In vitro digestion of dystrophin by calcium-dependent proteases, calpains I and II

Biochimie (1992) 74, 565-570 © Soci6t6 franqaise de biochimie et biologie mol6culaire / Elsevier, Pads

565

In vitro digestion of dystrophin by calcium-dependent proteases,

calpains I and II P C o t t i n l, S P o u s s a r d 1, D M o m e t 2, JJ Brustis l, M M o h a m m a d p o u r I, J L e g e r 2, A D u c a s t a i n g l IISTAB, Laboratoire de Biochimie et Technologie des Aliments, Universit~ Bordeattr I, 33405 Talence Cedex; 21NSERM, U 300, Facult~ de Pharmacie, Avenue Charles.Flahaut, 34060 Montpellier Cedex, France (Received l0 February 1992; accepted 16 April 1992)

Summary - - Dystrophin is a cytoskeletal protein which is thought to play an important role in membrane physiology since its absence (due to gene deficiency) leads to the symptoms of Duchenne muscular dystrophy (DMD). Some disruption in the regulation of intracellular free Ca 2+ levels could lead to DMD-like symptoms. In this study, calpains, which are very active calcium-dependent proteases, were examined for their capacity to hydrolyse dystrophin in vitro. The results show that calpains are able to split dystrophin and produce breakdown products of different sizes (the degree of cleavage being dependent on the incubation time with proteases). The time-course of protease degradation was examined by Western immunoblot using three polyclonal sera which were characterized as being specific to the central (residues 1173-1728) and two distal parts of the molecule ie specific to the N-terminal (residues 43760) or the C-terminal (residues 3357-3660) extremities of the dystrophin molecule. The cleavage patterns of dystrophin showed an accumulation of some major protease-resistant fragments of high relative molecular mass (250-370 kDa). These observations demonstrate that calpains digest dystrophin very rapidly when the calcium concentration is compatible with their activation. For instance, it is clear that calpains first give rise to large dystrophin products in which the C-terminal region is lacking. These observations suggest that dystrophin antibodies specific to the central domain of the molecule should be used to detect dystrophin for diagnostic purposes

and before any conclusion as to the presenceor absence of dystrophincan be deducedfrom results obtainedusing immunoanalysesof muscle biopsies.

ealpain I / calpain !I / Ca2+.dependent proteolysis / eytoskeletal proteins / dystrophin Introduction Calcium-dependent neutral proteases, also called calpains I and II isolated from the cytosolic fraction of muscular tissue but which may be translocated to the cell membrane under certain conditions, seem to have distinct catalytic 80-kDa subunits and an identical regulatory 30-kDa subunit [1-5]. Their precise biological functions are still poorly understood despite the current increase in knowledge on their structural and enzymatic properties [6-131. Many reports have already shown that in vitro some membranous structures, myofibrillar and cytoskeletal proteins, are substrates for calpain I and II [14-21]. It is known that rapid protein degradation in vivo is always an important feature in several muscle wasting diseases. Three years ago the normal muscle cell membrane was found to contain a large molecule, named dystrophin, whose absence in Duchenne muscular dystrophy leads to rapid muscle degeneration in patients without the gene coding for dystrophin [22-25]. In some other myopathies, such as a

dystrophic myotony, dystrophin appears to be abnormal [26, 27]. In these specific cases it is not yet known if the abnormality results from repression of synthesis or from a putative degradation process of this protein. Elevated free calcium levels found in some pathological muscles [28] can promote an activation of the proteolytic activity of both calpains I and II. In the present study we investigated the degradation of dystrophin to determine whether this molecule is a potential substrate for these calpains. Dystrophin antibodies specific to different regions of the protein have been used to examine breakdown fragments. The results provide evidence that caution must be exercised before concluding from immunoanalyses that the detection of dystrophin, in its native form, becomes negative as the result of such a proteolytic process. Materials and methods Unless otherwise indicated, all chemicals and reagents were from Sigma in the highest grade and purity available.

566 Preparation of proteins Calpains I and II were purified from 2 kg of fresh rabbit muscle as previously described [29] and stored in 20 mM Tris-HCl, pH 7.5, 2 mM EDTA, 2 mM EGTA, 2 mM dithiothreitol, 1 mM NaN s at 2°C. Calpastatin (calpain-specific inhibitor) was purified from fresh muscle as previously described [30] and stored lyophilized at -80°C. The dystrophin-enriched extract was prepared according to [3 !] and the procedure terminated by elution on a Gzs (Pharmacia) column to eliminate ammonium sulfate and the excess of protease inhibitors added during dystrophin extraction.

Digestion with calpains The dystrophin-enriched extract in 20 mM Tris-HCI, pH 7.5, 2 mM dithiothreitol, 1 mM NaN3 was incubated with purified calpain I and I! at the same weight ratio (1000 dystrophin]! calpain) for 0 to 10 min at 30°C. Proteolysis was initiated by the addition of 0.01 mM calcium (final concentation) for calpain I and 1 mM for calpain II. At the indicated times, the proteolytic digestion was stopped

by boiling the sample in 4% SDS denaturing solution before the SDS-PAGE analysis. Controls were carried out under various experimental conditions (ie no calpain, calpain in presence of calpastatin in excess) and dystrophin-enriched preparations were found to be stable for 30 rain at 30°C. Controls were also performed to determine whether calpain solutions were still active at the end of the proteolytic time-course studies. Anti-dystrophin antibodies

Rabbit polyclonal antibodies were produced using fusion protein as immunogens [31]. The three sera used were obtained from different parts of dystrophin molectde ie the N-terminal segment (residues 43-760), the central region (residues 11731728) and the C-terminal segment (residues 3357-3660). SDS-PAGE and immunobiot analyses

Enzymatic digests were separated on a 2.5-9.5% gradient resolving polyacrylamide gel containing 25% glycerol and no stacking gel. The relative molecular mass markers were as follows: myosin heavy chain (200 kDa), beta-galactosidase (I 16 kDa), phosphorylase B (97 kDa), bovine serum albumin (68 kDa). The fractionated proteins were electro-transferred onto lmmobilon P sheets in a transfer buffer containing 0.1% SDS and incubated with each polyclonal antibody used at 1/50 dilution, then revealed with anti-rabbit lgG coupled to alkaline phosphatase.

Results

Comparative proteolytic patterns of dystrophinenriched preparation Limited proteolysis of dystrophin present in the dystrophin-enriched preparation of chicken smooth muscle was performed and compared using three dif-

ferent and specific sera (Materials and methods). The domain-specific polyclonal dystrophin antibodies allowed us to identify cleavage products relative to their original region. The gel pattern shown in figure 1A was revealed with antibodies directed against the dystrophin Cterminal domain. In the presence of calpain I (calcium: 0.01 mM) or calpain II (calcium: 1 raM) the degradation products were the same. The native dystrophin band (Mr = 420 kDa) was very quickly degraded (in 60 s) but no protein band corresponding to the breakdown products with sizes between Mr 420 kDa and 66 kDa were revealed (eg according to the migration of the markers). After 60 s of calpain proteolysis no protein band corresponding to the native form of dystrophin remained, thus indicating that the C-terminal region of the protein seemed to contain at least one or more proteolytie sites which were highly sensitive to the proteolytic activity. Using the antibody specific to the central part of dystrophin (fig 1B), the same proteolytic patterns were revealed either with calpain I or calpain II. Three main stable breakdown products were detected: these proteins have relative molecular masses of 370, 310 and 250 kDa, respectively. The comparative time-courses showed that with calpain I, cleavages occurred slightly faster than with calpain II. In our experimental conditions, the native 420 kDa protein decreased over the time of proteolysis and remained detectable after 10 rain of digestion by both calpains: this last observation (fig 1B) was due to the higher sensitivity of this antibody to the spectrin-like domain of dystrophin. The calpain proteolytic patterns (fig IC) revealed with the dystrophin antibodies directed against the Nterminal part of the molecule did not show significant differences between both proteolytic time courses with calpain I and calpain H. Two major proteolytic products were identified with Mr 370 and 310 kDa. Among these two products, only the 310-kDa breakdown product was found to be stable and resistant to further proteolytic degradation.

Discussion It was first concluded from this study that dystrophin was a good substrate, at least in vitro, for both calpains I and II. The use of domain specific dystrophin antibodies directed against the C-terminal part of the molecule clearly indicated that calpains could very rapidly degrade the C-terminal region of dystrophin. Different cleavage sites, determined according to the relative molecular mass of the main dystrophin fragments generated by the proteolytic activity of both

567 r - C a l p a i n I - - - -1 r - o a l p a i n II - - ~ I

i

I

i.

I

I I

I

f I

A

k, kDa

dystrophin Ib 420 kDa

205

,4 t

r - c a l p a i n I- - "-1, I I

f- --

c a l p a i n II - - 1

I

I

116 97 66

i

t

t x

\ a

B

dystr°phi" I~ ~ , ~ 4 2 0 kDa

b

•

d

•

f

kDa

h

I

~

_/

370 310

2so ~j I

r c a l p a i n I - - - -1 r - - c a l p a l n I I

I I t i

II 1 I I

kDo dystrophin •

C

4 2 0 kOa

370

~

~

310

Fig 1. Digestion of smooth muscle dystrophin by calpain I and calpain II. Smooth muscle dystrophin-enriched solution containing approximately 10 mg/ml of protein was incubated at 30°C in the presence of calpain I or calpain 1I (0.1 mg/ml) and free Ca ~+ (0.010 mM and 1 mM respectively). The times of incubation were: lanes a, e: 0 min; b, f: 0.5 min; c, g: 1 min; d, h: 10 min. Immunodetection of dystrophin and its breakdown products were performed with polyclonal antibodies and a secondary rabbit antibody coupled with alkaline phosphatase (as described in Materials and methods). A. Polyclonal antibody raised against the C-terminal domain (aminoacids 3357-3660). B. Polyclonal antibody raised against the spectrin-like domain (aminoacids 1173-1728). C. Polyclonal antibody raised against the actin-binding domain (aminoacids 43-760). The electrophoretic mobility of the molecular weight markers used is shown by black arrows in the right part of the gel.

568

I III1,,

1

1.....

L]II I

II

420KDa

370KDa

310kDa

250kDa

X

Y

Z

Fig 2. Localization of calpain I and lI cleavage regions within a smooth muscle dystrophin. A schematic drawing illustrating the different dystrophin domains as previously described [35]. Localization of the three regions used to obtain sites of recognition by the polyclonal antibodies for these experiments are shown in dashed area. -, Cleavage sites for calpains I and II. Z, preferential cleavage region; Y, secondary cleavage region; X, third cleavage region. The M, of the generated breakdown products are indicated in kilodaitons (kDa). calpains, are proposed in a linear schematic drawing of the dystrophin molecule in figure 2. It would certainly be preferable to identify the exact cleavage sites rather than guess at them by transferring individual blotted bands onto a sequencer, but accurate results have been difficult to obtain due to difficulties in obtaining pure dystrophin-enriched solutions. As shown in figure 2, the first cleavage site is denoted as Z at the C-terminal end. This result was suggested by the presence of a proteolytic product of M, = 370 kDa detected with both N-terminal and central-specific dystrophin sera. The 310-kDa breakdown product detected by the antibody directed against the N-terminal part of dystrophin must have been derived from the 370-kDa fragment shortened in its C-terminal region by a cleavage between the central spectrin-like domain and the cystein-rich region (cleavage site Y). In addition, the antibody directed against the central portion of the dystrophin molecule enabled detection of a 250-kDa protein band which was not detected by the antibodies directed against either of the distal pans of the molecule. This fragment must have been derived from another cleavage site splitting the N-terminal pan from the central spectrin-like domain (cleavage site X).

Although there are other possibilities for generating different 310- or 250-kDa fragments the cleavage sites X, Y and Z appear to be the most probable. In fact, there are no definitive ways of mapping these cleavage sites but the central region of dystrophin showed strong resistance towards calpain hydrolyses, which is an important result from this study. According to these results, the use of antibodies directed towards the central domain of dystrophin would be more optimal to prove the absence of dystrophin from immunohistochemical staining and immunoblot analyses of pathological muscles. The initial greater affinity for calpains towards the C-terminal domain is very important, because this region, via the glycoprotein complex, is thought to interact with the muscle membrane [32-34]. Moreover, the increase in free intracellular calcium ions described under some pathological conditions [28] could activate calpains and thus lead to dystrophin detachment from the muscle membrane. Some illnesses could therefore exhibit dystrophin-deficiency-like symptoms due to proteolytic dissociation of dystrophin from the muscle membrane. Experiments with artificial increases in calcium-ion concentration are being conducted in order to investigate whether the calcium stimulated proteolytic activities of both calpains I and I1 could effectively lead to dystrophin deficiency-like symptoms in vivo.

569

Acknowledgments We are grateful to Dr Jean J Leger for his helpful advice and discussions, P Lochet for technical assistance and S Oree for rereading the manuscript. This work was supported by Institut National de la Recherche Agronomique and ~.ssociation Fran~:aise contre les Myopathies.

References 1 Ohno S, Emori Y, laaajoh S, Kawasaki H, Kisaragi M, Suziki K (1984) Evo!utionary origin of a calcium-dependent protease by fusLm of genes for a thiol protease and calcium-binding protein. Nature 312, 566-570 2 Sakihama 1", Kakidani H, Zenita K, Yumoto N, Kikuchi T, Sasaki I", Karmagi R, Nakanishi S, Ohmori M, Takio K, Titani K, Murachi T (1985) A putative Ca2+-binding protein structure of the light subunit , " porcin calpain elucidated by molecular cloning and protein sequence analysis. Proc Natl Acad Sci USA 82, 6075--6079 3 Emori Y, Kawasaki H, Imajoh S, Kawashima S, Suzuki K (1986) Isolation and sequence analyses of cDNA clones from the small subunit of rabbit calcium-dependent prntease. J Bioi Chem 261, 9472-9476 4 Aoki K, Imajoh S, Ohno 5, Emori Y, Koike M, Kosaki G, Susuki K (1986) Complete amino acid sequence of human CaZ+-activated neutral proteases deduced from its cDNA sequence. FEBS Lett 205, 313-317 5 Ohno S, Emori Y, Suzuki K (1986) Nucleotide sequence of a cDNA coding for the small subunit of human calcium-dependent protease. Nucleic Acids Res 14, 5559 6 Mellgren RL (1987) Calcium activated proteases: an enzyme systeme active at cellular membranes? FASEB J 1, 110-115 7 Cottin P, Vidalenc PL, Ducastaing A (1981) Ca2+-dependent association between a Ca2÷-dependent neutral protease and its specific inhibitor. FEBS Lett 136, 221224 8 Takanako E, Maki M, Hatanaka M, Mori H, Zenita K, Sakihama 1", Kannagi R, Marti T, Titani K, Mumchi T (1986) Evidence for the repetitive domain structure of pig calpastatin as demonstrated by cloning complementary DNA. FEBS Lett 208, 199-202 9 Imajoh S, Kawasaki H, Emori, Sugihara H, Suzuki K (1987) A fragment of an endogenous inhibitor product in Escherichia coli for calcium-activated neutral protease (CANP) retains an inhibitory activity. FEBS Lett 215, 274-278 10 Mellgren RL (1988) On the mechanism of binding of calpastatin, the protein inhibitor of calpains, to biologic membranes. Biochem Biophys Res Commun 150, 170-176 11 Cong J, Goll DE, Peterson AM, Kaprell HP (1989) The role of autolysis in activity of the Ca2~'-dependent proteinases (u-calpain and m-calpain). J Biol Chem 264, 10096-10103 12 DeMartino GN, Huff CA, Croall DE (1986) Autoproteolysis of the small subunit of calcium-dependent protease II activates and regulates protease activity. J Biol Chem 261, 12047-12052 13 Tan FC, Goll DE, Otsuka Y (1988) Some properties of the mUlimolar Ca2+-dependcnt proteinase from bovine cardiac muscle. J Mol Cell Cardio120, 983-997

14 Johnson P (1990) Calpains (intracellular calcium activated cysteine proteinases): structure-activity relationships and involvement in normal and abnormal cellular metabolism. lntJBiochem 22, 8, 811-822 15 Wang KW, Villalobo A, Roufogalis BD (1989) Calmodulin-binding proteins as calpain substrates. Biockem J 262, 693-706 16 Boivin P, Galand C, Dhermy D (1990) In vitro digestion of spectrin, protein 4.1 and ankyrin by Calcium dependent neutral protease (calpain I). Int J Biochem 22, 1479-1489 17 GIaser T, Kosower NS (1986) Fusion of rats erythrocytes by membrane-mobility agent A2C depends on membrane proteolysis by a cytoplasmic calpain. Eur J Biochem 159, 387-392 18 Croall DE, Morrow JS, DeMartino GN (1986) Limited proteolysis of the erythn~yte membrane skeleton by calcium-dependent proteinases. Biochim Biophys Acta 882, 287-296 19 Fox JE, Reynolds CC, Morrow JS, Phillips DR (1987) Spectrin is associated with membrane-bound actin filaments in platelets and is hydrolyzed by the Ca2÷-dependent protease during platelet activation. Blood 69, 537-545 20 Azanza JL, Raymond J, Robin JM, Cottin P, Ducastaing A (1980) Prot6inase neutre Ca2+-d6pendante du muscle squelettique de lapin: activit6 sur les prot6ines myofibrillaires. Biochiraie 62, 481-486 21 Harris AS, Croall DE, Morrow JS (1989) Calmodulin regulates fodrin susceptibility to cleavage by calciumdependent protease. J Biol Chem 264, 17401-17408 22 Monaco A (1989) Dystrophin, the protein product of the Duchenne/Becker muscular dystrophy gene. Trends Biochem Sci 14, 412--415 23 Zubrzycka-Gaam EE, Hutter OF, Karpati G, Klamut HJ, Bulman DE, Hodges RS, Worton RG, Ray PN (1991) Dystrophin is tightly associated with the sarcolemma of mammalian skeletal muscle fibers, Exp Cell Res 192, 278-288 24 Bonilla E, 5amit CE, Miranda AF, Hays AP, Salviati G, Dimauro S, Kunkel LP, Hoffman EP, Rowland LP (1988) Duchenne muscular dystrophy: deficiency of dystrophin at the muscle cell surface. Cell 54, 447-452 25 Zubrzycka-Gaam EE, Bulman DE, Karpati G, Burghes AH, Belfall BK, Klatmut HJ, Talbot J, Hodgest RS, Ray PN, Worton RG (1988) The Duchenne muscular dystrophy gene product is localized in sarcolemma of human skeletal muscle. Nature 333, 466--469 26 Eiberg H, Molar J, Nielsen LS (1981) Sixth Int Congress Hum Genet, Jerusalem, Abstract 147 27 Arahata K, Hoffman EP, Kunkel LM, Ishiura S, Tsukahara T, Ishihara, T, Sunohara N, Nonaka 1, Osawa E, Sugita H (1989) Dystrophin abnormalities by immunofluorescence and immunoblot analyses. Proc Natl Acad Sci USA 86, 7154-7158 28 Turner PR, Westwood T, Chester MR, Steihardt RA (! 988) Increased protein degradation results from elevated free calcium levels found in muscle from mdx mice. Nature 335, 735-738 29 Garret C, Cottin P, Dufourcq J, Ducastaing A (1988) Evidence for Ca2+-independent association between calpain II and phospholipid vesicles. FEBS Lett 227, 209214 30 Otsuka Y, Goll DE (1987) Purification of the Ca2÷-dependent proteinase inhibitor from bovine cardiac muscle and its interaction with the millimolar Ca2+-dependent proteinase. J Biol Chem 262, 5839-5851

570 31

Pons E Augier N, Heilig R. I.,6ger J, Momet D, L6ger JJ (1990) Isolated dystrophin molecules as seen by electronic microscopy. Proc Natl Acad Sci USA 87,7851-7855 32 Ervasti JM, Campbell KP (1991) Membrane organization of the dystrophin complex. Cell 66, 1121-1131 33 Hoffman EP, Kunkel LM (1989) Dystrophin abnormalities in Duchenne/Becker muscular dystrophy. Neuron 2, 1019-1029

34 Carpenter S, Karpati G, Zubrzycka-Ga~m E, Bulman DE, Ray PN, Worton RG (1990) Dystrophin is localized to the plasma membrane of human skeletal fibers by electronmicroscopic cytochemical study. Muscle Nerve 13, 376380 35 I_.6gerJJ, Augier N, L6ger J, Mornet D, Pons F (1991) La ou les dyslrophine(s), trois a n n . s apr~s leur d~couverte. Medecine/Sciences 7, 805-819