Expression of Lysosomal Cathepsin B during Calf Myoblast-Myotube ...

THEJOURNAL OF BIOLOGICAL CHEMISTRY I(:’

Vol. 266, No. 21, Issue of July 25, pp. 14104-14112,1991 Printed in U.S.A .

1991 by The American Society for Biochemistry and Molecular Biology,Inc.

Expression of Lysosomal Cathepsin B during Calf Myoblast-Myotube Differentiation CHARACTERIZATION OF A cDNAENCODINGBOVINECATHEPSIN

B* (Received for publication, September 4, 1990)

Daniel M. BechetS, Marc J. Ferrara, SylvieB. Mordier, Marie-Paule Roux, ChristianeD. Deval, and Alain Obled From the Unite de Recherches sur 1’Expression des Proteases, SRV Theix, Institut National de la Recherche Agronomique, 63122 Ceyrat France

Expression of lysosomal cysteine proteinases was studied during fetal calf myoblast-myotube differentiation. Activitiesof cathepsin Band L, but not cathepsin H, increase during bovine myogenic differentiation. In fetal muscle, cathepsin B and L activities are 2-4-fold orders of magnitude lower than in cultured Emyoblasts. Active-site titrations of cathepsin B with 6 4 nevertheless reveal similar concentrations of active cathepsin B in myoblasts and myotubes, but 5-6-fold lower concentrations in fetal muscle. To specify whether concentrations of cathepsin B a r e related to levelsof cathepsin B transcript, a cDNA clone encoding bovine cathepsin B was isolated and liquid hybridizations were performed with 32P-riboprobescomplementaryto the mRNA. In agreement with active-site titrations, there is no difference in cathepsin B mRNA levels between cultured myoblasts and myotubes, but lower levelsof mRNA are found in fetal muscle. Concentrations of activecathepsin B therefore reflect levelsof cathepsin B mRNA. Kinetic studies revealedthat the catalyticefficiency (kcat/Km) of cathepsin B is 2-3-fold higher in myotubes than inmyoblasts. The increase in cathepsin B activity during calf myoblast-myotube differentiation is thus due to modifications of enzymatic properties, and not of enzyme concentrations. The different catalytic efficiency of cathepsin Bin myotubes and myoblasts was related neither to modifications of mRNA size, as revealed by Northern blot analysis, nor toa different M , of the active enzyme, as revealed by affinity labeling with benzyloxycarbonyl-Tyr(-’251)-Ala-CHN~, but to limited differences in cathepsinB isozymes.

H (Ishido et al., 1987), and cathepsin D (Faust et al., 1985) have been isolated. As a rule, cathepsins are synthesized as large molecular weightprecursors containing signal- and prosequences. Post-transcriptional processing has been extensively studied for cathepsin D (Conner et al., 1987), and more recently for cathepsins B (Nishimura et al., 1988a, 1988b), L (Nishimura et al., 1988a), andH (Nishimura et al., 1988b). There are now several examples of tumor cells exhibiting over-expression and secretion of cathepsin B (Sloane et al., 1987), cathepsin L (Gal and Gottesman, 1986), or cathepsin D (Rochefort et al., 1987).Over-expression of cathepsin is most probably due t o induction of cathepsin gene transcription (Doherty et al., 1985; Cavaill6s et al., 1989), but there also is evidence that high levels of secretion from tumorcells are associated with alteration in the secretion pathway (Dong et al., 1989). Secretion of cathepsin by tumor cells may be important in developing metastasis (Sloane et al., 1987; Rochefort et al., 1987). The recent descriptions of the genomic structures of mouse cathepsin B (Ferrara et al., 1990) and rat cathepsin L (Ishido et al., 1989) will probably lead to a better understanding of the regulation of cathepsin gene transcription. The mechanisms bywhich cathepsin expression iscontrolled under non-pathological conditions are poorly understood on the other hand. Important variations in lysosomal cathepsin activities existbetween species (Bbchetet al., 1986). There is a tissue-specific distribution of cathepsins (Kominami et al., 1985; Bando et al., 1986), and cathepsinB contents have been shown to be partly correlated with levels of the corresponding transcripts in different tissues (San Segundo et al., 1986). Despite very low levels in skeletal muscle, the presence of cathepsins within muscle fibers has been demonstrated by histochemical (Stauber and Ong, 1981) and immunohistochemical (Stauber et al., 1985; Taylor et al., 1987) Lysosomal endopeptidases are thought to play an important role in protein catabolism. These proteinases comprise ca- investigations.Differentiation of skeletal muscleinvolves thepsins B, H, and L, which are cysteine proteinases, and the withdrawal of undifferentiated myoblasts from thecell cycle, asparticproteinasezathepsin D. Completeaminoacid se- fusion into multinucleated myotubes, and coordinate inducquences have been determined for cathepsin B (Takio et al., tion of muscle-specific gene products. Activities of lysosomal 1983; Ritonja et al., 1985), cathepsin L (Dufour et al., 1987; cysteine proteinases also increase during myogenic differen1988), cathepsin tiation(Bird et al., 1981; Kirschke et al., 1983),butthe Ritonja et al., 1988; Towatari and Katunuma, H (Takio et al., 1983; Ritonja et al., 1988) and cathepsin D mechanism by which cathepsinactivitiesincreaseduring (Shewale and Tang,1984), and several cDNA clones encoding myogenesis isnot fully understood,andit is particularly cathepsin B (San Segundo et al., 1985; Chan et al., 1986), uncertain whether this involves a transcriptional activation cathepsin L (Portnoy et al,, 1986; Troen et al., 1987), cathepsin of cathepsin gene expression (Colella et al., 1986) or posttranscriptional controls. In the present study we have isolated * The costs of publication of this article were defrayed in part by a cDNA clone encoding bovine cathepsin B and have investhe payment of page charges. This article must therefore be hereby of cathepsin B during bovine myoblastmarked “advertisement” in accordance with 18 U.S.C. Section 1734 tigated the expression myotube differentiation. Evidence is presented showing that solely to indicate this fact. theincreasein lysosomal activities of cathepsin B is not $ To whom correspondence should be addressed.

14104

Expression of Cathepsin B during Myoblast-Myotube Differentiation

14105

between confluent myoblasts (80 f 17% ( n = 9)), myotubes (78 f 13% ( n = 8 ) ) ,and fibroblasts (74 f 9% ( n = 3)), although lower for fetal muscle(11.5 f 0.6 ( n = 4)). Repeated freezing-thawing and extraction of post-cytosolic pellet with sodium phosphate buffer, pH 5.8, showed that 90-95% recovery of cathepsin total activity was EXPERIMENTALPROCEDURES consistently obtained in the first supernatant. Protein concentration was determined according to Bradford (1976) using bovine serum Materials-Cell culture media (Hanks' modified Eagle's medium, albumin as standard. Dulbecco's modified Eagle's medium, and medium 199) and preseAssays for Bz-Phe-Val-Arg-NMec (Bromme et al., 1989), Z-Arglected fetal calf serum were purchased from Flow Laboratories (Pu- Arg-NMec (cathepsin B), Z-Phe-Arg-NMec (cathepsins B and L), teaux, France). Cathepsin substrates (Z'-Phe-Arg-NMec,Z-Arg-Arg- and Arg-NMec (cathepsin H ) hydrolysis, activation studies with Cys NMec,Arg-NMec) were obtainedfromCambridgeResearch Bio- or DTT (Barrett and Kirschke, 1981), and inhibition studies withZchemicals (Cambridge, UnitedKingdom).LeupeptinandBz-PhePhe-Phe-CHNp (Kirschke and Shaw,1981) were as described in the Val-Arg-NMec were from Bachem (Bubendorf, Switzerland) and E- references cited. Active-site titration of cathepsin Bby E-64 was 64 from Protein Research Foundation (Ozaka, Japan). Z-Phe-Pheperformed according to Barrett et al. (1982). Kinetic contants ( L , CHN, and Z-Tyr-Ala-CHNy were kindly provided by Dr. E. Shaw V,) were determined by the method of Wilkinson (1961); kcat was (Basel, Switzerland). FPLC apparatus, Mono-S columns (H 10/10) estimated as V,,,/Eo, with E, being theconcentration of enzyme and electrophoretic materialswere purchased from Pharmacia (Upps- estimated by active-site titration. ula, Sweden).Ampholytes and Gel-Fix adhesive plastic sheets for FPLC on Mono-S columns of lysosomal extracts was performed in isoelectrofocusing wen: obtained from Serva (TEBU, France) and 30 mM sodium phosphate buffer, p H 5. 8, with a linear gradient of cellulose acetatepaperfromSartorius(France).Plasmidvectors NaCl (0-0.8 M ) and ata flow rate of 0.8 ml min-~'. Isoelectrofocusing (pUC18, pGemini-blue)andSP6RNA polymerase were obtained was performed as previously described (Deval etal., 1990) in 0.2-mm from Promega-Biotech (COGER, Paris). M13 phage vectors and T 7 thick polyacrylamide gels. After focusing, the gel was rapidly blotted DNA polymerase were from Pharmacia (Uppsula, Sweden) and reon a cellulose-acetate sheet soaked with 100 PM Z-Arg-Arg-NMec in striction enzymes fromBoehringerMannheim(Meylan,France). 0.1 M sodium acetate buffer, pH 5.0, containing 20 mM DTT. After Na'"1 (100 mCi/ml) and [?'P]CTP (400 Ci/mmol) were provided by 10 min at 42 "C, NMec was revealed on the blot underUV light (360 Amersham(Les Ulis, France). All otherenzymes and chemicals nm) and thefluorescence at 460 nm photographed using a UV filter (analytical grade) were purchased from Sigma. (ZA, Kodak). Cell Cultures-Primary cultures of calf muscle cells were prepared Affinity Labeling of Cathepsin B-Z-Tyr-Ala-CHN? was iodinated from 3- to 6-monthold fetal calf muscle (BicepsFemoris). The using theIodogen method (Fraker and Speck,1978) and according to procedure was essentially that described by Buckingham et al. (1976), the protocol of Mason et al. (1989). Cathepsin B, eluted in 30 mM with the modifications previously reported (Bbchet et aL, 1990). Calf sodium phosphatebuffer, pH 5.8, in Mono-S FPLC chromatography, myoblasts were confluent a t 3 days andby the 4th day theybegan to was preincubated with 1 mM DTT and 1 mM EDTA for 2 h a t 4 "C fuse into multinucleatedmyotubes. The percentage of nuclei in myo- to activate the enzyme. Affinity labeling was performed in the prestubes (fusion index) was determined on parallel cultures using Giemsa ence of 1p M radiolabeled inhibitor for 30 min at 30 "C. Proteins were staining. Fusion index was maximal (70-90%) by 5 days of culture. recovered by precipitation with 5% trichloroacetic acid, 0. 1% Triton Spontaneous contractions of the myotubes were observed 3-4 days X-100, washed three times with HyO, and dissolved in sample buffer after fusion. On the basis of morphological differences, muscle cell for SDS-PAGE (Laemmli, 1970). For SDS-PAGE, a 1 2 % polyacrylcultures contained less than 10%fibroblasts. amide resolving, 7.5% polyacrylamide stacking gel was used. AutoraFor cultures of bovine fibroblasts a 3-5-cm2 portion of the fetal diography of iodinated proteins was performed with Hyperfilm-MP skin was minced and digested for 1 h at room temperature in 30 ml x-ray films (Amersham Corps) at -70 "C with intensifying screens. of 0.05% trypsin in phosphate-buffered saline, p H 7.4. Dissociated Molecular mass markers for all polyacrylamide gels were phosphocells were collected by centrifugation at 1000 X g for 10 min, resus- rylase b (94 kDa), bovine serum albumin (67 kDa), ovalbumin (43 pended in Dulbecco's modified Eagle's medium containing 10% fetal kDa), carbonic anhydrase (30 kDa), soybean trypsin inhibitor (20 calf serum and 40 pg/ml Gentamycin, and plated in 75-cm2 culture kDa), and a-lactalbumin (14 kDa). flasks. Incubation was at 37 "C with 5% COz, and fibroblasts were Isolation of a Bovine Cathepsin B cDNA and Dot-Blot Hybridizafed fresh medium at 24 h. Bovine fibroblasts reached confluence by tion-Total RNA was isolated from cultured cells or fetal muscles by 4-5 days of culture. a slightly modified guanidinium LiCl procedure (Cathalaet al., 1983) Cathepsin Actiuities-The presence of cytosolic inhibitors for cysand poly(A)+ mRNA selected by oligo(dT)-cellulose chromatography. teine proteinases in muscle (Kirschke et al., 1983; Bige et al., 1985; A bovine fibroblast cDNA library was constructed by cloning, into Ouali et al., 1986) precludes direct assay of lysosomal cathepsins in EcoRI sites of Xgtll, double-stranded cDNA obtained from bovine unfractionated extracts. In this study, cathepsin activities were therefibroblast poly(A)' mRNA. A cDNA probe (prCB3; San Segundo et fore only studied after discarding cytosolic fractions. The cell monoal., 1985) encoding rat cathepsin B was used to screen the cDNA layer was washed with phosphate-buffered saline and detached from library mainly as described by Maniatis et al. (1982). One positive the flask with a rubber scraper. Subsequent procedures were per- recombinant clone (Xmp8) was isolated and purified. The phage DNA formed a t 0-4 "C, unlessspecified otherwise. Cells were homogenized was isolated, digested with EcoRI, and the released cDNA fragment with a Potter-Elvehjem homogenizer in 10 mM potassium phosphate was subcloned into plasmid vector pUC18 for restriction endonuclease buffer, pH 7.4, containing 0.25 M sucrose, 50 mM KC1 and 1 mM EDTA. An aliquot of cellular homogenate was made in 0.2% Triton mapping. The cDNA insert was digested with appropriate restriction X-100 and stored at -20 "C until further analysis of N-acetyl-0-D- enzymes,subcloned into M13 phage vectors, and DNAsequences were determined by the dideoxynucleotide chain termination method glucosaminidase activity and protein content. The cellular homogenate was centrifuged for 20 min at100,000 X g, the pellet resuspended (Sanger et al., 1977) using T7 DNA polymerase (Pharmacia). Dotblots were made with GeneScreen Plus membranes according to the in 30 mM sodium phosphate buffer, p H 5.8, and frozen overnight. instructions of the manufacturer (Du Pont-New England Nuclear). After thawing, an aliquot was also made 0.2% in Triton X-100 and Riboprobes and RNAse Protection Assay-The cDNAinsert of stored for N-acetyl-P-D-glucosaminidase activity measurements. The supernatant recovered after 20-min centrifugation at 100,000 X g was Amp8 was ligated into the EcoRI site in the polylinker of pGeminiused for cathepsin activity studies and was referred to as lysosomal blue Vector. The plasmid (pGem8) was linearized by BglII and :'*Pextract. Preparation of lysosomal extracts from muscle tissues were riboprobes were synthesized (Firestein et al., 1987) using SP6 RNA polymerase. The riboprobe is 550 bases long and is complementary as described previously (Obled et al., 1984). N-Acetyl-6-D-glucosaminidase total activities were determined in cell and lysosomal ho- to a fragment containing pGemini-bluepolylinker and the 3' end of mogenates to estimate yield in lysosomes (BCchet et al., 1986). Yield cathepsin B mRNA (490 bases). Myoblasts or myotubeswere washed in lysosomes in post-cytosolic pellets were not significantly different in phosphate-buffered saline and directly dissolved in buffer D (5 M guanidinium thiocyanate, 0.1 M EDTA, pH 7.0) a t a final concentration of 3.10' nuclei/ml. Total RNA from fetal muscles or myotubes I The abbreviations used are: Z-, benzyloxycarbonyl; FPLC, fast was also dissolved in buffer D (1 mg/ml). Solubilized cells or total protein liquid chromatography; Bz-, benzoyl-; -NMec, 7-(4RNA were mixed with 1-2 ng of riboprobe in a total volume of 15 pl methy1)coumarylamide; SDS, sodium dodecyl sulfate; E-64, L-3-car- of buffer D, heated 5 min at 60 "C, and hybridized 18 h at 25 "C. hoxy-trans-2,3-epoxypropionyl-leucylamido-(4-~anidino)butane; RNase A digestion was carried as described by Firestein etal. (1987). DTT, dithiothreitol; PAGE, polyacrylamide gel electrophoresis. Hybridization to cathepsin B mRNA protects a 490-base riboprobe

related to any change in active enzyme concentration nor modification in levels of mRNA encoding cathepsin B, but to major changes in enzymatic properties.

ExpressionCathepsin of

14106

B during Myoblast-Myotube Differentiation

fragment from RNase digestion. After proteinase K treatment, phenol/chloroform extraction, and ethanol precipitation, protected fragments were separated on 6% polyacrylamide sequencing gels. The gels were dried and analyzed by autoradiography. Northern Blot Analysis-Poly(A)+ mRNA from myoblasts or myotubes was subjected to 1% agarose gel electrophoresis containing formaldehyde and transferred overnight to Hybond N+ membrane (Amersham Corp.) with 20 X SSC (1 X SSC is 0.15 M NaCl, 15 mM trisodium citrate, pH 7.0). The transfer was fixed 5 min in 0.05 M NaOH and neutralized by washing 2-3-fold with 2 X SSC. The membrane was washed for 1 h at 65 "C in 2 X SSC, 0.5% SDS, and prehybridized for 2 h at 55 "C in 50% formamide containing 5X SSC, 5 X Denhardt's reagent, 50 mM sodium phosphate buffer, pH 6.5, 0.1% SDS, 0.1 mg/ml sonicated salmon sperm DNA, and 10 rg/ml yeast tRNA. Hybridization was carried out for 16 h at 55 "C in 10 ml of the same solution containing 100 ng of riboprobe (lo7 cpm). The membrane was washed for 20 min at 55 "C successively in 2 X SSC containing 1%SDS, 0.5 X SSC containing 0.1% SDS, and 0.1 X SSC containing 0.1% SDS. Labeled bands were detected by autoradiography. Similar results were obtained when, after hybridization, the membrane was treated for 1 h at 30 "C with 1pg/ml RNase A in 2 X

ssc.

RESULTS

Characterization of Cysteine Proteinase Actiuities-To clearly distinguish the different cysteine proteinase activities, crude lysosomal extracts were fractionated by FPLC on a Mono-S column (Fig. 1).Similar profiles of Z-Phe-Arg-NMec, Z-Arg-Arg-NMec, and Arg-NMec hydrolysis were observed in lysosomal extracts from myoblasts, myotubes, or fetal muscle (Fig. 1 a-c, respectively). Cathepsin B, as revealed by hydrolysis of its specific substrate Z-Arg-Arg-NMec, was not retained on Mono-S column at pH 5.8. Cathepsin B activity also corresponded to the major peak of Z-Phe-Arg-NMec hydrolysis. Two other minor peaks of cathepsin L-like activity 40

with Z-Phe-Arg-NMec, whichwere not associated with ZArg-Arg-NMechydrolysis, were eluted at 0.5 M NaCl (referred to as CL1) and in the 2 M NaCl wash (referred to as CL2). Only very weak cathepsin-H Arg-NMec hydrolyzing activity was detected in myoblasts or myotubes, and was eluted at 0.2 M NaCl on Mono-S at pH 5.8. For comparison, we also show that primary cultures of bovine fibroblasts exhibited very low CL1 activity (Fig. Id) and were thus clearly different from myoblasts, myotubes, or fetal muscle. The different activities recovered from Mono-Sshared properties that characterize lysosomal cysteine proteinases. Hydrolysis of Z-Phe-Arg-NMec, Z-Arg-Arg-NMec, and ArgNMec required a thiol activator, and maximum activities were obtained with 5-10 mM Cys or 2-4 mM DTT. Complete inhibition was achieved inthe presence of M E-64. In addition, cathepsin HArg-NMec activity was not affected by bestatin, an inhibitor of Arg- and Leu-aminopeptidases. In addition to absence of activity on Z-Arg-Arg-NMec, cathepsin L can also be distinguished from cathepsin Bby its inhibition by Z-Phe-Phe-CHN? (Kirschke and Shaw, 1981; B6chet et al., 1986). Inhibition by Z-Phe-Phe-CHN? of the ZPhe-Arg-NMec activity of cathepsin B, L1, and L2 eluted from Mono-S are compared in Fig. 2. Only 5-10% inhibition of cathepsin B activity was observed with 1 g M Z-Phe-PheCHNz. Cathepsin L1 was totally inhibited on 15-min incubation with 0.5 p~ Z-Phe-Phe-CHNz. Cathepsin L2 was more resistant to Z-Phe-Phe-CHN2 inhibition, but exhibited only 20-30% residual activity at 1 g~ Z-Phe-Phe-CHN2 concentration. Similar patternsof inhibition were observed whether myoblasts (Fig. 2a), myotubes (Fig. 2b), or fetal muscle (Fig. Zc) were considered. In Fig. 2 4 we show that in lysosomal extracts, the proportion of Z-Phe-Arg-NMec hydrolyzing activityinhibited by Z-Phe-Phe-CHNz was not significantly different between myoblasts, myotubes, or fetal muscle, and wasclose to 20-25%. This value also corresponded to the

20

0

80

40

0 20 (c) Fetal muscle 10

0 60

(d) Fibroblast 40 70

20

J

-8

"I

-6

Log Z-Phe-Phe-CHN2 Coneenlmtioo

0

e

FIG. 2. Comparative inhibition by Z-Phe-Phe-CHNZ of cathepsins B, L1, and L2. Cathepsin B (O), L1 (O), and L2 (A) FRACTION No fractions recovered from Mono-S FPLC of myoblasts ( a ) , myotubes FIG. 1. Comparative Mono-SFPLC of lysosomal extracts ( b ) , and fetal muscle (c) extracts were preincubated for 15 min with myotubes ( b ) ,fetal muscle (c),and increasing concentrations of Z-Phe-Phe-CHN2, and assessed for refrom bovine myoblasts (a), fibroblasts ( d ) . Elution involved a linear NaCl gradient (0-0.8 M) sidual Z-Phe-Arg-NMec hydrolyzing activity. In d, inhibition by Zfrom fractions 5 to 20, and a pulse of 2 M NaCl at fraction 23. Assays Phe-Phe-CHNn of Z-Phe-Arg-NMec hydrolyzing activity was asfor Z-Arg-Arg-NMec (A),Arg-NMec (m), and Z-Phe-Arg-NMec (0) sessed inunfractionated lysosomal extracts from myoblasts (A), myotubes (O),and fetal muscle (W). were carried out for each fraction. 10

20

30

Expression of Cathepsin B during Myoblast-Myotube Differentiation proportion of Z-Phe-Arg-NMec hydrolyzing activity eluted as CL1 CL2 from Mono-S. Cathepsin B, H, andL1 activities detected in these studies therefore exhibit properties very similar to those already described for cathepsin B, H, andL purified from other tissues or other species (Kirschke and Barrett, 1987; Dufour et al., 1987; Deval et al. , 1990). Cathepsin L2, however, was not as strongly inhibited as L1 by Z-Phe-Phe-CHN2, and in this respect may resemble bovine spleen cathepsin S activity (Kirschke et al., 1986; Bromme et al., 1989). The ratio of the activities with Bz-Phe-Val-Arg-NMec and Z-Phe-Arg-NMec has been reported to clearly distinguish cathepsin L from cathepsin S (Kirschke et al., 1984, 1986) and was then compared for L1 andL2. This ratio was similar for both enzymes (0.097 for L1 and 0.092 for L2) and close to values reported to characterize cathepsin L (Kirschkeet al., 1984,1986). From these data we conclude that CL1 and CL2 probably correspond to different isozymes of bovine cathepsin L. Cathepsin ActivityLevels-Specific activities of cysteine proteinases were then directly compared in unfractionated lysosomal extracts. Z-Arg-Arg-NMec and Arg-NMec were used as specific substrates for cathepsin B andH, respectively. According to the results described earlier, cathepsin L (CL1 + CL2) activity was accurately assessed in lysosomal extracts by comparing Z-Phe-Arg-NMec hydrolysis with or without inhibition by 1 ~ L MZ-Phe-Phe-CHNn.As reported in Fig. 3, bovine myoblast-myotube differentiation was associated with a significant 2-3-fold increase in cathepsin B and activities. L In fetal muscle, cathepsin B and L activities were also much lower than those measured in primary cultures. In every case, cathepsin H Arg-NMec hydrolyzing activity was very weak, with no significant difference between myoblasts, myotubes, or fetal muscle. For comparison, we also show that primary cultures of bovine fibroblasts reveal similar levels of cathepsin B- and H-specific activity, but lower levels of cathepsin L than myoblasts. Active Cathepsin B Concentrations-Further studies were then carried out to determine whether similar mechanisms could explain the differences in cathepsin B lysosomal activity observed between fetal muscle, myoblasts, and myotubes. Modifications in total activities of cathepsin B could result from eitherdifferent concentrations of active enzyme, or from modifications in enzymatic properties. The epoxysuccinyl

+

Q

14107

peptide E-64 is a specific and irreversible inhibitor of cysteine proteinases (Hashida et al., 1980; Barrett et al., 1982). Its stoichiometric interaction with cysteine proteinases allows determination of active molarity of these enzymes. In practice, active-site titrationsperformed with crude lysosomal extracts would result in simultaneous binding of E-64 to cathepsin B, L, and H. Cathepsins were therefore separated by FPLC on Mono-S before active-site titration. Titrationwith E-64 could not be accurately performed with cathepsin L or cathepsin H fractions, probably because of insufficient enzyme concentrations, and further studies were then focused on cathepsin B expression. As shown in Fig. 4, no significant difference in active cathepsin Blevels wasdetected between myoblasts and myotubes. Incontrast, in fetal muscle, the lowerlevel of cathepsin B activity was also associated with levels of active enzyme 5-6-fold lower than those measured in primary cultures. Isolation of a cDNA Encoding Bovine Cathepsin B-In order to determine whether changes in active cathepsin B concentrations between fetal muscle and primary cultures were related to a modification in cathepsin B gene expression, a cDNA clone (Amp8)encoding bovine cathepsin Bwas isolated from a bovine cDNA library using prCB3 as a probe. The restriction map, nucleotide sequence, and deduced amino acid sequence for Amp8 are presented in Fig. 5. Amp8 encodes for 123 amino acids of the COOH-terminal part of bovine cathepsin B, and encompasses 1kb of nucleotides in the 3”untranslated region of the mRNA. The amino acid sequence encoded by Amp8 is the same as the primary structure recently determined for bovine cathepsin B (Meloun et al., 1988), except for Ala-218 which is replaced by Gly-218. The predicted carboxyl terminus is also longer than the mature protein by 3 aminoacid residues, and similar COOH-terminal extensions were identified from cDNA encoding cathepsin B in other species (Chan et al., 1986). Levels of Cathepsin B Transcripts in Myoblasts, Myotubes, and Fetal Muscle-Dot-blot hybridization studies carried out 0.8

,

t

*

lZ]

Muscle Fetal Myotube Myoblast

FIG. 4. Active-site titration by E-64 of cathepsin B in myoblasts, myotubes, and fetal muscle. Cathepsin B recovered from Mono-S FPLC was preincubated for 15 min with increasing concenFIG. 3. Specific activities of cathepsins B, L, and H in bo- trations of E-64, and residual activity against Z-Phe-Arg-NMec was vine myoblasts, myotubes, fibroblasts, and fetal muscle. Ca- assayed. The inset shows atypical titration experiment. Linear regresthepsinactivities were measuredinunfractionated lysosomalexsion coefficients were generally greater than 0.97. From the slope of tracts. Z-Arg-Arg-NMec and Arg-NMec were used as specific sub- the regression lines, we calculated that the absolutespecific activity strates for cathepsin B and cathepsin H, respectively. Cathepsin L of cathepsin B in these studies was 21.1 t 4.5 (8) milliunits/pg, 35.9 activity was assessed with Z-Phe-Arg-NMec as substrate and with Z- k 11.6 (14) milliunits/pg, and 24.1 t 1.6 (3) milliunits/pg (mean f Phe-Phe-CHN, (1 p ~ as) specific inhibitor. Cathepsin activities are S.D. ( n ) ) for myoblasts, myotubes, and fetal muscle,respectively. reported to the yield in lysosomes and to total protein content. One From the intercept with the base line, the operational molarity of unit ( U ) of activity is therelease of 1 pmol of aminomethylcoumarin cathepsin B under our assay conditions was estimated to be 40-100 (Mec) per min. Results are means t S.D. of four to nine separate nM for myoblasts, myotubes, or fetal muscle. The histogram shows experiments, and statistical significance levels (*p < 0.01; **p < 0.001) levels of active cathepsin B calculated from the intercept with the were determined by using Student’t test.B, cathepsin B; H, cathepsin base line and after correctionfor yield in lysosomes and cell protein L; 0, cathepsin H. content (means f S.D.). Myoblast Myotube Fibroblast

Foetal Muscle

14108


Expression Cathepsin of

FIG. 5. Identification of a cDNA clone encoding bovine cathepsin B. The restrictionendonucleasemapand sequencing strategy for Amp8 are shown at the top of the figure. Arrows indicate thestrandandextent of DNA sequenced. The open box denotes the regionencoding amino acids 135-256 of bovine cathepsin B (numbering accordsolid ing toMeloun et al. (1988)), and the bar the 3"untranslated region. The nucleotide sequence of the coding region of Amp8 is presented at the bottom of the figure. The predictedamino acid sequence is similar to that determined by amino acid sequencing of bovine cathepsin B (Meloun et al., 1988), except for a COOH-terminal extension of three additional amino acids (underlined). The active-site His residue is boxed.

I 0 0 bp

135 C l yT y rS e rP r oS e rT y rL y s

140

A GGC TAC AGC CCG TCC TAC

150 C l u A s p L y s H i s P h eC l y C y s S e rS e rT y rS e r AAA GAA GAC AAG CAT TTT GGA TGC ACT TCC T A C AGC

I60 V a l A l a A s n A s n C l u L y s G l u I l eM e t A l a Clu GTC GCC A A C AAC GAG A A G GAG ATC ATG GCA GAG ATC TAC

I70 I l eT y rL y s

A s n C l yP r o

Val A A A AAT GGC CCA GTC

I80 G l u C l y A l a Phe S e r V a l Tyr S e r Asp Phe Leu Leu Tyr Lys S e r G l y V a l Tyr GAG GGG GCC T T C TCT GTG TAC TCG GAC TTC CTG CTA TAC AAG TCT GGG GTG TAC 190 Gln His

Val

S e rG l y

Clu

I l eM e tG l yC l y

I S

200 Ala

I l eA r gI l eL e uG l yT r p

210 220 V a l G l y A s n S e rT r p Asn Thr Asp C l y V a l C l u A s n C l yT h rP r oT y rT r pL e u GGA G r G GAG A A C GGC ACC ccc TAC TGG C T G GTC GGC A A C rcc TGG A A C A C T GAT

230 240 T r p C l y Asp Asn G l y Phe Phe Lys I l e Leu Arg C l y G l n Asp H i s C y s C l y I l e TGG GGT GAC A A T GGC TTC TTC A A A ATC CTC A G A GGA CAG GAC CAC T G T GGA ATC

250 C l u S e r G l u I l e V a l A l a G l yM e P t r oC y T s hr t i i s G l n T y r Stop GAG TCG GAA ATC GTG GCT GGA ATG CCC TGC ACT CAT C A G TAC TAG

FIG.6. Quantitation of cathepsin B mRNA by solution hybridization to pGem8 riboprobe. The riboprobe synthesized on linearized pGem8 using SP6 RNA polymerase is 550 bases long. B mRNA Hybridizationtocathepsin protects 490-base a fragment from RNase digestion.Autoradiography was for 2 days (lanes a-i) or 7 days (lanes j and k). Lane a,riboprobe; lane b,protection by tRNA (20 p g ) ; lane c, DNA ladder (RNA migrates 5% slower than DNA of the same size (Sambrook et al., 1989)); lanes d-f, hybridization to solubilized myoblasts (0.7, 1.4, and 2.8 X lo6 nuclei, respectively); lanes g-i, hybridization to solubilized myotubes (0.8, 1.6, and 3.3 X 10"nuclei, respectively). T o compare cultured cells and fetal muscle, assays were performed with 2 pg of total RNA (lane J,myotubes; lane k, fetal muscle).

a

b

with bovine cDNA probe Amp8 did not indicate any significant difference between bovine myoblasts and myotubes, but hybridization signals were lower for bovine fetal muscle than for primary cultures (not shown). To more precisely determine the concentrations of cathepsin B mRNA in muscle cells, liquid hybridizations were performed with "P-riboprobes complementary to cathepsin B mRNA (Fig. 6). Hydrolysis with RNase Aof cRNA-mRNA hybrids protected a fragment 490 bases long in myoblasts, myotubes, and fetal muscle. These studies confirmed that there are similar amounts of cathepsin B mRNA in myoblasts and myotubes, but lower levels in fetal muscle. When theband absorbance of the protectedfragments was compared with that of known amounts of riboprobes, we estimated that there are 80-100 cathepsin transcripts/nucleus in myoblasts or myotubes, and 10-20 transcripts/nucleus in fetal muscle. Modifications of Cathepsin B Enzymatic Properties during

c

d

I

I

Myoblast-Myotube Differentiation-Comparison of cathepsin

B mRNA levels therefore confirmed our titration studies of active cathepsin B. On the one hand, differences in cathepsin B activity between fetal muscle and muscle cell cultures were related to different levels of cathepsin B mRNA, and to parallel changesin active cathepsin B concentrations. On the other hand, an increase in cathepsin B activity between confluent myoblasts and myotubes was not associated with corresponding modifications in cathepsin B mRNA levels or in active enzyme concentration. Another possibility is that the increase in total activity of cathepsin B during myoblastmyotube differentiation is due to modifications of enzymatic properties. To test thishypothesis, kinetic studies were performed with cathepsin B fractions recovered from Mono-S chromatography of lysosomal extracts from myoblasts, myotubes, and fetal muscle (Table I). These experiments revealed that myoblast-myotube differentiation was associated with a

Expression of Cathepsin B during Myoblast-Myotube TABLE I Kinetic constants for cathepsin B The kinetic constants were determined a t p H 5.5 and at 37 "C in the concentration range 10-60 p M Z-Arg-Arg-NMec. Results were obtained from three to five separate experiments and are presented with standard deviations. The concentration of cathepsin B for the evaluation of kcatwas determined by active-site titration with E-64. Values for cathepsin B purified from bovine liver were obtained from Deval et al. (1990).

Myoblast 127 Myotube Fetal muscle Cathepsin R

K,

kc,,

SM

s-1

f3 167 & 14 125 f 24 119 f 8

76 f 17 216 & 62 26 & 6 63 f 16

Kmlkm rnM".s"

598 1293 208 529

2 8 s -m

1 8 s -m

Differentiation

14109

Amp8 wasused as aprobe. These data indicate that the different catalytic propertiesof cathepsin B in myoblasts and myotubes are not related toa difference in size of cathepsin B mRNA. Affinity labeling of Cathepsin B with Z-Tyr(-'251)-Ala-CHN.L in Myoblasts and Myotubes-Even if myoblasts andmyotubes express the same cathepsin B mRNA, theactive enzymecould differbecause of differentpost-translational processing. There are onlya limited number of methods to determineM, values of active cathepsins. Comparison of cathepsin B purified from myoblasts and myotubes requires large scale cell culture of myoblasts andmyotubes, and alsoassumes that no preparation artifacts arise during purification. Immunoprecipitation of ["S]Met-labeled proteins from myoblasts and myotubes will not indicatewhich forms of enzyme are active. An alternative approachwas recently described by Crawford et al. (1988). They identified Z-Tyr(1)-Ala-CHN? as a selective inhibitor of cathepsins L and B that bindscovalently only to the active formsof these enzymes by alkylating thereactivesite cysteine residue. Furthermore, Mason et al. (1989) have shown that Z-Tyr(-'"1)-Ala-CHN2 does not label other proteins when incubatedwith KNIH 3T3cells or cell extracts. When we incubated Z-Arg-Arg-NMec active fractions recovered fromMono-SFPLCwith 1 p~ Z-Tyr(-'"1)-AlaCHN?, onemajor radiolabeled-band of M, 32,000 was detected by SDS-PAGE (Fig. 8). This corresponds to the M, of the uncleavedform of cathepsin Bpurifiedfrom bovine liver (Devalet al., 1990).Affinitylabeling of cathepsin B was abolished when samples were preincubated with E-64 (Fig. 8d) but not after pretreatment with Z-Phe-Phe-CHN2 (Fig. 8, b and c). A directcomparison betweenmyoblasts and myotubes of the M, size of active cathepsin B is presented in Fig. 8 (f andg). In both myoblasts and myotubes, cathepsin B recovered from Mono-S FPLC displayed M, 32,000. This wasalsothe majorradiolabeled bandwhen myoblasts or myotubes were cultured for3 h directly in thepresence of 0.5 p~ Z-Tyr(-'"II)-Ala-CHN2 (Fig. 8, h and i). In addition to demonstrating that affinity labeling with Z-Tyr(-'""I)-AlaCHN, can be used to determine theM, of active cathepsin B

h 94

Northern blot analysis of cathepsin B mRNA in bovine myoblasts and myotubes. Poly(A)+ mRNA (10 p a ) from FIG.7.

bovine myoblasts or myotubes was subjected to electrophoresis in a 1%agarose-formaldehyde gel, transferred to Hybond-N' membrane, and probed with ['"PIRNA complementary to bovine cathepsin B mRNA. Autoradiography with an intensifying screen was for 5 days at -70 "C. Migration of the 18 S and 28 S ribosomal RNA is indicated.

i

-

67 43

-

30

-

2-3-fold increase in catalytic efficiency (kcat&,). In addition t o reduced concentrations of active enzyme, cathepsin B in fetal muscledisplayeda lower catalytic efficiency than in 20 myoblasts. K, and kc,, of myoblast cathepsin B were also close to values obtained for cathepsinB purified from bovine (Deval et al., 1990) or other species (Baricos etal., 1988). FIG.8. Inhibitor-labeling of cathepsin B inbovine myoNorthern Blot Analysis of Cathepsin B mRNA in Myoblasts blasts and myotubes. a-e, cathepsin R recovered from Mono-S and Myotubes-Different enzymatic properties of cathepsin FPLC of myoblast lysosomal extract was activated andincubated for B from myoblasts and myotubescould result from the trans- 10 min a t 30 "Cwithout ( a and e ) , or with 0.5 p~ Z-Phe-Phe-CHN2 lation of different mRNAs. To test this hypothesis, Northern( b ) , 5 p~ Z-Phe-Phe-CHN, (e), or 1 p~ E-64 ( d ) , before affinity blots of poly(A)+ mRNA from myoblasts and myotubeswere labeling with Z-Tyr(-'"1)-Ala-CHN2 and SDS-PAGE (as described probed with the r2P]cRNA. As shown in Fig. 7, both myo- under "Experimental Procedures"). f and g, lysosomal extracts from myoblasts ( f ) or myotubes (g)were fractionated by Mono-S FPLC, blasts and myotubes revealed a major 2.6-kb transcript, and and cathepsin B fractions labeled with 1 p~ labeled inhibitor. h and a minor 3.2-kb band.Therelativeintensities of the two i, SDS, 12% PAGE of trichloroacetic acid precipitates of myoblasts messengers were the same in myoblasts myotubes. and Similar ( h )or myotubes ( i )directly cultured for 3 h with 0.5 p M Z-Tyr(-"'I)Northern blots were obtained when full-length radiolabeled Ala-CHN2.

14110

Expression of Cathepsin B during Myoblast-Myotube Differentiation

from a single culture flask of muscle cells, these results reveal that the different enzymatic properties of cathepsin B in , of the active myoblasts andmyotubes are not related to Mthe enzyme. Probably because of the low concentration, affinity labeling of active cathepsin L could only be performed with fractions recovered from Mono-S FPLC, and revealed radiolabeled bands of 31,29, and 26 kDa (not shown).

but not H, increase between confluent and fused cells. In contrast, cathepsins B, H, and L, but not D, increased during differentiation of rat L6 muscle cells (Kirschke et al., 1983). In addition to likelydifferencesbetween transformed cell lines and primary cultures, speciesspecificity may alsobe expected. Inthis regard, ourdata with fetal calf muscle support previous results stating that only very low levels of Isoelectrofocusing of Cathepsin B in Myoblasts and Myo- cathepsin H can be detected in bovine species (Bkchet et al., tubes-Isoelectrofocusing in thin layersof polyacrylamide gel 1986). T o circumvent the uncertainty that cysteine proteinases was then used to detect anymodifications in the cathepsinB isozyme population. The different cathepsin B isozymes were may originate from non-muscle cells, previous studies essenrevealedimmediately after focusing by blotting the gel on tially focused on established myogenic lines (Kirschke et al., paper prewetted with Z-Arg-Arg-NMec containing solution. 1983; Colella et al., 1986). In agreement with other laboratoNo signal was evident when lysosomal extracts were prein- ries (Buckingham et al.,1976; Gospodarowicz et al., 1976))a was prewet with 70-90% fusion index was obtained in bovine myotubes inthis cubated with lo-" M E-64, or when the paper a solution containing both Z-Arg-Arg-NMec and E-64. As study, with the remaining mononuclear cells corresponding fibroblasts. shown in Fig. 9, limited but reproducible differences in ca- to undifferentiated myoblasts or to contaminating Fetal calf fibroblasts were also studied for cysteine proteinase thepsin B Z-Arg-Arg-NMec hydrolyzing isozymes were eviactivities. The increase in cathepsin activities observed during dent on isoelectrofocusing. myoblast-myotube differentiation could in no way be attributed to fibroblast proliferation, as fibroblast primary cultures DISCUSSION revealed similar levels of cathepsin B activity to myoblasts. Fetal calf myoblast-myotube cell cultures have been exten- The catalytic propertiesof cathepsin B were also the same in sively characterized for synthesis of contractile proteins and fibroblasts andmyoblasts. of mRNAs encoding contractile proteins (Buckinghamet al., The present work further shows that levels of cysteine 1976; Daubas et al., 1981). Herein we show that bovine myo- proteinase activities are2 to 4 orders of magnitude greater in blast-myotube differentiation is also characterized by a 2-3cultured myoblasts (or fibroblasts) than in fetalmuscle from fold increase in cathepsin B and L lysosomal activities. Sim- which they derive. Histological studies reveal that at90 days ilar increases in cathepsin B- and L-specific activities were of age, fetal calf muscles already contain highly organized reported for myoblast-myotube differentiation in cell lines multinucleated myotubes and myofibers, in addition toundifsuch as rat L6 (Kirschkeet al., 1983) and mouse C2 (Colella ferentiated myoblasts.' Primary culturesdeveloped from fetal et al., 1986). The enzymatic propertiesdescribed in this study muscle result in the proliferationof these myoblasts andwill for Z-Arg-Arg-NMec-, Arg-NMec-, andZ-Phe-Phe-CHN,a t best reproducemyoblast-myotube differentiation.Later sensitive 2-Phe-Arg-NMechydrolyzing activities are also sim-stages of myogenesis, including the formation of secondary ilar to those characterizing cathepsin B, H, and L, respecmyotubes and myofibers are, however, far from being simutively, in other tissues species or (Dufour et al.,1987; Kirschke lated in vitro. Lower levels of cathepsin activities in fetal and Barrett, 1987; Deval et al., 1990). It is noteworthy that muscle than in cultured muscle cells could therefore mean bovine myoblast-myotube differentiationisnotassociated there is a drop in cathepsin activities during fetal muscle with parallel changesof all lysosomal cysteine proteinases,as development. Indeed, recent experiments in our laboratory cathepsin H remained at low levels of activity. This is in indicate alargedecrease of cathepsin-specific activities in agreement with the data reportedby Colella et al. (1986) for muscles during bovine fetal growth.:' High levels of cathepsins a mouse C2 muscle cell line in which cathepsin B, L, and D, in culturedcells are also in agreement with higher proteolysis rates in cultured muscle cells than in muscle in situ (Harper a bed et al., 1987). Modifications of specific activity of lysosomal cathepsins could result from many mechanisms encompassing alterations in cathepsin mRNA levels, mRNA translational rates, posttranslational processes, and/or interaction with endogenous inhibitors. Levels of mRNA encoding cathepsin B in bovine tissues could not be assessed with a rat cathepsin B cDNA probe (prCB3). Such species specificity is in agreement with previous data indicating that rat cathepsinB cDNA does not hybridize to chicken muscle cell RNA (Colella et al., 1988) or that cathepsin L mRNA from human cells is not detected 1988). In this with a mouse cDNA probe (Gal and Gottesman, report we therefore also describe the isolation and partial characterization of a cDNA probe (Xmp8) encoding bovine cathepsin B. The COOH-terminal amino acid sequence deduced from the Amp8 coding region is consistent with the FIG. 9. Isoelectrofocusing of cathepsin B isoforms. Lysoso- primary structure previously reported for bovine cathepsin B mal extracts from myoblasts (lanes a and b ) or myotubes (lanes c and (Meloun et al., 1988), except for one amino acid (Gly-218). d ) were submitted to isoelectrofocusing. Lanes a and c: 0.1 milliunits This difference could be due to sequence polymorphism. The of 2-Arg-Arg-NMec hydrolyzing activity.Lanes b and d, 0.2 milliunits of Z-Arg-Arg-NMec hydrolyzing activity. Isoelectrofocusing and detection of cathepsin B isoforms were performed as previously described (Deval et al., 1990). Standards forisoelectrofocusingwere carbonic anhydrase (PI = 5.85), @-lactoglobulin A (PI = 5.13), and t.r-ypsin inhibitor (PI = 4.55).

J. Robelin, A. Lacourt, D. BBchet, M. Ferrara, Y. Briand, and Y. Geay, unpublished data. "D. M. Bkchet, M. J. Ferrara, C . D. Deval, and J. Robelin, manuscript in preparation.

Expression of Cathepsin B during Myoblast-Myotube Differentiation Amp8 sequence further reveals that the predicted COOH terminus of bovine cathepsin B is extended by three amino acids. The removal of this COOH-terminal tripeptide probably occurs during post-translational processing of cathepsin B. COOH-terminal extensions were also identified from the cDNA encoding rat, mouse, and human cathepsin B (San Segundo et al., 1985; Chan et al., 1986).Dot-blot hybridization, as well as a highly sensitive liquid hybridization assay with a ''2P-riboprobe complementary to cathepsinB mRNAwere used to specify levels of mRNA encoding cathepsin B. Both techniques confirm that myoblasts and myotubes contain similar amounts of mRNA encoding cathepsin B, but that this transcriptis poorly represented inmuscle from 5-monthold fetuses. It is interesting that, in every cases, levels of cathepsin B mRNA paralleled the concentrations of active cathepsin B estimated by active-site titrationwith E-64. This does not support an important translational control of cathepsin B expression in this system. In fetal muscle, it appears that low amounts of cathepsin B mRNA may probably account for low concentrations of active cathepsin B. In contrast, during myoblast-myotube differentiation, despite an increase in specific activity, there were constant levels of cathepsin B mRNA and constant concentrations of active cathepsin B. These observations do not support transcriptional activation of cathepsin B expression during bovine myoblast-myotube differentiation. This is consistent with data obtained with L6 or C2 lines (Colella et al., 1986) where no change in amounts of cathepsin B mRNA were detected between confluent and fused cells. In fact, we further show that therise in cathepsin B-specific activity during fetal calf myoblast-myotube differentiation can be attributed to modifications of its enzymatic properties. Several mechanisms could explain such changes in the enzymatic properties of cathepsin B. One explanation could be that the different properties are due to different cathepsin B molecules being expressed in myotubes than in myoblasts. It has been shown that, like other lysosomal enzymes, cathepsin B is translated as alarge M, precursor containing signal- and pro-sequences. In the endoplasmic reticulum and Golgi compartments, this large M , precursor undergoes limited proteolysis, glycosylation, anda series of modifications of the glycosyl units. Any of these post-translational modifications could be different during myoblast-myotube differentiation, result in changes in molecular size or charge, and eventually induce different catalytic properties. Northern blot analyses demonstrated a cathepsin B mRNA length of2.6 kb in bovine musclecells. The size of this transcript is in agreement with cathepsin B mRNA described in muscle cell lines (Colella et al., 1986) or in other tissues (San Segundo et al., 1985). Another faint 3.2-kb transcript was also evident in myoblasts and myotubes, and further studiesare required to determinewhether it representsa precursor form of cathepsin B mRNA, or results from alternative splicing or use of an alternative poly(A) site. However, we could not detect any difference in size or proportion of 2.613.2 transcripts between myoblast and myotube that could account for modifications of cathepsin B. Furthermore, using affinity labeling with Z-Tyr(-lz51)-Ala-CHN2, myoblasts and myotubes also revealed the same active form of M , 32,000. From these data, we conclude that both myoblasts and myotubes express qualitatively and quantitatively similar cathepsin B mRNAs, and that the translation of these transcripts ultimately results in similar concentrations of lysosomal active cathepsin B exhibiting the same Mr. Takahashi et al. (1984,1986) have nevertheless reported the existence in porcine spleen of two cathepsin B isozymes (CB-I and CB-11)

14111

that exhibit different catalytic properties, distinct Asn-linked carbohydrates, and an amino acid replacement. We thus cannot exclude the possibility that cathepsin B in myoblasts and myotubes may differ by some amino acid substitution(s). RNase protection assays with 32PcRNAcovering the complete translated region will be required to detect single-base substitutions in cathepsin B mRNA. Only minor changes incathepsinB isozymeswere evidenced between myoblasts and myotubes by isoelectrofocusing. In a recent study, different populations of cathepsin B isozymes were purified from bovine liver and they revealed similar enzymatic properties (Deval et al., 1990). Despite similar catalytic properties(see Table I), cathepsin Bpurified from bovine liver (Deval et al., 1990) and cathepsin B from bovine myoblasts (Fig. 7), also exhibit different profiles of isozymes. Smith et al. (1989) further report thatglycosylation of cathepsin L does not account for differences in enzymatic properties. Whether post-translational modifications, other than those found in myoblasts or bovine liver, occur in myotubes and alter the catalytic properties remains to be demonstrated. Another explanation may be that cathepsin B is activated (in myotubes) or inhibited(in myoblasts) bysome compound(s) present in crude lysosomal extracts. The presence of intralysosomal inhibitors of cathepsin B and L has been reported in rabbit liver (Pontremoli et al., 1983). Katunuma (1989) has also speculated that cystatins may be secreted and endocytosed to regulate lysosomal cathepsin activities. However, because cathepsin B in crude lysosomal extracts from myoblasts exhibits enzymatic properties similar to purified cathepsin B(Deval et al., 1990;Baricos et al., 1988),inhibition may not be predominant in myoblast extracts. To our knowledge, no activator of cathepsin B activity has yet been described in lysosomal extracts or other cellular compartments. But thisexplanation is also one of the possibilities that could account for the results obtained in this study. Acknowledgments-We gratefully thank Dr. D. Steiner and Dr. S. Chan for providing prCB3, and Dr. E. Shaw for providing Z-PhePhe-CHNz andZ-Tyr-Ala-CHNz. REFERENCES Bando, Y., Kominami, E., and Katunuma, N. (1986) J . Biochem. (Tokyo) 100, 35-42 Baricos, W. H., Zhou, Y., Mason, R. W., and Barrett, A. J. (1988) Biochem. J. 252,301-304 Barrett, A. J., and Kirschke, H. (1981) Methods Enzymol. (Lorand, L., ed) 80,535-561 Barrett, A. J., Kembhavi, A. A., Brown, M. A., Kirschke, H., Knight, C. G., Tamai, M., and Hanada, K. (1982) Biochem. J. 201, 189198

BLchet, D., Obled, A., and Deval, C. (1986) Biosci. Rep. 6,991-997 BCchet, D. M., Listrat, A., Deval, C., Ferrara, M., and Quirke, J. F. (1990) A m . J. Physiol. 259, E822-E827 Bige, L., Ouali, A., and Valin, C. (1985) Biochim. Biophys. Acta 843, 269-275

Bird, J. W. C., Roisen, F. J., Yorke, G., Lee, J. A., McElligott, M. A., Triemer, D. F., and St. John, A. (1981) J. Histochem. Cytochem. 29,431-439

Bradford, M. (1976) Anal. Biochem. 72, 248-254 Bromme, D., Steinert, A., Friebe, S., Fittkau, S., Wiederanders, B., and Kirschke, H. (1989) Biochem. J. 264,475-481 Buckingham, M. E., Cohen, A., and Gros, F. (1976) J. Mol. Biol. 103, 611-626

Cathala, G. Savouret, J. F., Mendez, B., West, B. L., Karin, M., Martial, J. A., and Baxter, J. D. (1983) D N A 2, 329-335 Cavaillis, V., Garcia, M., and Rochefort, H. (1989) Mol. Endocrinol. 3,552-558

Chan, S. J., San Segundo, B., McCormick, M. B., and Steiner, D. F. (1986) Proc. Natl. Acad. Sci. U. S. A . 83,7721-7725

14112

Expression Cathepsin of


Colella, R., Roisen, F. J., and Bird, J. W. (1986) Biomed. Biochim. Acta 45, 1413-1419 Conner, G. E., Blodel, G., and Erickson, A. H. (1987)in Lysosomes: Their Role in Protein Breakdown (Glaumann, H., and Ballard, F. J., eds) pp. 151-161, Academic Press, New York Crawford, C., Mason, R.W., Wikstrom,P.,andShaw,E. (1988) Biochem. J . 253, 751-758 Daubas, P., Caput, D., Buckingham, M., and Gros, F. (1981) Deu. Biol. 84, 133-143 Deval, C., Bechet, D., Obled, A., and Ferrara,M. (1990) Biochem. Cell Biol. 68,822-826 Doherty, P., Hua, L., Liau, G., Gal, S., Graham, D. E., Sobel, M., and Gottesman, M. M. (1985) Mol. Cell Biol. 5, 466-473 Dong, J., Prence, E. M., and Sahagian, G. G. (1989) J. Biol. Chem. 264,7377-7383 Dufour, E., Obled, A, Valin, C., Bechet, D., Ribadeau-Dumas, D., and Huet, J. C. (1987) Biochemistry 26,5689-5695 Faust, P. L., Kornfeld, S., and Chirgwin, J. M. (1985) Proc. Natl. Acad. Sci. U. S. A. 82, 4910-4914 Ferrara, M., Wojcik, F., Rhaissi, H., Mordier, S., Roux, M. P., and Bechet, D. (1990) FEBS Lett. 273, 195-199 Firestein, G. S., Gardner, S. M., and Roeder, W. D. (1987) Anal. Biochem. 167, 381-386 Fraker, P. J., and Speck, J. R., Jr. (1978) Biochern. Biophys. Res. Commun. 80,849-857 Gal, S., and Gottesman, M. M. (1986) Biochem. Biophys. Res. Commun. 139,156-162 Gal, S., and Gottesman, M. M. (1988) Biochem. J. 253,303-306 Gospodarowicz, D., Weseman, J., Moran, J. S., and Lindstrom, J. (1976) J. Cell Biol. 70, 395-405 Harper, J. M., Soar, J . B., and Buttery, P. J. (1987) J. Endocrirzol. 112,87-96 Hashida, S., Towatari, T., Kominami, E., and Katunuma, N. (1980) J . Biochem. (Tokyo)88, 1805-1811 Ishido, K., Imajoh, S., Emori, Y., Ohno, S., Kawasaki, H., Minami, Y., Kominami, E., Katunuma, N., and Suzuki, K. (1987) FEBS Lett. 226, 33-37 (1989) Ishido, K., Kominami, E., Suzuki, K., andKatunuma,N. FEBS Lett. 259,71-74 Katunuma, N. (1989) in Intracellular Proteolysis (Katunuma, N., and Kominami, E., eds) pp. 3-23, Japan Scientific Societies Press Kirschke, H., and Shaw, E. (1981) Biochem. Biophys. Res. Commun. 101,454-458 Kirschke, H. and Barrett, A. J. (1987) in Lysosomes: Their Role in Protein Breakdown (Glaumann, H., and Ballard, F. J.,eds) pp. 193-238, Academic Press, New York Kirschke, H., Wood, L., Roisen, F. J., and Bird, W. J. (1983) Biochem. J. 214,871-877 Kirschke, H., Locnikar, P., and Turk, V. (1984) FEBS Lett. 174, 123-127 Kirschke, H., Schmidt, I., and Wiederanders, B. (1986) Biochem. J. 240,455-459 Kominami, E., Tsukahara, T., Bando, Y., and Katunuma, N. (1985) J. Biochem. (Tokyo)98, 87-93 E. F., andSambrook, J. (1982) Molecular Maniatis,T.,Fritsch,

Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY Mason, R. W., Wilcox, D., Wikstrom, P., and Shaw,E. N. (1989) Biochem. J. 257, 125-129 Meloun, B., Baudys, M., Pohl, J., Pavlik, M., and Kostka, V. (1988) J. Biol. Chem. 263,9087-9093 Nishimura, Y., Kawabata, T., and Kato, K. (1988a) Arch. Biochem. Biophys. 261,64-71 Nishimura, Y., Amano, J., Sato, H., Tsuji, H., and Kato, K. (1988b) Arch. Biochem. Biophys. 262,159-170 Obled, A., Ouali, A,, and Valin, C. (1984) Biochimie 66, 609-616 Ouali, A., Bige, L., Obled, A., Lacourt, A,, and Valin, C. (1986) in Cysteine Proteinases and their Inhibitors (Turk, V., ed) pp. 545554, Walter de Gruyter, Berlin Pontremoli, S., Melloni,E., Salamino, F., Sparatore, B., Michetti, M., and Horecker, B. L. (1983) Proc. Natl. Acad. Sci. U. S. A. 80, 1261-1264 Portnoy, D. A., Erickson, A. H., Kochan, J., Ravetch, J. V., and Unkeless, J. C. (1986) J. Biol. Chem. 261, 14697-14703 Ritonja, A., Popovic, T., Turk, V., Wiedenmann, K., and Machleidt, W. (1985) FEBS Lett. 181, 169-172 Ritonja, A., Popovic, T., Kotnik, M., Machleidt, W., and Turk, V. (1988) FEBS Lett. 228, 341-345 Rochefort, H., Caponi, F., Garcia,M., Cavaillis, V., Freiss, G., Chambon, M., Morisset, M., and Vignon, F. (1987) J. Cell. Biochem. 35, 17-29 Sambrook,J.,Fritsch, E. F., andManiatis, T. (1989) Molecular Cloning:A Laboratory Manual, 2nd ed., p. 7.76, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY Sanger, F., Nicklen, S., and Covlson, A.R. (1977) Proc. Natl. Acad. Sci. U. S. A. 74,5463-5467 San Sermndo, B., Chan, S. J., and Steiner. D. F. (1985) Proc. Natl. Acad.-Sci. U . S . A. 82, 2320-2324 San Segundo, B., Chan, S. J., and Steiner, D. F. (1986) FEBS Lett. 201,251-256 Shewale. J. G.. and Tang. -. J. (1984) . . Proc. Natl. Acad. Sci. U. S. A. 81, 3703-3707 Sloane, B. F., Rozhin J., Hatfield, J. S., Crissman, J. D., and Honn, K. (1987) Exp. Cell Biol. 55, 209-224 Smith, S. M., Kane, S. E., Gal, S., Mason, R. W., and Gottesman, M. M. (1989) Biochem. J. 262,931-938 Stauber, W. T., and Ong, S. H. (1981) J. Histochern. Cytochem. 29, 866-869 Stauber, W. T., Fritz, V., Dahlmann, B., Gauthier, F., Kirschke, H., and Ulrich, R. (1985) Histochem. J. 17, 787-796 Takahashi, T., Schmidt, P. G., and Tang, J. (1984) J . Biol. Chem. 259,6059-6062 Takahashi, T, Yonezawa, S., Dehdarani, A. H., and Tang, J. (1986) J. Biol. Chem. 261,9368-9374 Takio, K., Towatari, T., Katunuma, N., Teller, D. C., and Titani, K. (1983) Proc. Natl. Acad. Sci. U. S. A. 80, 3666-3670 Taylor, M. A,, Almond, R. E., and Etherington, D. J. (1987) Histochemistry 86,379-383 Towatari, T., and Katunuma, N.(1988) FEBS Lett. 236, 57-61 Troen, B. R., Gal, S., and Gottesman,M. M. (1987) Biochem. J. 246, 731-735 Wilkinson, G. N. (1961) Biochem. J. 80, 324-332