka, T., Yamazaki, H. & Tanoue, K.

TO o oka T. Shin W.S. +. i1985) &ire. Re Truelia. J. A. & St

. , "Yl-YyU

s. 64,407-410 , Okai, Y., Dan, Y., Morita, M., Iizuka, M. & Sugimoto,

racher. A. (1981) Biochem. Bioohvs. Res. Commun. 100, 8C4-4-822

. . . "

Tsu maka T. Kajikawa Y Kambayashi J. Sakon, M., H r r h i , N. Tanaka,

Verhallen P. F. J., Bevers, E. M., Comfunus, P. & Zwaal, R. F. A. (1987) T! & M&i, ?'. (1988) bio&m. Biophysl R&. Commun. 1 3 1201 1208

Biochid. Biophys. Acta 903,206-217