A high-molecular-weight cysteine endopeptidase from rat skeletal muscle

Biochimica et Biophysica Acta, 742 (1983) 399-408 Elsevier Biomedical Press

399

BBA 31478

A HIGH-MOLECULAR-WEIGHT CYSTEINE ENDOPEPTIDASE FROM RAT SKELETAL MUSCLE FIRHAAD ISMAIL and WIELAND GEVERS

Department of Medical Biochemistry, University of Cape Town, Medical School, Observatory, 7925 Cape Town (Republic of South Africa) (Received August 30th, 1982)

Key words: Cysteine endopeptidase; Proteinase," (Muscle)

A cytosolic enzyme of high molecular weight (about 500000), which attacks native or denatured proteins (inter alia, casein, globin and hexokinase) was purified about 1000-fold from mixed rat skeletal muscles, including muscles freed of mast cells by prior treatment of the animals with the degranulator, compound 4 8 / 8 0 . Peptides of varying size were generated from radioactively labelled globin, but no free amino acids were formed; free tyrosine was also not released from azocasein. The pH optimum was 7.5 and the presence of an essential cysteine group was suggested because dithiothreitol (1 mM) stimulated the activity and N-ethylmaleimide (5 mM) and p-chloromercuriphenylsulphonic acid (1 mM) were inhibitors. The activity was markedly inhibited by Zn2 ÷ but not by ieupeptin, chymestatin or pepstatin. The enzyme was stabilized by ATP, at concentrations as low as 0.1 mM, against inactivation at 42°C. The endopeptidase was clearly separated on gel chromatography from another large protease, also sensitive to Zn2+ , but with marked aminopeptidase activity and the properties of hydrolase H. The activity levels of the protease, assayed after chromatography on Sepharose 6B of high-speed supernatant fractions, did not vary significantly in skeletal muscle samples which were derived from denervated, starved, diabetic or hyperthyroid animals, in all of which the abnormal physiological states expressed themselves as enhanced rates of tyrosine released by incubated soleus and extensor digitorum Iongus muscles. Nevertheless, the enzyme described here may he part of an ATP-dependent, multi-component proteolytic system similar to that already known to he present in reticulocytes.

Introduction The description of a high-molecular-weight proteinase, apparently involved in a multi-component, ATP-dependent protein degradation pathway in reticulocytes [1-4], has aroused interest in the possible general importance of similar enzymes in other tissues. ATP-activated or -stabilized proteinases have in fact been partially purified from rat liver [5,6]. Skeletal muscles contribute significantly to whole-body protein turnover [7]. The only enzyme Abbreviation: EGTA, ethylene glycol bis(/~-aminoethyl ether)N,N'-tetraacetic acid. 0167-4838/83/0000-0000/$03.00 © 1983 Elsevier Biomedical Press

of high molecular weight (340000) which has been fully described in skeletal muscles is the zinc-sensitive aminoendopeptidase (hydrolase H) isolated from rabbit tissue by Okitani et al. [8]; this enzyme apparently does not interact with ATP. Two brief reports of successful extractions, from human and rat muscles, of high-molecular-weight proteinases possibly related to the reticulocyte and liver enzymes mentioned above, have recently appeared; one of these activities was inhibited by ATP [9], while the other was activated by ATP [10], in a manner apparently involving a heat-stable 'APF-I' fraction similar to ubiquitin [11]. Since a large number of other non-lysosomal, soluble or particulate proteinases have been char-

400 acterized in skeletal muscle (for review, see Ref. 12), a principal problem in the field remains the definition of enzymes participating in what may well be a variety of pathways of endogenous protein degradation, and the correlation of the activities of such entities with overall flux rates in intact cell systems, as controlled by exogenous hormones or inhibitors. We have accordingly studied a soluble, cysteine proteinase of high molecular weight (500000), which is demonstrably different from hydrolase H and which has been purified by an apparent factor of 1000, using column chromatographic procedures. The enzyme is an endopeptidase and is stabilized by ATP against thermal denaturation. It is very sensitive to inhibition by Zn 2÷. The activity of the partially purified endopeptidase was measured in skeletal muscles derived from denervated, starved, hyperthyroid and diabetic animals, and also from rats in whom mast cell degranulation had been produced by treatment with compound 48/80; in each case, a comparison was made between in situ proteolytic rates (of isolated, intact soleus and extensor digitorum longus muscles) and the activity of the high-molecular-weight protease. Materials and Methods Labelled protein substrates [14C]Methylglobin (2.5.106 dpm/mg) was prepared by labelling bovine haemoglobin [13], followed by haem extraction [14]. Fractionation of skeleta ! muscle extracts Long-Evans hooded rats (250-350 g) of both sexes, inbred in this laboratory and maintained on standard rat chow, were anaesthetized with intraperitoneal injections of pentobarbitone and killed by decapitation plus exsanguination. Muscles from both hind limbs were rapidly dissected out, freed of fat and connective tissue, weighed, and cut into fine pieces, which were chilled on ice. A homogenate (1:4, w/v, in 20 mM sodium phosphate buffer, pH 7.1, also containing 0.5 mM dithiothreitol) was prepared from the tissue fragments with an Ultraturrax homogenizer, operated at setting 6, using 2 x 30-s bursts and with continuous cooling in ice.

The homogenate was initially centrifuged in the cold at 8 0 0 × g for 15 min, and the resulting supernatant centrifuged twice further, first at 20000 x g for 15 min, and then at 100000 x g for 60 min in a Beckman SW 36 rotor. This left a particle- and membrane-free, high-speed supernatant fraction (fraction 1), which was dialysed for 18-24 h against 250 vol. of the ice-cold homogenized buffer solution. The dialysed fraction 1 was then kept at 70°C for 10 min after the addition of ATP (final concentration, 2 mM). Denatured protein was pelleted by centrifugation at 20000 x g for 20 min. The supernatant fraction (42 ml) was applied to a DEAE-cellulose (Whatman DE 54) column (1.4 x 20 cm), previously equilibrated with the homogenization buffer. The adsorbed protein was eluted with a linear KCI gradient (0-0.5 M), and fractions (2 ml) were collected for assays of protease activity (see below), and absorbance measurements at 280 nm. Sepharose 6B (or 4B) columns (60 x 0.9cm) were equilibrated in the cold with 50 mM Tris-HC1/0.5 mM dithiothreitol, pH 7.5, before the application of various samples at 1-5% of bed volumes. Fractions of 1 ml were collected at a flow rate of 2 ml/h. The columns were calibrated with known marker proteins. Protease assays Protease activity was routinely measured by the hydrolysis o f [14C]methylglobin. Each assay contained 100 mM Tris-HC1/5 mM MgC12/0.5 mM dithiothreitol, pH 7.5, together with the radioactive substrate (1.5 • l 0 4 dpm), all in a final volume of 250 #l. Incubations were performed at 37°C for 60 min, after which 200/~1 10% ice-cold trichloroacetic acid containing 5 mg/ml carrier bovine serum albumin was added to each to stop the reactions. Aliquots (200 #1) of the supernatant fractions obtained by centrifugation were mixed with 5 ml Instagel (Packard Instrument Co., Downer's Grove, IL, U.S.A.) and counted in a Beckman LS 9000 Scintillation Counter. Protease activity was expressed as dpm released in acidsoluble form per h, after subtraction of values for control reactions not containing added enzyme. Acid-soluble products of proteolysis were examined by chromatography of incubated assay mixtures on a Sephadex G25 (30 × 0.9 cm) col-

401

umn, followed by radioactivity analysis as described above. The rates of enzymatic hydrolysis of L-leucine2-naphthylamine (Leu-N-Nap) and a-N-benzoylL-arginine 2-naphthylamine (Bz-Arg-N-Nap) were measured by a modified method [15]. With Leu-NNap as substrate, each assay mixture consisted of 50 mM Tris-HC1, 1 mM EDTA, 0.5 mM substrate and sample in a final volume of 0.5 ml at pH 7.5. After incubation for 1 h at 37°C, 2 ml of ice-cold ethanol was added to each tube and the liberated naphthylamine measured fluorimetrically (excitation, 339 nm; emission, 403 nm). A standard curve using pure naphthylamine was obtained and the enzyme activity accordingly expressed as nmol naphthylamine released/h. In the case of assay mixtures containing Bz-Arg-N-Nap as substrate, the same procedure was followed but in these instances the substrate concentration was 2 mM. Protein determinations were performed by the method of Lowry et al. [16] as modified by Hartree [ 17] with bovine serum albumin as standard. Mast cell degranulation Prior treatment of the rats with compound 48/80 effectively removed histologically detectable mast cells from various tissues (including muscles) of these rats [18]. Rats weighing 200 g received twice-daily, intraperitoneal injections of compound 48/80 in the following manner: day 1, 100 itg/100 g; day 2, 200 /~g/100 g; day 3, 300 #g/100 g; day 4, 400 /xg/100 g; and day 5, 500/~g/100 g. The rats were killed on day 6. Physiological manipulations to enhance protein degradation in rat skeletal musics Hyperthyroidism. Rats received catabolic doses of L-triiodothyronine [19] (25 ktg/100g body mass), injected daily intraperitoneally over 14-18 days. Denervation. Unilateral sciatic nerve denervations were performed with the contralateral limbs serving as controls [20]. Diabetes mellitus. Diabetes was induced with alloxan (16 m g / 1 0 0 g body mass) injected intraperitoneally as a single dose, and the rats were killed on day 4 [21]. Starvation. This was carried out for a 48 h period.

Protein turnover in intact muscles Soleus and extensor digitorum longus muscles were dissected, weighed on a torsion balance and then incubated at 37°C in Krebs-Ringer 'bicarbonate buffer, under 95% 02/5% CO 2, for 2 h after a 30 min preincubation period [22]. Rates of protein breakdown were determined by measuring the net release of tyrosine from tissue protein with a fluorimetric assay [23]. Cycloheximide at 0.5 mM was added to the incubation medium to block re-utilization of tyrosine from protein synthesis [221. Materials [ 14C]Methylformaldehyde (16.7 mCi/nmol) was obtained from Amersham, U.K. Chymostatin, leupeptin and pepstatin were purchased from the Peptide Institute, Protein Research Foundation, Japan. Dithiothreitol, ATP, creatine kinase, creatine phosphate, c o m p o u n d 4 8 / 8 0 , L-triiodothyronine, L-Leu-N-Nap, Bz-Arg-N-Nap and naphthylamine were obtained from Sigma Chemical Company, U.S.A., and all other chemicals used were analytical reagent grade. Results

Purification of protease The dialysed high-speed supernatant (fraction

A......-.-°_'°°° -1600

-o,, ~

•

~

/"/'

~ooo

.0.2

/",. ~o

7~-

"1200

3'0

", ,b

5b

o,

°.°°

60

Fractions

Fig. l. DEAE-cellulose chromatography of a soluble muscle fraction. Fraction 1 obtained from 8 g of homogenized muscle was dialysed and subjected to heat-treatment (see Materials and Methods). It w a s t h e n chromatographed on a DEAE-cellulose column as described, with a linear KCI gradient (. .). Fraction' volumes were 2 ml and the assays of protease activity (A) and absorbance measurements ( O ) w e r e conducted as also described in Materials and Methods.

402 TABLE I PURIFICATION OF HIGH-MOLECULAR-WEIGHT ENDOPEPTIDASE All procedures and assays were carried out as described in Materials and Methods, using 8 g of rat hindleg muscle as starting material. Fraction

Protein (mg)

Total activity (dpm)

Spec. act. (dpm/mg)

Apparent purification (fold)

Yield (~)

Dialysed high-speed supernatant (fraction 1) Heat-treated material Activity peak on DEAE-cellulose Activity peak on Sepharose 6B

340 106 1.6 0.5

200000 325 000 320 000 308000

588 3 066 200 000 616000

1 5 340 1048

-

1) p r e p a r e d from the m i x e d muscles of r a t h i n d legs c o n t a i n e d p r o t e a s e activity d e t e c t e d with radioactive m e t h y l g l o b i n as s u b s t r a t e ( T a b l e I). This activity n o t o n l y survived t r e a t m e n t at 70°C, p r o v i d e d A T P (2 r a M ) was present, b u t was actually increased. T h e m a t e r i a l was a p p l i e d to an a n i o n exchange c o l u m n (DEAE-cellulose); all the redp i g m e n t e d p r o t e i n ( p r e s u m a b l y residual m y o g l o bin) was washed off the c o l u m n a n d c o n t a i n e d no p r o t e a s e activity. T h e virtually colourless p r o t e i n r e m a i n e d a d s o r b e d o n t o the column, a n d was eluted d u r i n g the a p p l i c a t i o n o f a salt gradient, so that all the p r o t e a s e activity a p p e a r e d b e t w e e n KC1 c o n c e n t r a t i o n s of 0.20 a n d 0.30 M (Fig. 1).

100 98 95

This activity p e a k (representing a b o u t 2% of the total p r o t e i n a p p l i e d to the c o l u m n ) was p o o l e d a n d the p r o t e i n p r e c i p i t a t e d overnight at 90% s a t u r a t i o n with a m m o n i u m sulphate. T h e resulting pellet was taken up in 2 ml 50 m M T r i s - H C l c o n t a i n i n g 0.5 m M dithiothreitol, p H 7.5, a n d a p p l i e d to a Sepharose 6B (or 4B) column. The activity p e a k eluted with a Kay indicative of an a p p r o x i m a t e m o l e c u l a r weight o f 500000 (Fig. 2). T h e same Kav was o b t a i n e d when the c o l u m n was r u n at high ionic strength (0.6 M KC1) or in the presence o f 0.5% T r i t o n X-100, a n d the elution

80-

% _

/

lo8

0.400 Enzyme

Lo,, •

. . . .

inhibition

/

.0,300

/ /

5.0 5,5 6,0 Log Molecular weight

"~

300

' Kay

',:

Hydr H

40-

•

.0,200 1o00 o o~

'

..2

lO

20

•=x•e.

30 Fraction

40

t ~ .=

-O.100

0

II

4O

50

50

Fractions

Fig. 2. Scpharose 6B chromatography: Separation of aminopeptidase from endopeptidase. A pooled and concentrated fraction containing all the protease activity from the DEAE-cellulose column described in Fig. 1 was chromatographed on a Sepharose 6B column as described in Materials and Methods. Fractions of 1 ml wer.e collected and aliquots (100 # 1) were assayed separately in each case with [taC]methylglobin and Leu-N-Nap as substrates. [14C]methylglobinase activity, =; leucine-naphthylaminase activity, e; absorbanc¢ at 280 rum, O. Inset: Calibration of the column with the marker proteins ferritin, catalase and haemoglobin, in order of decreasing size (El) (Hydr. H, hydrolase H). Fig. 3. Apparent absence of a specific inhibitor of the protease in muscle extracts. Undialysed fraction 1 from 1 g of homogenized leg muscle was chromatographed on Sepharose 6B (0.9 × 60 cm), previously equilibrated with 50 mM Tris-HCl/0.5 mM dithiothreitol (pH 7.5), as described in Fig. 2. The protease peak was identified and pooled. Samples of this enzyme were assayed in the presence and absence of 100-/~1aliquots of the whole range of other column fractions, and the percentage inhibition (if any) (B) was calculated in each case. Absorbance at 280 nm, O.

403 behaviour was also independent of the presence or absence of dithiothreitol. These findings suggest that the enzyme is genuinely of high molecular weight. (The subunit structure is currently being investigated.) The final apparent purification was 1000-fold over the dialysed fraction I, and the yield through the two column steps was 95% (Table I). It should be mentioned that the radioactive assay depends on competing unlabelled protein substrates and that these values can only serve as approximations.

Distinction of protease from aminoendopeptidase Fraction 1 catalysed significant hydrolysis of Leu-N-Nap at pH 7.5. This aminopeptidase activity was also adsorbed to the DEAE-cellulose column and eluted at 0.20 M KC1 in a peak which overlapped extensively with the [laC]methylglobinase activity (data not shown). Measurement of aminopeptidase activity in each of the fractions from the Sepharose 6B column showed that a well-defined, separate peak of activity, of approximate molecular weight 320 000, was responsible for the degradation of Leu-N-Nap (Fig. 2). This latter enzyme was partially inhibited by iodoacetate (10 mM) and by leupeptin (0.5 mM) and markedly inhibited by zinc ions (results not shown). The optimal pH for this activity was 7.5 and the enzyme hydrolysed Bz-Arg-N-Nap less well than Leu-N-Nap. This entity was thus almost certainly the aminoendopeptidase, hydrolase H, recently described in rabbit muscle extracts [8]. It is of interest that when the heat-step was omitted from the purification scheme, t h e Sepharose 6B (or 4B) protease activity profile revealed two peaks; one in the 500000 molecular weight range (which Was also rountinely observed when the heat-step was included) and a second peak of activity in the excluded volume (Vo). The latter activity was pepstatin-inhibitable, and had a pH optimum around 5.0; it presumably represented an aggregated protease(s) of lysosomal origin, probably cathepsin D (data not shown). Treatment of this peak with Triton X-100 failed to produce a shift in its position of elution, indicating that this was not a membrane-associated protease activity.

Absence of endogenous inhibitors In order to detect the presence of proteins in

the crude extracts which could be physiologically important inhibitors of the high-molecular-weight proteinase, we applied a fraction 1 preparation which had not been subjected to heat treatment to a Sepharose 6B column; the peak with high protease activity was pooled. This enzyme was then assayed in the presence and absence of 100-/~1 aliquots of all the other inactive fractions. Decreases in protease activity correlated generally with the total amounts of protein added in other fractions, and there was no indication of a well-defined peak of inhibition (Fig. 3). (Precise superimposition of the protein and 'inhibition' curves was not possible in areas of activity overlap due to the protease peak.) The results were thus in keeping with the idea that native proteins present in the extracts were all more or less effective substrates for the protease, and that no particularly effective inhibitor was present.

Effects of A TP Enzyme activities throughout the purification procedure were measured in the presence or absence of 2 mM ATP and an ATP-regenerating system consisting of creatine phosphate and creatine kinase. Both the DEAE activity peak and the Sepharose activity peak showed a variable 'stimulation' in the presence of the ATP system. This inconsistent effect was examined further by preincubating the purified enzyme with and without the ATP system for 60 min at 37 and 42°C, in the

I

I

A

B

&--

Activity remaining 50-

3~ Time (min)

6~

35

6~

Time (min)

Fig. 4. Effect of ATP on thermal stability of the protease. Samples of enzymes purified by column chromatography (Fig. 2) were pre-incubated at 37°C (A) and 42°C (B) for 30 and 60 min, in the presence (A) and absence (zx)of ATP. After this, the activityof the remainingenzymewas assayed,in both cases with ATP being present, at 37°C as describedin Materials and Methods.

404 TABLE II EFFECTS ON ENDOPEPTIDASE OF VARIOUS INHIBITORS AND ACTIVATORS An enzyme fraction, purified as described in Fig. 2, was incubated at 37°C for l h in the presence and absence of various inhibitors and/or activators. The assays with radioactive globin were carried out as described in Materials and Methods. Agent

% Inhibition

Agent

% Activation

ZnCI2 p-Chloromercuriphenyisulphonic acid

(0.1 mM)

80

CaC12

(1 mM)

(0.1 mM)

70

Dithiothreitol Cysteine

(0.1 mM) ( 1 mM)

N-ethylmaleimide

(1 mM)

60 ATP

(0.1-1.0 mM)

5-25

lodoacetate

(5 mM)

0 Triton X-100

1%

0

Chymostatin

(40 #g/ml)

0

Leupeptin

(40 txg/ml)

0

Pepstatin EDTA EGTA Bovine serum albumin Casein/azocasein

(40 p.g/ml) (5 mM) (5 mM)

0 0 0

(250- 1000 #g/ml) (250-1000#g/ml)

0 70 40

Triton X-100+ATP (1 mM)

25

lmidazole

25

(25 mM)

0 75-95

absence of exogenous substrate. Thereafter, [14C]methylglobin was added, as well as ATP to those tubes lacking it, and the remaining activity was then assayed at 37°C for 60 rain. The loss of activity was 20% when the pre-incubation was conducted at 37°C, and 90% at 42°C, in the absence of ATP. Pre-incubation in the presence of A T P protected the activity markedly (Fig. 4). A T P concentrations as low as 0.1 m M stabilized the protease to roughly the same extent. Citrate at 50 mM, creatine phosphate 20 mM, E D T A 1 m M and the non-hydrolysable ATP analogue, fly-methylene ATP failed to stabilize the enzyme.

Zinc inhibition ZnCI 2 addition was markedly inhibitory to the protease activity, such that 80% inhibition was obtained at 0.1 m M (Table II). Imidazole, a known chelator of Zn 2÷, activated in standard assays by 25%.

freezing. However, when the enzyme was kept at 4°C, no loss of activity was observed over the first 7 days, following which a gradual diminution in activity was noted, reaching 60% of the initial activity at 14 days. The p H range of the protease with radioactive globin extended from 6.5 to 9.0 but the optimum was 7.5. The protease appeared to contain an essential cysteine group, since dithiothreitol and cysteine addition enhanced its activity, while Ne t h y l m a l e i m i d e and p - c h l o r o m e r c u r i p h e n y l sulphonic acid were inhibitory (Table II). Curiously, iodoacetate did not inhibit the enzyme and we cannot explain this finding at this stage. Chymostatin, leupeptin and pepstatin had no effect at concentrations known to inhibit the activities of entities such as the mast cell serine protease [25], Ca2+-activated protease [26] and the lysosomal acid hydrolases cathepsins B, H, L and D [27]. E D T A and E G T A did not inhibit the enzyme.

Other properties The purified enzyme was unstable in the frozen state, losing 80% of its activity within 4 days of

Substrate selectivity In addition to [14C]methylglobin, which may be

405

400-

considered to be a denatured but soluble protein, the enzyme preparation attacked soluble denatured casein and azocasein, because addition of these proteins to the assay markedly diminished the production of radioactive products from the globin substrate; addition of bovine serum albumin did not 'dilute' the product radioactivity in standard assays and this protein may thus not be attacked by the endoprotease (Table II). The activity of crystalline hexokinase (assayed by the method of [28]) was diminished by 50% after incubation for 1 h with the protease. The protease appeared to be an endopeptidase since analysis of the trichloroacetic acid-soluble products on Sephadex G25 revealed radioactive peptide ranging in molecular weight from 1500 to 150 000 (Fig. 5).

L/\

300"

d.p.m. 200-

/

100 -

\

~ s.s.s.9,.s,s-~ • /, 10

20

!

310 Fractions

40

50

Fig. 5. Size analysis of protease digestion products. A standard assay mixture, constituted as described in Materials and Methods and containing enzyme purified as in Fig. 2, was incubated with []4C]methylglobin and stopped by addition of trichloroacetic acid. After ccntrifugation, the supernatant was chromatographed on Sephadex G25, and radioactivity in the fractions (0.4 ml each) was measured. The unincubated and unprecipitated radioactive giobin gave a single peak appearing at the Vo position.

TABLE III PHYSIOLOGICAL MODULATION OF PROTEOLYTIC RATES IN SKELETAL MUSCLES: PROTEASE ACTIVITIES Proteolytic rates were measured in intact soleus and extensor digitorum longus muscles, as described in Materials and Methods. The high-molecular-weight protease activities in hindleg skeletal muscles dissected from control, compound 48/80-treated, starved, hyperthyroid, diabetic and denervated rats, were determined in (unheated) high-speed supernatant preparations subjected to Scpharose 6B chromatography as also described in Materials and Methods. Expt. No.

Control

Compound 48/80-treated

Starvation

Hyperthyroidism

Denervation

Diabetes mellitus

Proteolytic rates in intact tissue (nmol tyrosine/ mg per 2 h) Soleus

Extensor digitorum longus

0.31 0.30 0.32 0.29 0.28 0.30 0.87 0.92 0.85 0.49 0.47 0.48 0.55 0.59 0.52 0.47 0.44 0.42

0.23 0.21 0.24 0.22 0.21 0.20 0.64 0.61 0.65 0.37 0.39 0.35 0.56 0.60 0.55 0.33 0.35 0.32

Protease activities Total (dpm/min)

Specific ( d p m / m g per min)

1 380 1 200 1250 l 100 1400 1 300 1000 1300 1 350 1 300 1 500 1 700 1 200 i 350 1600 1 250 1400 1 100

27000 26000 25 000 27000 27 500 25 000 22000 29 000 27 500 28 000 25 000 30000 27000 26 500 29 500 26000 28 500 25000

406

Absence of "ubiquitin effect" A crude 'APF-I' preparation, obtained by the method of Ciechanover et al. [11] from rat erythrocytes as well as from fresh skeletal muscle, was added in varying amounts, in the presence and absence of ATP, to enzyme preparations at various levels of purification. This was done in order to detect the possible participation, in an ATP-dependent system, of an SH-bearing 'activating' enzyme [29]. No complementation between these 'APF-I' preparations and the muscle extracts was detected at any stage, and the release of labelled globin fragments was in fact slightly diminished, presumably because of substrate competition (data not shown).

Altered states of muscle proteolysis Because of the difficulties inherent in assaying a protease activity in the presence of varying quantities of competing protein substrates and endogenous inhibitors, activity levels in rat skeletal muscles (obtained from control, compound 48/80-treated, starved, denervated and hyperthyroid rats) were routinely estimated after Sepharose 6B chromatography of fraction 1 preparations, as the column-purified material showed a linear relationship between activity and enzyme concentration (data not shown). The fractions obtained all contained similar amounts of protein, since the tissue to homogenization-fluid ratio was kept constant at 3:1, w/v. The activities (both total and specific) of the high-molecular-weight protease did not differ significantly in skeletal muscle samples which were derived from controls and from denervated, diabetic, hyperthyroid, starved or compound 48/80-treated animals (Table III). Except for the degranulated group, all the 'myopathic' models expressed themselves in enhanced rates of tyrosine release by incubated soleus and extensor digitorum longus muscles, compared with muscles derived from control animals. Discussion

There appear to be important differences between the high-molecular-weight proteinase we have prepared from rat skeletal muscle and the proteinase isolated by Hardy et al. [9] from human skeletal muscle. Examples are the rapid action of

our enzyme on casein and globin, the absence of aminopeptidase activity, lack of Ca 2+ stimulation, marked dithiothreitol stimulation, and absence of inhibition by ATP. No data are available in respect of the powerful Zn 2+ inhibition and the heat-stabilization afforded by ATP, which are here reported as features of our enzyme. The crude muscle enzyme briefly described by Etlinger and co-workers [10] displays marked activation by ATP which is diminished by hemin addition, and is apparently dependent on a ubiquitin-like fraction also obtained from muscle; no other details are available. Another cysteine proteinase isolated from smooth muscle has been found to have a very large size (1 300000) and a mildly alkaline pH optimum [30]. Both casein and albumin are attacked and Zn 2÷ inhibits weakly (30% at 1 mM); no effects of ATP were reported, either in terms of activation or stabilization of the very heat-labile enzyme. Clearly, one of the important requirements of our study was also to ensure a clear distinction between the high-molecular-weight protease and various other, previously well-characterized proteases, apart from those mentioned above. Our enzyme was optimally active at pH 7.5 and was not inhibited by a range of Umezawa antibiotic inhibitors nor stimulated by Ca 2÷ , even in the presence of EDTA. This latter finding elimitated the possibility that one of the Ca: ÷ -requiring cysteine proteases was involved, and leupeptin inhibition shown by such enzymes was also absent [31]. Other enzymes which were excluded from consideration are the mast cell-derived chymase (chymostatin-inhibited), the aminoendopeptidase hydrolase H which is inhibited by leupeptin [9] and was in any case chromatographically separatable from the endoprotease we have studied, and the various cathepsins B, D, L and H [27]. Stimulation by thiol agents and inhibition by p-chloromercuriphenylsulphonic acid and N-ethylmaleimide suggested that a thiol group is essential for activity. One of the points of greatest interest in the area of high-molecular-weight tissue proteinases is the question of ATP involvement. ATP stimulation of protease activity in our enzyme at various levels of purification was variable and slight. However, the heat-stability of the enzyme at 42°C was very

407

significantly enhanced by the nucleotide even at concentrations as low as 0.1 mM, indicating that the enzyme had some kind of binding site for ATP. The ATP-stabilized liver protease described in Ref. 5 was, in addition, stabilized by citrate; this polyanion was ineffective in preventing the thermal denaturation of our enzyme. Nevertheless, our muscle-derived enzyme had several other properties in common with this liver enzyme and with that described by De Martino and Goldberg [6]. Thus, its pH optimum was also between 7 and 8, and the activation by thiol compounds, inhibition by N-ethylmaleimide and p-chloromercuriphenylsulphonic acid, heat stabilization by ATP and endopeptidase character were all similar. Our enzyme differed from that of De Martino and Goldberg [6], however, in that their enzyme showed a more marked inhibition caused by iodoacetate (mild or absent in the case of our enzyme), a marked stimulation (almost 100%) caused by ATP (slight and variable in our case) and enhancement of ATP dependence by addition of the non-ionic detergent, Triton X-100 (absent in our case). Again, the high-molecular-weight (450 000) protease characterized in reticulocyte extracts required ATP for thermal stabilization and participates together with ubiquitin in a multi-component, ATP-dependent system which catalyses the selective degradation of denatured globin to the level of amino acids [32]. It is noteworthy that albumin is attacked by the reticulocyte system while our muscle enzyme apparently did not act on this protein. Since /~,methylene ATP failed to stabilize the muscle protein, a possible role for ATP hydrolysis may eventually be found for a system of which this highmolecular-weight protease may be a part. The observation that the activity of the protease did not correlate with changes in overall muscle proteolytic rates in situ suggests that this protease is not rate-limiting for the degradation process. However, the assayable activities of enzyme systems participating in intracellular protein degradation may well remain constant in muscle cells while regulation is mediated by variations in the levels of endogenous inhibitor(s) and/or conformtional changes altering substrate protein susceptibility or changes in assayable levels of separate, rate-limiting enzymes. Since mast cell degranulation caused no change in the activity of the high-

molecular-weight protease, the enzyme may well be muscle cell-derived, although immunochemical evidence will be needed to establish the exact location of the protease in muscle tissue. No specific inhibitor was detected in the high-speed supernatant fraction, but endogenous native proteins competed with the denatured radioactively labelled substrate for degradation. The role of the enzyme is thus not limited to the degradation of 'abnormal' or partially denatured proteins, as appears to be the case in the reticulocyte system [2,4]. Thus, we have characterized a high-molecularweight, cysteine endopeptidase from rat skeletal muscle present in the particle-free fraction of muscle extracts and optimally active at pH 7.5. Because of the significant thermal stabilization of the enzyme caused by ATP, this protease may form part of a multi-component ATP-dependent proteolytic system similar to that reported in reticulocytes. Direct evidence for a role of ubiquitin in muscle has not yet been obtained, however.

Acknowledgements This work was supported by grants from the South African Medical Research Council and Atomic Energy Board. The assistance of B. Sedres and the helpful comments of Dr. D.R. Van der Westhuyzen are gratefully acknowledged. F.I. was a Guy Elliott Medical Research Fellow.

References 1 Etlinger, J.D. and Goldberg, A.L. (1977) Proc. Natl. Acad. Sci. U.S.A. 74, 54-58 2 Goldberg, A.L., Strnad, N.P. and Sre.e.dhara Swamy, K.H. (1980) in Protein Degradation in Health and Disease, Ciba Foundation Syrup. 75, pp. 227-25 I 3 Hershko, A., Cioehanovcr, A. and Rose, I.A. (1980) Proc. Natl. Acad. Sci. U.S.A. 77, 1783-1786 4 Ciechanover, A., Hellcr, H., Elias, S., Haas, A.L. and Hershko, A. (1980) Proc. Natl. Acad. Sci. U.S.A. 77, 1365-1368 5 Rose, I.A., Warms, J.V.B. and Hershko, A. (1979) J. Biol. Chem. 254, 8135-8138 6 De Martino, G. and Goldberg, A.L. (1979) J. Biol. Chem. 254, 3712-3715 7 Young, V.R. and Munro, H.N. (1978) Fed. Proc. Fed. Am. Soc. Exp. Biol. 37, 2291-2300 8 0 k i t a n i , A., Nishimura, T. and Kato, H. (1981) Eur. J. Biochcm. 115, 269--74 9 Hardy, M.F., Mantle, D., Edmunds, T. and Pennington,

408 R.J.T. (1981) B. Soc. Trans. 9, 218-219 10 Etlinger, J.D., Speiser, J., Wajnberg, E. and Glucksman, M.J. (1981) Acta Biol. Med. Germ. 40, 1285-1291 11 Ciechanover, A., Elias, S., Heller, H., Ferber, S. and Hershko, A. (1980) J. Biol. Chem. 255, 7525-7528 12 Bird, J.W.C., Carter, J.H., Triemer, R.E., Brooks, R.M. and Spaniel A.M. (1980) Fed. Proc. 39, 20-25 13 Rice, R.H. and Means, G.E. (1971) J. Biol. Chem. 246, 831-832 14 Schapira, G., Rosa, J., Maleknia, N. and Padieu, P. (1968) Methods Enzymol. 12, part B, 47-769 15 Hardy, M.F. and Pennington, R.J.T. (1979) Biochim. Biophys. Acta 577, 253, 266 16 Lowry, O.H., Rosebrough, N.D., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275 17 Hartree, E.F. (1972) Anal. Biochem. 48, 422-427 18 Pastan, I. and AImqvist, S. (1966) Endocrinology 78, 361-366 19 Goldberg, A.L., Tischler, M., De Martino, G.N. and Griffin, G. (1980) Fed. Proc. 39, 31-36 20 Goidspink, D.F. (1976) Biochem. J. 156, 71-80 21 Mayer, M., Amin, R. and Shafrir, E. (1974) Arch. Biochem. Biophys. 161, 20-25

22 Fulks, R.M., Li, J.B. and Goldberg, A.L. (1975) J. Biol. Chem. 250, 290-298 23 Waalkes, T.P. and Udenfriend, S. (1957) J. Lab. Clin. Med. 50, 733-736 24 Porzio, M.A. and Pearson, A.M. (1977) Biochim. Biophys. Acta 490, 27-34 25 Libby, P. and Goldberg, A.L. (1980) Biochem. J. 188, 213-220 26 Reville, W.J., Goll, D.E., Stromer, M.H., Robson, R.M. and Dayton, W.R. (1976) J. Cell. Biol. 70, 1-8 27 Barrett, A.J. (1977) in Proteinases in Mammalian Cells and Tissues (Borrett, A.J., ed.), pp. 41-44, North-Holland Amsterdam 28 Crabtree, B. and Newsholme, E.A. (1972) Biochem. J. 126, 49-58 29 Haas, A.L., Warms, J.V.B., Hershko, A. and Rose, I.A. (1982) 257, 2543-2548 30 Roth, M., Hoechst, M. and Afting, E.-G. (1981) Acta Biol. Med. Germ. 40, 1357-1363 31 Dayton, W.R., Schollmeyer, J.V., Lepley, R.A. and Cortes, L.R. (19810 Biochim. Biophys. Acta 659, 48-61 32 Hershko, A., Ciechanover, A. and Rose, I.A. (1979) Proc. Natl. Acad. Sci. U.S.A. 76, 3107-3110