ARCHIVES
OF BIOCHEMISTRY
AND
Vol. 283, No. 1, November
BIOPHYSICS
15, pp. 6%74,199O
Phosphorylation of the Multicatalytic Proteinase Complex from Bovine Pituitaries by a Copurifying CAMP-Dependent Protein Kinase’ Maria
E. Pereira
and Sherwin
Mount Sinai School of Medicine
Wilk2
of CUNY,
New York, New York 10029
Received May 16,1990, and in revised form July 6,199O
The multicatalytic proteinase complex is a hollow cylindrical- or disc-shaped, high molecular mass (700 kDa) particle, organized as a stack of four rings (1) and composed of multiple polypeptide subunits ranging in molec-
ular mass between 21 and 34 kDa (for a review see (2)). MPC,3 first isolated and characterized from bovine pituitaries by Wilk and Orlowski (3-5) is present mainly in the cytosolic fraction of the cell, although immunohistochemical studies have localized the enzyme also in the nucleus (6). The complex is present in considerable amounts in all eukaryotic cells that have been examined (reviewed in Ref. (2)), and was found to be identical to the “prosome,” a 19 S ribonucleoprotein particle (6, 7). MPC constitutes a major extralysosomal proteolytic system and may play a key role in intracellular protein turnover (8). The MPC contains three distinct proteolytic activities that split peptide bonds on the carboxyl side of hydrophobic (chymotrypsin-like), basic (trypsin-like) and acidic (peptidylglutamyl-peptide hydrolyzing) amino acids. The integrity of the complex appears to be required for expression of all three catalytic activities, since dissociation leads to inactivation (9). The three activities can be selectively activated or inhibited. For example, leupeptin exclusively inhibits the trypsin-like activity, sodium dodecyl sulfate (SDS) significantly inhibits the trypsin-like activity, but stimulates up to 20-fold the peptidylglutamyl-peptide bond-hydrolyzing activity, and the chymotrypsin-like activity is selectively inhibited by Cbz-Gly-Gly-leucinal(9). The proteinase degrades in vitro a variety of naturally occurring peptides such as bradykinin, angiotensin II, LH-RH, neurotensin, and substance P (3) and has been shown to generate Met- and Leu-enkephalin from precursor molecules (10). Phosphorylated and dephosphor-
r This work was supported by Grants 2T32 DA07135-11 (training grant to M.E.P.) from the National Institute on Drug Abuse, and NS17392 (S.W.) and MH-00350 (Research Scientist Award to SW.) from the National Institutes of Health. * To whom correspondence should be addressed at Mount Sinai School of Medicine of CUNY, Pharmacology Dept., Box 1215, One Gustave L. Levy Place, New York, NY 10029. Fax: (212) 831-0114.
a Abbreviations used: MPC, multicatalytic proteinase complex; PAGE, polyacrylamide gel electrophoresis; LH-RH, luteinizing hormone releasing hormone; PK-A, CAMP-dependent protein kinase; 8-N3-CAMP, 8-azido adenosine 3’,5’-cyclic monophosphate; CbzGly-Gly-Leu-pNA, benzyloxycarbonyl-Gly-Gly-Leu-p-njtroanilide; SDS, sodium dodecyl sulfate; Mes, 2-(N-morpholino) ethanesulfonate; DTT, DL-dithiothreitol.
The multicatalytic proteinase complex (MPC) constitutes a major nonlysosomal proteolytic system that may play an important role in the processing of biologically active peptides and enzymes, as well as in intracellular metabolism. We report that at least two of its subunits of MW 28,800 (S,) and 27,000 (S,) are phosphorylated by a CAMP-dependent protein kinase (PK-A) that copurifies with the complex isolated from bovine pituitaries. The CAMP-induced phosphorylation was time dependent and inhibited by a PK-A inhibitor. Although not an integral part of the complex, PK-A activity was still present even in 1700-fold-purified and apparently homogeneous preparations by criteria of nondissociating polyacrylamide gel electrophoresis. Furthermore, we present evidence that the copurification of the two enzymes is not species or tissue specific, or dependent on a single method of purification. The copurifying kinase was stimulated lo-fold by CAMP (10 FM) and 2- to 3fold by a peptide substrate of the MPC, but was unaffected by protein kinase C activators (calcium and a phospholipid mixture). These findings suggest that protein phosphorylation may represent a mechanism for regulating the activity of the multicatalytic proteinase complex. 0 1990 Academic Press, Inc.
0003-9&n/90
68 All
$3.00
Copyright 0 1990 by Academic Press, Inc. rights of reproduction in any form reserved.
PHOSPHORYLATION
OF BOVINE
MULTICATALYTIC
ylated casein, human serum albumin, lysozyme, phosphorylase b (ll), (Y- and /3-crystallins (12), and oxidized forms of some enzymes (13) are examples of protein substrates degraded by MPC, some of them being extensively broken down to acid-soluble products. Despite the recent increase in the number of publications on MPC, many fundamental questions concerning its properties and mode of action remain to be answered. It is believed that in uiuo MPC is normally present in a latent form (14, 15), thereby protecting the cell against uncontrolled proteolysis. The physiological mechanisms responsible for activation or inhibition of MPC are not known. In vitro studies have shown that low concentrations of SDS, fatty acids (4, 9, 16), and polylysine (17) are activators of the complex, while organic mercurials and thiol blocking agents have been reported to strongly inhibit MPC activity (11). Protein phosphorylation is recognized as an important mechanism regulating cellular metabolism. Pathways involved in carbohydrate, lipid, cholesterol, and protein metabolism demonstrate similar effects of protein phosphorylation or dephosphorylation (18). In an effort to determine if the catalytic properties of MPC were altered by phosphorylation, we discovered that the purified bovine pituitary MPC preparation had phosphotransferase activity. The studies presented herein show that at least two of the subunits of a 1700-fold-purified MPC preparation from bovine pituitaries are phosphorylated by a CAMPdependent protein kinase that copurifies with the complex even after four chromatographic steps and gel electrophoresis. EXPERIMENTAL
PROCEDURES
Supplies Frozen bovine pituitaries were obtained from Pel-Freez Inc (Rogers, AK). CAMP (Tris salt), ATP (Tris salt), EDTA, DTT, Kemptide (Leu-Arg-Arg-Ala-Ser-Leu-Gly), and synthetic PK-A inhibitor (Thr-Thr-Tyr-Ala-Asp-Phe-Ile-Ala-Ser-Gly-Arg-Thr-Gly-ArgArg-Asn-Ala-Ile-His-Asp) were purchased from Sigma Chemical Co. (St. Louis, MO). [3”P]-8-N3-cAMP (50-80 Ci/mmol) and [y-32P]ATP (lo-25 Ci/mmol) were from ICN (Irvine, CA). P-81 cellulose phosphate paper and DEAE-cellulose were obtained from Whatman Inc. (Clifton, NJ). DEAE-Sephacel and Sephacryl S 300 were from Pharmacia (Piscataway, NJ), Ultrogel AcA22 was from IBF Biotechnics Inc. (Savage, MD). and hydroxylapatite was from Calbiochem (San Diego, CA). Gel electrophoresis reagents were obtained from Bio-Rad (Richmond, CA). The following molecular mass marker proteins for SDS gel electrophoresis were obtained from Sigma Chemical Co.: myosin, 205 kDa; P-galactosidase, 116 kDa; phosphorylase B, 97.4 kDa; bovine serum albumin, 66 kDa; ovalbumin, 45 kDa; and carbonic anhydrase, 29 kDa. The proteinase peptide-substrate was synthesized as described (19).
Enzyme Purification MPC was prepared from 100 g of frozen bovine pituitaries following a previously described method (11). After ammonium sulfate precipitation, the cytosol was fractionated by two alternating ion-exchange
PROTEINASE
COMPLEX
69
(DEAE-Sephacel) and gel filtration (AcA-22) chromatography steps. For each column, every third fraction collected was assayed for kinase and proteolytic activities. MPC was also prepared from 100 g of fresh rabbit liver following the same above described procedure (1 l), and from 35 g of frozen bovine pituitaries following a different method of purification (14). The cytosolic fraction was mixed with DEAE-cellulose in a buffer A containing pH 7.6,20 mM NaCl, 1 mM MgCl,, 0.1 mM EDTA, 20 mM Tris-HCl, 0.5 mM DTT, and 20% glycerol, filtered and washed with buffer A containing 0.5 M NaCl. The filtrate was then fractionated by ion-exchange (DEAE-Sephacel) and gel filtration (Sephacryl S 300) chromatography, and applied onto a column of hydroxylapatite equilibrated with the same components of buffer A, except that the 20 mM Tris-HCl was replaced by 25 mM potassium phosphate buffer, pH 7.6 (buffer B). MPC was eluted from the column with a gradient of phosphate (25 to 150 mM phosphate) in buffer B. The MPC-containing fractions were finally rechromatographed on a Sephacryl S 300 column. Kinase and proteolytic activities were measured in each chromatographic step.
Enzyme Activities The activity of the CAMP-dependent protein kinase was determined by its ability to phosphorylate a synthetic substrate (kemptide) in the presence and absence of 10 pM CAMP. Fifty microliters of the column fractions was incubated in a total volume of 125 ~1. The reaction mixture contained 10 mM Tris-HCl, pH 7.4, 5 mM MgCl,, 200 FM [y32P]ATP (100-200 cpm/pmol), and 130 pM kemptide. After a 5-min incubation at 3O”C, the phosphorylation of the synthetic peptide was determined as described (20). The proteolytic activity was determined with the chromogenic substrate Cbz-Gly-Gly-Leu-pNA (400 wM), as previously described (3). Fifty microliters of the column fractions was incubated for 30 min at 37°C. Both reactions were linear with time and enzyme concentration.
Gel Electrophoresis Nondissociating PAGE. After the second AcA-22 column (last step of purification), 8 pg of MPC was resolved by PAGE on a 4.5% gel under nondissociating conditions in a 0.05 M Tris-HCl buffer, pH 8.4, and stained for protein with Coomassie blue or Silver stain. In a parallel experiment, MPC was electrophoresed at 4”C, the gel was cut into 5-mm sections from the top (slice 1) to the dye front (slice 22) and the proteolytic and kinase activities were determined in each slice. The incubation periods were 2 h. The enzyme after the second AcA-22 column SDS-PAGE. also subjected to SDS-PAGE on a 12.5% gel, and electrophoresis continued until the tracking dye migrated to a distance of 25-28 as described (11, 21). The gels were stained with Coomassie blue dried. When appropriate, the incorporation of ‘*P was visualized autoradiography. DuPont-cronex film was exposed for 48 h.
Phosphorylation
was was cm, and by
of MPC by the Associated Kinase
Aliquots (60 ~1) of purified MPC (9.6 pg) were incubated in a total volume of 120 ~1, in the presence and absence of 10 pM CAMP. The reaction mixture contained 10 mM Tris-HCl, pH 7.4, 5 mM MgCl,, and 200 pM [-y-32P]ATP (750-800 cpm/pmol). After 5-, 15., and 30min incubation periods at 3O”C, the reactions were terminated with SDS-stop solution (2% SDS, 5% 2-mercaptoethanol, 10% sucrose, and 0.002% bromphenol blue) and the samples were run on SDSPAGE. In a parallel experiment aliquots of MPC were incubated in the same conditions as described above, but in the presence of 40 pM PK-A inhibitor (22). After 30 min of incubation, the reaction was stopped and the samples run on SDS-PAGE as described above.
70
PEREIRA
AND
WILK
RESULTS
23
1
Phosphmylation of MPC The results of SDS-PAGE and Coomassie blue staining of the bovine pituitary MPC-preparation obtained following the method of (11) revealed the usual multisubunit pattern of polypeptides of MW 22,000-32,000, marked by arrows in Fig. 1, Lane 1. The molecular masses of the seven major resolved polypeptides identified by SDS-PAGE analysis were approximately: 31.0, 28.8, 27.0, 25.6, 24.7, 23.3, and 22.5 kDa. The components were designated S, to S, in decreasing order of their molecular masses (S, = 31.0 kDa). MPC proved to be a very effective substrate for a CAMP-dependent protein kinase that copurified with our MPC preparation, even after the last chromatographic step (Table I). The kinase phosphorylated the complex in a CAMP- and time-dependent manner (Fig. 1, Lanes 4-7). When the MPC preparation was subjected to phosphorylating conditions, using [y-32P]ATP in the presence of CAMP, and the phosphorylated reaction products were separated by SDS-PAGE and detected by autoradiography, we observed labeling of at least six major bands (Fig. 1, Lanes 4-7). Two of the MPC components with molecular masses of 28.8 kDa (S,) and 27.0 kDa (S,) exhibited high levels of phosphorylation, when compared with two other components (S, and S,), which were only slightly radiolabeled. The Sz and S3 components reacted in immunoblots with the rabbit antiserum raised in our laboratory against the purified bovine pituitary preparation (results not shown). In addition, two bands with molecular masses of 45 and 55 kDa, not visible by Coomassie blue or silver staining, were also phosphorylated. The 55-kDa band likely corresponds to the regulatory subunit of PK-A Type II, which undergoes autophosphorylation by the catalytic subunit (25). The 45-kDa band probably results from the proteolytic cleavage of the 55kDa subunit. Phosphorylation of MPC was inhibited by 40 PM PK-A inhibitor (Fig. 1, Lane 7). The labeling of
4567
FIG. 1. Lane 1, subunit composition of MPC after SDS-PAGE, visualized with Coomassie blue staining. The arrows represent the seven major subunits identified, labeled &ST, in decreasing order of molecular mass; Lanes 2 and 3, gel autoradiograph of the incorporation of [32P]-8-N3-cAMP into the PK-A, in the absence (Lane 2) and presence (Lane 3) of excess nonradiolabeled CAMP (100 PM). The band (55 kDa) represents the regulatory subunit of the PK-A, Type II; Lanes 4 to 6, gel autoradiograph of the phosphorylation of the MPC preparation in the presence of 10 pM CAMP, after 5 (Lane 4), 15 (Lane 5), and 30 min (Lane 6) of incubation. Lane 7 represents a 30-min incubation with the addition of PK-A inhibitor. The arrows on the left of Lane 4 indicate phosphorylated bands; molecular mass markers are indicated on the right of Lanes 1 and 7 in kDa. Approximately 5 pg of the purified bovine pituitary MPC preparation was loaded onto each lane of the gel.
Photoafinity Labeling of the CAMP-Dependent Protein Kinase Aliquots (50 ~1) of purified MPC (8 pg) were incubated in a total volume of 100 ~1 in the presence and absence of an excess of CAMP (100 NM). The reaction mixture contained 50 mM Mes, pH 6.2,lO mM MgCls, 1 mM 2-mercaptoethanol, and 1 pM [32P]-8-Ns-cAMP and the incubating conditions were as in (23). The reaction was stopped with SDS-stop solution and 100~~1 aliquots were subjected to SDS-PAGE and autoradiography as described above.
Protein Assay Protein
concentrations
were determined
by the Lowry method (24).
TABLE Summary
of the Activity
of the PK-A
in the Consecutive
I Steps of MPC
Purification
step
Homogenate Supernatant Ammonium sulfate fractionation DEAE-Sephacel chromatography (pH 7.5) Ultrogel chromatography (pH 8.3) DEAE-Sephacel chromatography (pH 8.3) Second Ultrogel chromatography (pH 7.5)
Protein hdml)
Control
CAMP
0.041 0.080
26.64 28.27
0.167
10.37
0.052 0.037 0.048 0.104
7.40 1.85 2.31 0.08
370 620
40.80 13.60
0.072 0.046
80 188 110 152
15.40
0.130 0.039
6
0.38 0.24
0.017
0.12
0.015
0.18
0.013
from
Control
Bovine
Pituitaries”
Sp act (U/mg)
Total U
U/ml Volume (ml)
Purification
CAMP
15.10 49.85 13.34 9.72
4.05 7.28 0.62
Control
0.002 0.003
CAMP
Total (%)
0.001
100
0.006
330 88 64 27 48 4
0.008
0.011
0.105 0.071 0.127 0.070
0.138 0.157 0.399
0.576
’ Activity was determined with kemptide as the substrate (0.130 mM) as described under Experimental Procedures; U (units) defined as the amount of enzyme that transfers one pmol of 32P to the substrate per hour, in the absence (control) and presence of 10 ticM CAMP.
PHOSPHORYLATION
OF BOVINE
MULTICATALYTIC
2
?a! 206 , 3i6 9 10 11
12 13 14 15
16 17
18 19 20 21 22
0.0
1.0
FIG. 2. A, Nondissociating PAGE of the purified MPC (8 pg). B, profiles of the proteolytic activity @) (absorbance at 540 nm X 3) and kinase activity (m) (cpm X 10m4) of the 4.5% gel.
purified MPC CAMP followed vealed a major consistent with Purification
with the affinity reagent [32P]-8-N3by SDS-PAGE and autoradiography reband at 55 kDa (Fig. 1, Lanes 2 and 3), the regulatory subunit of PK-A (23).
of MPC
The purity of the MPC preparation after the final AcA22 column was verified by PAGE. MPC migrated as a single band in nondissociating PAGE, when stained with either Coomassie blue (Fig. 2A) or silver stain (not shown). When the gels were sliced into 5-mm segments and both kinase and proteinase activities determined in
TABLE
PROTEINASE
COMPLEX
71
each segment, kinase activity could always be measured in slices with proteolytic activity (Fig. 2B). Kinase and proteinase activities were monitored throughout the purification procedure and, as shown in Tables I and II, CAMP-dependent phosphotransferase activity could be detected in all of the purification steps. Table I shows that three times more CAMP-dependent protein kinase activity was present in the supernatant fraction than in the homogenate. This could be due to removal of endogenous PK-A inhibitors. No PK-A activity could be detected in proteinase-containing fractions of the ion-exchange columns, due to the high EDTA concentrations of the gradient buffers. The MPC-containing fractions eluted from both ion-exchange columns at the end of the gradient (0.3 M Tris, 96 mM EDTA, pH 7.5, in the first and 0.4 M Tris, 70 mM EDTA, pH 8.3, in the second column). High EDTA concentrations chelated the magnesium ions present in the assay mixture, essential for phosphotransferase activity. To measure PK-A activity, the samples were dialyzed against a 0.01 k Tris-EDTA, pH 7.5, buffer, where the EDTA concentration was 2.7 mM. Purification of MPC in Tris-HCl buffers containing lower EDTA concentrations led to unacceptably low recoveries, excluding their practical use. The phosphotransferase activity of the second DEAE-Sephacel column was variable, sometimes being higher (as shown in Table I) other times lower (not shown) than in the preceding step, but it was always readily measured after dialysis. Although a 1700-fold purified MPC preparation was obtained (Table II), a significant amount of the CAMPdependent protein kinase activity remained associated with the proteinase as shown in the last gel filtration step (Table I and Fig. 3). Furthermore, CAMP-dependent phosphotransferase activity could also be measured in MPC-preparations obtained either from rabbit liver following the same purification method as in (ll), or from bovine pituitaries after the purification method of
II
Summary of the Purification of the MPC from Bovine
Pituitaries”
Activity Purification
step
Homogenate Supernatant Ammonium sulfate fractionation DEAE-Sephacel chromatography (pH 7.5) Ultrogel chromatography (pH 8.3) DEAE-Sephacel chromatography (pH 8.3) Second Ultrogel chromatography (pH 7.5)
Volume (ml)
Protein bx/mU
U/ml
Total U
370 620
40.80 13.60
0.400 0.260
148.0 161.2
15.40
1.440
115.2
0.38 0.24 0.12
0.280 0.420 0.210 3.080
52.6 46.2
80 188
110 152 6
0.18
Spec act W/md
Recovery (%:o)
0.010 0.019 0.094
31.9
0.747 1.787 1.750
18.5
17.111
n Activity was determined with Cbz-Gly-Gly-Leu-pNA as the substrate (0.4 mM) as described under Experimental defined as the amount of enzyme that cleaves one pmol of the substrate per hour.
Purification (fold)
100 109 78 36 31 22 12
Procedures.
1 2
10 76 182
179 1745
U (units)
‘72
PEREIRA
o.ol 0
AND
Cbz-Gly-Gly-Leu-pNA W)
58
28
Fraction
#
FIG. 3. Distribution of MPC activity (open circles) and the CAMPdependent protein kinase activity (closed circles) in the last gel filtration fractions. Proteolytic activity (0) is expressed in U/ml and kinase activity (0) is expressed in U/ml X 1O-5; (0) absorbance at 280 nm. Every third fraction collected (12 ml/fraction) was assayed for chymotrypsin-like and kinase activities.
(14) (Table III). The ratio (R) of the specific activities of PK-A and MPG in the three different preparations after the last step of purification was of the same order of magnitude (0.03-0.09), as can be seen in Table III. Phosphotransferase
Activity
WILK
of the MPC Preparation
The kinase activity copurifying with MPC was stimulated 8- to lo-fold by CAMP (10 PM) and was inhibited by 40 FM PK-A inhibitor (results not shown). No protein kinase C was present in our preparation since calcium (0.48 mM) and/or a phospholipid mixture (20 PM phosphatidyl-L-serine and 0.32 PM dioctanoyl glycerol) failed to stimulate any phosphotransferase activity.
TABLE
Chymotrypsin-Like
FIG. 4. Right ordinate (closed symbols), proteolytic activity (0) expressed in U/mg of protein X 10, as a function of the concentration of the Cbz-Gly-Gly-Leu-pNA substrate (abscissa). Left ordinate (open symbols), kinase activity (0) expressed as U/mg of protein X 10 measured in the presence of Cbz-Gly-Gly-Leu-pNA (abscissa). Data from a representative experiment.
The kinase activity remaining associated with MPC after the last step of purification was measured in the presence of the chromogenic MPC-peptide substrate, Cbz-Gly-Gly-Leu-pNA. A two- to threefold activation of the phosphotransferase activity was observed (Fig. 4). The sigmoid relationship between the reaction velocity (ordinate) versus the activator concentration (abscissa) is suggestive of an allosteric interaction. The proteinase substrate did not activate the PK-A previously resolved from the complex after the first DEAE-Sephacel column (results not shown). The activation of the copurifying PK-A by Cbz-GlyGly-Leu-pNA was not due to protection of the kinase substrate kemptide from proteolysis by MPC. Degradation of kemptide by MPC in the absence of Cbz-GlyGly-Leu-pNA was monitored by HPLC and was found
III
and Phosphotransferase Activities of Three Different MPC Preparations, after the Last Step of Purification” Phosphotransferase
Source and method of MPC preparation
Chymotrypsinlike activity
activity lo PM
Control
CAMP
R
17.1 U/mg
0.070 U/mg
0.576 U/mg
0.034
4.2 U/mg
0.043 U/mg
0.229 U/mg
0.055
3.5 U/mg
0.105 U/mg
0.316 U/mg
0.090
A. Bovine pituitaries,
after Orlowski
and Michaud
(1989). 100 g of tissue.
B. Rabbit livers, after Orlowski
and Michaud
(1989). 100 g of tissue.
C. Bovine pituitaries,
after McGuire
et al. (1989). 35 g of tissue.
’ Chymotrypsin-like activity was determined with the substrate Cbz-Gly-Gly-Leu-pNA as in Table II; phosphotransferase activity was determined with the substrate kemptide as in Table I; units for each activity as described in Tables I and II; the numbers represent specific activities; R, ratio of specific activities of PK-A (in the presence of cAMP)/MPC (measured as in Tables I and II).
PHOSPHORYLATION
OF BOVINE
MULTICATALYTIC
to be less than 10% during the period of incubation (results not shown). The relationship between concentration of Cbz-Gly-Gly-Leu-pNA and its hydrolysis by MPC is also shown in Fig. 4 for comparison. DISCUSSION
In the present studies, we show that at least two MPC subunits (S,, MW 28,800 and S3, MW 27,000) of the bovine pituitary preparation are strongly phosphorylated in a CAMP-dependent fashion by a PK-A that copurifies along with the 1700-fold-purified MPC. Two other subunits (S,, MW 31,000 and S5, MW 24,700) are also phosphorylated, but to a lesser degree. Reports by Haass and Kloetzel (26) on Drosophila MPC have shown that 28kDa subunits are substrates for phosphorylation. They suggest that this post-translational modification might be responsible for the development-dependent variation in subunit patterns of the MPC. In addition, cDNA cloning and sequencing studies of the 35kDa subunit of Drosophila MPC revealed a consensus sequence for tyrosine phosphorylation identical to sites present in lactate dehydrogenase and several viral particles (27). Since the amino acid sequence of the 35-kDa subunit demonstrated no significant homology with any of the known proteases, Haass et al. (27) postulated that this MPCcomponent was involved in activity regulation rather than enzymatic cleavage. Furthermore, Tanaka and his co-workers (28) have demonstrated that the C, subunit (MW 25,800) of the rat liver MPC contains a sequence of 70 amino acids, homologous to the autophosphorylation tyrosine residues of various cellular tyrosine kinases such as src gene products and some receptor proteins. Taken together, the above findings suggest that MPC phosphorylation might be a mechanism for regulating its proteolytic activities. In vitro studies of other enzymes show their activities to be regulated by phosphorylation. For example, Drosophila topoisomerase II (an enzyme producing cleavage and resealing of supercoiled DNA) undergoes rapid phosphorylation by protein kinase C, resulting in the stimulation of its activity (29). Moreover, purified preparations of Drosophila topoisomerase II contain an endogenous protein kinase, and the topoisomerase is a phosphoprotein (30). Monkey brain aminopeptidase activity was inhibited by CAMP-dependent phosphorylation (31). Tyrosine hydroxylase from bovine adrenal cells (32) was phosphorylated and activated in vitro, by at least three protein kinases, namely PK-A, protein kinase C, and calcium/calmodulin-dependent protein kinase (32). The PK-A, which copurifies with the bovine pituitary MPC, has a CAMP-binding regulatory subunit of 55 kDa and is stimulated 8- to lo-fold by CAMP (10 PM), inhibited by a synthetic PK-A inhibitor, affinity labeled with 8-Azido-CAMP, and unaffected by protein kinase C activators. Its specific activity in the presence of CAMP, cal-
PROTEINASE
COMPLEX
73
culated to be 0.576 U/mg of protein (U = pmol of 32P transferred/h) is 5.6-fold less than the specific activity reported for PK-A type II purified from bovine brain (3.26 U/mg of protein) (33). Since the bulk of the protein in our preparation is contributed by MPC, the lower kinase specific activity is expected. Our evidence suggests that the copurifying PK-A activity resides in a polypeptide other than the MPC, since its regulatory component (visible only by radiolabeling with 32P) has a greater MW (55,000) than any of the MPC subunits. The phosphorylation of the MPC subunits by the copurifying kinase, and the two- to threefold stimulation of the PK-A by a synthetic MPC substrate, is suggestive of a functional interaction between the two systems. Furthermore, we show that several fractionation techniques resolving proteins according to different molecular properties only partially separated the PK-A from the MPC. PK-A copurified with the complex from rabbit liver following the same purification procedure (ll), or from bovine pituitaries using a different isolation method (14), indicating that the copurification of the two enzymes is not species or tissue specific, or dependent on a single method of purification. This may indicate that in vivo the two protein species are tightly associated. Recent studies have indicated that MPC is associated with at least two other factors in a larger 26 S complex. This macromolecular complex degrades ubiquitinated proteins, with MPC postulated to be the catalytic core (34-36). Furthermore, the assembly of the three components of the 26 S complex was implied to be ATP dependent (34-36). Ganoth et al. (37) have indicated that, if ATP is required for complex formation, a protein kinase could be one of the three components of the 26 S system. The phosphorylation of MPC subunits by PK-A and/ or tyrosine kinase represents a potential mechanism for controlling its activity in response to extracellular stimuli. The identity of the MPC phosphorylation sites for PK-A requires future analysis. Studies currently under investigation in our laboratory will determine if the state of MPC phosphorylation in the bovine pituitary will affect its proteolytic activity, and/or its incorporation into a larger ATP-dependent proteolytic system, the 26 S complex. REFERENCES 1. Kopp, F., Steiner, R., Dahlmann, B., Kuehn, L., and Reinauer, (1986) B&him. Biophys. Actu 872,253-260.
H.
2. Rivett, A. J. (1989) Arch. Biochem. Biophys. 268, 1-8. 3. Wilk, S., and Orlowski, 4. Orlowski, M., and Wilk, mun. 101,814-822. 5. Orlowski,
M. (1980) J. Neurochem. S. (1981) B&hem.
35, 1172-1182.
Biophys.
Res. Com-
M., and Wilk, S. (1988) Biochem. J. 255, 750-751.
6. Arrigo, A. P., Tanaka, K., Goldberg, Nature (London) 331,192-194.
A., and Welsh, W. J. (1988)
74
PEREIRA
7. Falkenburg, P. E., Haass, C., Kloetzel, P. M., Niedel, B., Kopp, F., Kuehn, L., and Dahlmann, B. (1988) Nature (London) 331,190192. 8. Rivett, A. J. (1989) J. Biol. Chem. 264, 12,2X-12,219. 9. Wilk, S., and Orlowski, M. (1983) J. Neurochem. 40,842-849. 10. Orlowski, M., Michaud, C., and Wilk, S. (1980) Biochem. Biophys. Res. Commun. 94,1145-1153. 11. Orlowski, 9278.
M., and Michaud,
12. Ray, K., and Harris, 7545-7549. 13. Rivett,
C. (1989) Biochmistry
28, 9270-
H. (1985) Proc. Natl. Acad. Sci. USA 82,
A. J. (1985) J. Biol. Chem. 260,12,600-12,606.
14. McGuire, M. J., McCullogh, M. L., Croall, D. E., and DeMartino, G. N. (1989) Biochim. Btiphys. Acta 996,181-186. 15. Tanaka, K., Ii, K., Ichihara, A., Waxman, (1986) J. Biol. Chem. 261,15,197-15,203.
L., and Goldberg,
16. Dahlmann, B., Rutschmann, M., Kuehn, (1985) Biochem. J. 228,171-177.
A. L.
L., and Reinauer,
H.
17. Tanaka, K., Yoshimura, T., Ichihara, A., Ikai, A., Nishigai, Morimoto, Y., Sato, M., Tanaka, N., Katsube, Y., Kameyamo, and Takagi, T. (1988) J. Mol. Biol. 203,985-996.
M., K.,
18. Shenolikar, 19. Wilk, 464.
S. (1988) FASEB
J. 2,2753-2764.
S., Pearce, S., and Orlowski,
M. (1979) Life Sci. 24, 457-
20. Roskoski, R., Jr. (1983) in Methods in Enzymology (Corbin, J. D., and Hardman, J. G., Eds.), Vol. 99, pp. 3-6, Academic Press, San Diego. 21. Laemmli,
U. K. (1970) Nature
(London)
227,680-685.
22. Cheng, H. C., Kemp, B. E., Pearson, R. B., Smith, A. J., Misconi, L., Van Patten, S. M., and Walsh, D. A. (1986) J. Biol. Chem. 261,
989-992.
AND
WILK
23. Walter, U., and Greengard, P. (1983) in Methods in Enzymology (Corbin, J. D., and Hardman, J. G., Eds.), Vol. 99, pp. 154-162, Academic Press, San Diego. 24. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275. R., and Rosen, 0. M. (1976) J. Biol. Chem. 251, 25. Rangel-Aldao, 7526-7532. 26. Haass, C., and Kloetzel, P. M. (1989) Erp. CelZRes. 180.243-252. B., Multhaup, G., Beyreuther, K., and 27. Haass, C., Pesold-Hurt, Kloetzel, P. M. (1989) EMBO J. 8,2373-2379. 28. Tanaka, K., Fujiwara, T., Kumatori, A., Shin, S., Yoshimura, T., Ichihara, A., Tokunaga, F., Aruga, R., Iwanaga, S., Kakizuka, A., and Nakanishi, S. (1990) Biochemistry 29,3777-3785. 29. Sahyoun, N., Wolf, M., Besterman, J., Hsieh, T., Sander, M., Levine III, H., Chang, K., and Cuatrecasas, P. (1986) Proc. Natl. Acad. Sci. USA 83,1603-1607. 30. Sander, M., Nolan, J. M., and Hsieh, T. (1984) Proc. N&l. Acad. Sci. USA8 1,6983-6942. 31. Ramamoorthy, S., and Balasubramanian, A. (1989) Biochem. J. 258,777-783. 32. George, J. R., Haycock, J. W., Johnston, J. P., Craviso, G. L., and Waymire, J. C. (1989) J. Neurochem. 52,274-284. 33. Miyamoto, E., Petzold, G. L., Kuo, J. F., and Greengard, P. (1973) J. Biol. Chem. 248,179-189. 34. Matthews, W., Tanaka, K., James, D., Ichihara, A., and Goldberg, A. L. (1989) Proc. Natl. Acad. Sci. USA 86,2597-2601. 35. Eytan, E., Ganoth, D., Armon, T., and Hershko, A. (1989) Proc. Natl. Acad. Sci. USA 86,7751-7755. 36. Driscoll, J., and Goldberg, A. L. (1990) J. Biol. Chem. 266,47894792. 37. Ganoth, D., Leshinsky, E., Eytan, E., and Hershko, A. (1988) J. Biol. Chem. 263,12,412-12,419.
OF BIOCHEMISTRY
AND
Vol. 283, No. 1, November
BIOPHYSICS
15, pp. 6%74,199O
Phosphorylation of the Multicatalytic Proteinase Complex from Bovine Pituitaries by a Copurifying CAMP-Dependent Protein Kinase’ Maria
E. Pereira
and Sherwin
Mount Sinai School of Medicine
Wilk2
of CUNY,
New York, New York 10029
Received May 16,1990, and in revised form July 6,199O
The multicatalytic proteinase complex is a hollow cylindrical- or disc-shaped, high molecular mass (700 kDa) particle, organized as a stack of four rings (1) and composed of multiple polypeptide subunits ranging in molec-
ular mass between 21 and 34 kDa (for a review see (2)). MPC,3 first isolated and characterized from bovine pituitaries by Wilk and Orlowski (3-5) is present mainly in the cytosolic fraction of the cell, although immunohistochemical studies have localized the enzyme also in the nucleus (6). The complex is present in considerable amounts in all eukaryotic cells that have been examined (reviewed in Ref. (2)), and was found to be identical to the “prosome,” a 19 S ribonucleoprotein particle (6, 7). MPC constitutes a major extralysosomal proteolytic system and may play a key role in intracellular protein turnover (8). The MPC contains three distinct proteolytic activities that split peptide bonds on the carboxyl side of hydrophobic (chymotrypsin-like), basic (trypsin-like) and acidic (peptidylglutamyl-peptide hydrolyzing) amino acids. The integrity of the complex appears to be required for expression of all three catalytic activities, since dissociation leads to inactivation (9). The three activities can be selectively activated or inhibited. For example, leupeptin exclusively inhibits the trypsin-like activity, sodium dodecyl sulfate (SDS) significantly inhibits the trypsin-like activity, but stimulates up to 20-fold the peptidylglutamyl-peptide bond-hydrolyzing activity, and the chymotrypsin-like activity is selectively inhibited by Cbz-Gly-Gly-leucinal(9). The proteinase degrades in vitro a variety of naturally occurring peptides such as bradykinin, angiotensin II, LH-RH, neurotensin, and substance P (3) and has been shown to generate Met- and Leu-enkephalin from precursor molecules (10). Phosphorylated and dephosphor-
r This work was supported by Grants 2T32 DA07135-11 (training grant to M.E.P.) from the National Institute on Drug Abuse, and NS17392 (S.W.) and MH-00350 (Research Scientist Award to SW.) from the National Institutes of Health. * To whom correspondence should be addressed at Mount Sinai School of Medicine of CUNY, Pharmacology Dept., Box 1215, One Gustave L. Levy Place, New York, NY 10029. Fax: (212) 831-0114.
a Abbreviations used: MPC, multicatalytic proteinase complex; PAGE, polyacrylamide gel electrophoresis; LH-RH, luteinizing hormone releasing hormone; PK-A, CAMP-dependent protein kinase; 8-N3-CAMP, 8-azido adenosine 3’,5’-cyclic monophosphate; CbzGly-Gly-Leu-pNA, benzyloxycarbonyl-Gly-Gly-Leu-p-njtroanilide; SDS, sodium dodecyl sulfate; Mes, 2-(N-morpholino) ethanesulfonate; DTT, DL-dithiothreitol.
The multicatalytic proteinase complex (MPC) constitutes a major nonlysosomal proteolytic system that may play an important role in the processing of biologically active peptides and enzymes, as well as in intracellular metabolism. We report that at least two of its subunits of MW 28,800 (S,) and 27,000 (S,) are phosphorylated by a CAMP-dependent protein kinase (PK-A) that copurifies with the complex isolated from bovine pituitaries. The CAMP-induced phosphorylation was time dependent and inhibited by a PK-A inhibitor. Although not an integral part of the complex, PK-A activity was still present even in 1700-fold-purified and apparently homogeneous preparations by criteria of nondissociating polyacrylamide gel electrophoresis. Furthermore, we present evidence that the copurification of the two enzymes is not species or tissue specific, or dependent on a single method of purification. The copurifying kinase was stimulated lo-fold by CAMP (10 FM) and 2- to 3fold by a peptide substrate of the MPC, but was unaffected by protein kinase C activators (calcium and a phospholipid mixture). These findings suggest that protein phosphorylation may represent a mechanism for regulating the activity of the multicatalytic proteinase complex. 0 1990 Academic Press, Inc.
0003-9&n/90
68 All
$3.00
Copyright 0 1990 by Academic Press, Inc. rights of reproduction in any form reserved.
PHOSPHORYLATION
OF BOVINE
MULTICATALYTIC
ylated casein, human serum albumin, lysozyme, phosphorylase b (ll), (Y- and /3-crystallins (12), and oxidized forms of some enzymes (13) are examples of protein substrates degraded by MPC, some of them being extensively broken down to acid-soluble products. Despite the recent increase in the number of publications on MPC, many fundamental questions concerning its properties and mode of action remain to be answered. It is believed that in uiuo MPC is normally present in a latent form (14, 15), thereby protecting the cell against uncontrolled proteolysis. The physiological mechanisms responsible for activation or inhibition of MPC are not known. In vitro studies have shown that low concentrations of SDS, fatty acids (4, 9, 16), and polylysine (17) are activators of the complex, while organic mercurials and thiol blocking agents have been reported to strongly inhibit MPC activity (11). Protein phosphorylation is recognized as an important mechanism regulating cellular metabolism. Pathways involved in carbohydrate, lipid, cholesterol, and protein metabolism demonstrate similar effects of protein phosphorylation or dephosphorylation (18). In an effort to determine if the catalytic properties of MPC were altered by phosphorylation, we discovered that the purified bovine pituitary MPC preparation had phosphotransferase activity. The studies presented herein show that at least two of the subunits of a 1700-fold-purified MPC preparation from bovine pituitaries are phosphorylated by a CAMPdependent protein kinase that copurifies with the complex even after four chromatographic steps and gel electrophoresis. EXPERIMENTAL
PROCEDURES
Supplies Frozen bovine pituitaries were obtained from Pel-Freez Inc (Rogers, AK). CAMP (Tris salt), ATP (Tris salt), EDTA, DTT, Kemptide (Leu-Arg-Arg-Ala-Ser-Leu-Gly), and synthetic PK-A inhibitor (Thr-Thr-Tyr-Ala-Asp-Phe-Ile-Ala-Ser-Gly-Arg-Thr-Gly-ArgArg-Asn-Ala-Ile-His-Asp) were purchased from Sigma Chemical Co. (St. Louis, MO). [3”P]-8-N3-cAMP (50-80 Ci/mmol) and [y-32P]ATP (lo-25 Ci/mmol) were from ICN (Irvine, CA). P-81 cellulose phosphate paper and DEAE-cellulose were obtained from Whatman Inc. (Clifton, NJ). DEAE-Sephacel and Sephacryl S 300 were from Pharmacia (Piscataway, NJ), Ultrogel AcA22 was from IBF Biotechnics Inc. (Savage, MD). and hydroxylapatite was from Calbiochem (San Diego, CA). Gel electrophoresis reagents were obtained from Bio-Rad (Richmond, CA). The following molecular mass marker proteins for SDS gel electrophoresis were obtained from Sigma Chemical Co.: myosin, 205 kDa; P-galactosidase, 116 kDa; phosphorylase B, 97.4 kDa; bovine serum albumin, 66 kDa; ovalbumin, 45 kDa; and carbonic anhydrase, 29 kDa. The proteinase peptide-substrate was synthesized as described (19).
Enzyme Purification MPC was prepared from 100 g of frozen bovine pituitaries following a previously described method (11). After ammonium sulfate precipitation, the cytosol was fractionated by two alternating ion-exchange
PROTEINASE
COMPLEX
69
(DEAE-Sephacel) and gel filtration (AcA-22) chromatography steps. For each column, every third fraction collected was assayed for kinase and proteolytic activities. MPC was also prepared from 100 g of fresh rabbit liver following the same above described procedure (1 l), and from 35 g of frozen bovine pituitaries following a different method of purification (14). The cytosolic fraction was mixed with DEAE-cellulose in a buffer A containing pH 7.6,20 mM NaCl, 1 mM MgCl,, 0.1 mM EDTA, 20 mM Tris-HCl, 0.5 mM DTT, and 20% glycerol, filtered and washed with buffer A containing 0.5 M NaCl. The filtrate was then fractionated by ion-exchange (DEAE-Sephacel) and gel filtration (Sephacryl S 300) chromatography, and applied onto a column of hydroxylapatite equilibrated with the same components of buffer A, except that the 20 mM Tris-HCl was replaced by 25 mM potassium phosphate buffer, pH 7.6 (buffer B). MPC was eluted from the column with a gradient of phosphate (25 to 150 mM phosphate) in buffer B. The MPC-containing fractions were finally rechromatographed on a Sephacryl S 300 column. Kinase and proteolytic activities were measured in each chromatographic step.
Enzyme Activities The activity of the CAMP-dependent protein kinase was determined by its ability to phosphorylate a synthetic substrate (kemptide) in the presence and absence of 10 pM CAMP. Fifty microliters of the column fractions was incubated in a total volume of 125 ~1. The reaction mixture contained 10 mM Tris-HCl, pH 7.4, 5 mM MgCl,, 200 FM [y32P]ATP (100-200 cpm/pmol), and 130 pM kemptide. After a 5-min incubation at 3O”C, the phosphorylation of the synthetic peptide was determined as described (20). The proteolytic activity was determined with the chromogenic substrate Cbz-Gly-Gly-Leu-pNA (400 wM), as previously described (3). Fifty microliters of the column fractions was incubated for 30 min at 37°C. Both reactions were linear with time and enzyme concentration.
Gel Electrophoresis Nondissociating PAGE. After the second AcA-22 column (last step of purification), 8 pg of MPC was resolved by PAGE on a 4.5% gel under nondissociating conditions in a 0.05 M Tris-HCl buffer, pH 8.4, and stained for protein with Coomassie blue or Silver stain. In a parallel experiment, MPC was electrophoresed at 4”C, the gel was cut into 5-mm sections from the top (slice 1) to the dye front (slice 22) and the proteolytic and kinase activities were determined in each slice. The incubation periods were 2 h. The enzyme after the second AcA-22 column SDS-PAGE. also subjected to SDS-PAGE on a 12.5% gel, and electrophoresis continued until the tracking dye migrated to a distance of 25-28 as described (11, 21). The gels were stained with Coomassie blue dried. When appropriate, the incorporation of ‘*P was visualized autoradiography. DuPont-cronex film was exposed for 48 h.
Phosphorylation
was was cm, and by
of MPC by the Associated Kinase
Aliquots (60 ~1) of purified MPC (9.6 pg) were incubated in a total volume of 120 ~1, in the presence and absence of 10 pM CAMP. The reaction mixture contained 10 mM Tris-HCl, pH 7.4, 5 mM MgCl,, and 200 pM [-y-32P]ATP (750-800 cpm/pmol). After 5-, 15., and 30min incubation periods at 3O”C, the reactions were terminated with SDS-stop solution (2% SDS, 5% 2-mercaptoethanol, 10% sucrose, and 0.002% bromphenol blue) and the samples were run on SDSPAGE. In a parallel experiment aliquots of MPC were incubated in the same conditions as described above, but in the presence of 40 pM PK-A inhibitor (22). After 30 min of incubation, the reaction was stopped and the samples run on SDS-PAGE as described above.
70
PEREIRA
AND
WILK
RESULTS
23
1
Phosphmylation of MPC The results of SDS-PAGE and Coomassie blue staining of the bovine pituitary MPC-preparation obtained following the method of (11) revealed the usual multisubunit pattern of polypeptides of MW 22,000-32,000, marked by arrows in Fig. 1, Lane 1. The molecular masses of the seven major resolved polypeptides identified by SDS-PAGE analysis were approximately: 31.0, 28.8, 27.0, 25.6, 24.7, 23.3, and 22.5 kDa. The components were designated S, to S, in decreasing order of their molecular masses (S, = 31.0 kDa). MPC proved to be a very effective substrate for a CAMP-dependent protein kinase that copurified with our MPC preparation, even after the last chromatographic step (Table I). The kinase phosphorylated the complex in a CAMP- and time-dependent manner (Fig. 1, Lanes 4-7). When the MPC preparation was subjected to phosphorylating conditions, using [y-32P]ATP in the presence of CAMP, and the phosphorylated reaction products were separated by SDS-PAGE and detected by autoradiography, we observed labeling of at least six major bands (Fig. 1, Lanes 4-7). Two of the MPC components with molecular masses of 28.8 kDa (S,) and 27.0 kDa (S,) exhibited high levels of phosphorylation, when compared with two other components (S, and S,), which were only slightly radiolabeled. The Sz and S3 components reacted in immunoblots with the rabbit antiserum raised in our laboratory against the purified bovine pituitary preparation (results not shown). In addition, two bands with molecular masses of 45 and 55 kDa, not visible by Coomassie blue or silver staining, were also phosphorylated. The 55-kDa band likely corresponds to the regulatory subunit of PK-A Type II, which undergoes autophosphorylation by the catalytic subunit (25). The 45-kDa band probably results from the proteolytic cleavage of the 55kDa subunit. Phosphorylation of MPC was inhibited by 40 PM PK-A inhibitor (Fig. 1, Lane 7). The labeling of
4567
FIG. 1. Lane 1, subunit composition of MPC after SDS-PAGE, visualized with Coomassie blue staining. The arrows represent the seven major subunits identified, labeled &ST, in decreasing order of molecular mass; Lanes 2 and 3, gel autoradiograph of the incorporation of [32P]-8-N3-cAMP into the PK-A, in the absence (Lane 2) and presence (Lane 3) of excess nonradiolabeled CAMP (100 PM). The band (55 kDa) represents the regulatory subunit of the PK-A, Type II; Lanes 4 to 6, gel autoradiograph of the phosphorylation of the MPC preparation in the presence of 10 pM CAMP, after 5 (Lane 4), 15 (Lane 5), and 30 min (Lane 6) of incubation. Lane 7 represents a 30-min incubation with the addition of PK-A inhibitor. The arrows on the left of Lane 4 indicate phosphorylated bands; molecular mass markers are indicated on the right of Lanes 1 and 7 in kDa. Approximately 5 pg of the purified bovine pituitary MPC preparation was loaded onto each lane of the gel.
Photoafinity Labeling of the CAMP-Dependent Protein Kinase Aliquots (50 ~1) of purified MPC (8 pg) were incubated in a total volume of 100 ~1 in the presence and absence of an excess of CAMP (100 NM). The reaction mixture contained 50 mM Mes, pH 6.2,lO mM MgCls, 1 mM 2-mercaptoethanol, and 1 pM [32P]-8-Ns-cAMP and the incubating conditions were as in (23). The reaction was stopped with SDS-stop solution and 100~~1 aliquots were subjected to SDS-PAGE and autoradiography as described above.
Protein Assay Protein
concentrations
were determined
by the Lowry method (24).
TABLE Summary
of the Activity
of the PK-A
in the Consecutive
I Steps of MPC
Purification
step
Homogenate Supernatant Ammonium sulfate fractionation DEAE-Sephacel chromatography (pH 7.5) Ultrogel chromatography (pH 8.3) DEAE-Sephacel chromatography (pH 8.3) Second Ultrogel chromatography (pH 7.5)
Protein hdml)
Control
CAMP
0.041 0.080
26.64 28.27
0.167
10.37
0.052 0.037 0.048 0.104
7.40 1.85 2.31 0.08
370 620
40.80 13.60
0.072 0.046
80 188 110 152
15.40
0.130 0.039
6
0.38 0.24
0.017
0.12
0.015
0.18
0.013
from
Control
Bovine
Pituitaries”
Sp act (U/mg)
Total U
U/ml Volume (ml)
Purification
CAMP
15.10 49.85 13.34 9.72
4.05 7.28 0.62
Control
0.002 0.003
CAMP
Total (%)
0.001
100
0.006
330 88 64 27 48 4
0.008
0.011
0.105 0.071 0.127 0.070
0.138 0.157 0.399
0.576
’ Activity was determined with kemptide as the substrate (0.130 mM) as described under Experimental Procedures; U (units) defined as the amount of enzyme that transfers one pmol of 32P to the substrate per hour, in the absence (control) and presence of 10 ticM CAMP.
PHOSPHORYLATION
OF BOVINE
MULTICATALYTIC
2
?a! 206 , 3i6 9 10 11
12 13 14 15
16 17
18 19 20 21 22
0.0
1.0
FIG. 2. A, Nondissociating PAGE of the purified MPC (8 pg). B, profiles of the proteolytic activity @) (absorbance at 540 nm X 3) and kinase activity (m) (cpm X 10m4) of the 4.5% gel.
purified MPC CAMP followed vealed a major consistent with Purification
with the affinity reagent [32P]-8-N3by SDS-PAGE and autoradiography reband at 55 kDa (Fig. 1, Lanes 2 and 3), the regulatory subunit of PK-A (23).
of MPC
The purity of the MPC preparation after the final AcA22 column was verified by PAGE. MPC migrated as a single band in nondissociating PAGE, when stained with either Coomassie blue (Fig. 2A) or silver stain (not shown). When the gels were sliced into 5-mm segments and both kinase and proteinase activities determined in
TABLE
PROTEINASE
COMPLEX
71
each segment, kinase activity could always be measured in slices with proteolytic activity (Fig. 2B). Kinase and proteinase activities were monitored throughout the purification procedure and, as shown in Tables I and II, CAMP-dependent phosphotransferase activity could be detected in all of the purification steps. Table I shows that three times more CAMP-dependent protein kinase activity was present in the supernatant fraction than in the homogenate. This could be due to removal of endogenous PK-A inhibitors. No PK-A activity could be detected in proteinase-containing fractions of the ion-exchange columns, due to the high EDTA concentrations of the gradient buffers. The MPC-containing fractions eluted from both ion-exchange columns at the end of the gradient (0.3 M Tris, 96 mM EDTA, pH 7.5, in the first and 0.4 M Tris, 70 mM EDTA, pH 8.3, in the second column). High EDTA concentrations chelated the magnesium ions present in the assay mixture, essential for phosphotransferase activity. To measure PK-A activity, the samples were dialyzed against a 0.01 k Tris-EDTA, pH 7.5, buffer, where the EDTA concentration was 2.7 mM. Purification of MPC in Tris-HCl buffers containing lower EDTA concentrations led to unacceptably low recoveries, excluding their practical use. The phosphotransferase activity of the second DEAE-Sephacel column was variable, sometimes being higher (as shown in Table I) other times lower (not shown) than in the preceding step, but it was always readily measured after dialysis. Although a 1700-fold purified MPC preparation was obtained (Table II), a significant amount of the CAMPdependent protein kinase activity remained associated with the proteinase as shown in the last gel filtration step (Table I and Fig. 3). Furthermore, CAMP-dependent phosphotransferase activity could also be measured in MPC-preparations obtained either from rabbit liver following the same purification method as in (ll), or from bovine pituitaries after the purification method of
II
Summary of the Purification of the MPC from Bovine
Pituitaries”
Activity Purification
step
Homogenate Supernatant Ammonium sulfate fractionation DEAE-Sephacel chromatography (pH 7.5) Ultrogel chromatography (pH 8.3) DEAE-Sephacel chromatography (pH 8.3) Second Ultrogel chromatography (pH 7.5)
Volume (ml)
Protein bx/mU
U/ml
Total U
370 620
40.80 13.60
0.400 0.260
148.0 161.2
15.40
1.440
115.2
0.38 0.24 0.12
0.280 0.420 0.210 3.080
52.6 46.2
80 188
110 152 6
0.18
Spec act W/md
Recovery (%:o)
0.010 0.019 0.094
31.9
0.747 1.787 1.750
18.5
17.111
n Activity was determined with Cbz-Gly-Gly-Leu-pNA as the substrate (0.4 mM) as described under Experimental defined as the amount of enzyme that cleaves one pmol of the substrate per hour.
Purification (fold)
100 109 78 36 31 22 12
Procedures.
1 2
10 76 182
179 1745
U (units)
‘72
PEREIRA
o.ol 0
AND
Cbz-Gly-Gly-Leu-pNA W)
58
28
Fraction
#
FIG. 3. Distribution of MPC activity (open circles) and the CAMPdependent protein kinase activity (closed circles) in the last gel filtration fractions. Proteolytic activity (0) is expressed in U/ml and kinase activity (0) is expressed in U/ml X 1O-5; (0) absorbance at 280 nm. Every third fraction collected (12 ml/fraction) was assayed for chymotrypsin-like and kinase activities.
(14) (Table III). The ratio (R) of the specific activities of PK-A and MPG in the three different preparations after the last step of purification was of the same order of magnitude (0.03-0.09), as can be seen in Table III. Phosphotransferase
Activity
WILK
of the MPC Preparation
The kinase activity copurifying with MPC was stimulated 8- to lo-fold by CAMP (10 PM) and was inhibited by 40 FM PK-A inhibitor (results not shown). No protein kinase C was present in our preparation since calcium (0.48 mM) and/or a phospholipid mixture (20 PM phosphatidyl-L-serine and 0.32 PM dioctanoyl glycerol) failed to stimulate any phosphotransferase activity.
TABLE
Chymotrypsin-Like
FIG. 4. Right ordinate (closed symbols), proteolytic activity (0) expressed in U/mg of protein X 10, as a function of the concentration of the Cbz-Gly-Gly-Leu-pNA substrate (abscissa). Left ordinate (open symbols), kinase activity (0) expressed as U/mg of protein X 10 measured in the presence of Cbz-Gly-Gly-Leu-pNA (abscissa). Data from a representative experiment.
The kinase activity remaining associated with MPC after the last step of purification was measured in the presence of the chromogenic MPC-peptide substrate, Cbz-Gly-Gly-Leu-pNA. A two- to threefold activation of the phosphotransferase activity was observed (Fig. 4). The sigmoid relationship between the reaction velocity (ordinate) versus the activator concentration (abscissa) is suggestive of an allosteric interaction. The proteinase substrate did not activate the PK-A previously resolved from the complex after the first DEAE-Sephacel column (results not shown). The activation of the copurifying PK-A by Cbz-GlyGly-Leu-pNA was not due to protection of the kinase substrate kemptide from proteolysis by MPC. Degradation of kemptide by MPC in the absence of Cbz-GlyGly-Leu-pNA was monitored by HPLC and was found
III
and Phosphotransferase Activities of Three Different MPC Preparations, after the Last Step of Purification” Phosphotransferase
Source and method of MPC preparation
Chymotrypsinlike activity
activity lo PM
Control
CAMP
R
17.1 U/mg
0.070 U/mg
0.576 U/mg
0.034
4.2 U/mg
0.043 U/mg
0.229 U/mg
0.055
3.5 U/mg
0.105 U/mg
0.316 U/mg
0.090
A. Bovine pituitaries,
after Orlowski
and Michaud
(1989). 100 g of tissue.
B. Rabbit livers, after Orlowski
and Michaud
(1989). 100 g of tissue.
C. Bovine pituitaries,
after McGuire
et al. (1989). 35 g of tissue.
’ Chymotrypsin-like activity was determined with the substrate Cbz-Gly-Gly-Leu-pNA as in Table II; phosphotransferase activity was determined with the substrate kemptide as in Table I; units for each activity as described in Tables I and II; the numbers represent specific activities; R, ratio of specific activities of PK-A (in the presence of cAMP)/MPC (measured as in Tables I and II).
PHOSPHORYLATION
OF BOVINE
MULTICATALYTIC
to be less than 10% during the period of incubation (results not shown). The relationship between concentration of Cbz-Gly-Gly-Leu-pNA and its hydrolysis by MPC is also shown in Fig. 4 for comparison. DISCUSSION
In the present studies, we show that at least two MPC subunits (S,, MW 28,800 and S3, MW 27,000) of the bovine pituitary preparation are strongly phosphorylated in a CAMP-dependent fashion by a PK-A that copurifies along with the 1700-fold-purified MPC. Two other subunits (S,, MW 31,000 and S5, MW 24,700) are also phosphorylated, but to a lesser degree. Reports by Haass and Kloetzel (26) on Drosophila MPC have shown that 28kDa subunits are substrates for phosphorylation. They suggest that this post-translational modification might be responsible for the development-dependent variation in subunit patterns of the MPC. In addition, cDNA cloning and sequencing studies of the 35kDa subunit of Drosophila MPC revealed a consensus sequence for tyrosine phosphorylation identical to sites present in lactate dehydrogenase and several viral particles (27). Since the amino acid sequence of the 35-kDa subunit demonstrated no significant homology with any of the known proteases, Haass et al. (27) postulated that this MPCcomponent was involved in activity regulation rather than enzymatic cleavage. Furthermore, Tanaka and his co-workers (28) have demonstrated that the C, subunit (MW 25,800) of the rat liver MPC contains a sequence of 70 amino acids, homologous to the autophosphorylation tyrosine residues of various cellular tyrosine kinases such as src gene products and some receptor proteins. Taken together, the above findings suggest that MPC phosphorylation might be a mechanism for regulating its proteolytic activities. In vitro studies of other enzymes show their activities to be regulated by phosphorylation. For example, Drosophila topoisomerase II (an enzyme producing cleavage and resealing of supercoiled DNA) undergoes rapid phosphorylation by protein kinase C, resulting in the stimulation of its activity (29). Moreover, purified preparations of Drosophila topoisomerase II contain an endogenous protein kinase, and the topoisomerase is a phosphoprotein (30). Monkey brain aminopeptidase activity was inhibited by CAMP-dependent phosphorylation (31). Tyrosine hydroxylase from bovine adrenal cells (32) was phosphorylated and activated in vitro, by at least three protein kinases, namely PK-A, protein kinase C, and calcium/calmodulin-dependent protein kinase (32). The PK-A, which copurifies with the bovine pituitary MPC, has a CAMP-binding regulatory subunit of 55 kDa and is stimulated 8- to lo-fold by CAMP (10 PM), inhibited by a synthetic PK-A inhibitor, affinity labeled with 8-Azido-CAMP, and unaffected by protein kinase C activators. Its specific activity in the presence of CAMP, cal-
PROTEINASE
COMPLEX
73
culated to be 0.576 U/mg of protein (U = pmol of 32P transferred/h) is 5.6-fold less than the specific activity reported for PK-A type II purified from bovine brain (3.26 U/mg of protein) (33). Since the bulk of the protein in our preparation is contributed by MPC, the lower kinase specific activity is expected. Our evidence suggests that the copurifying PK-A activity resides in a polypeptide other than the MPC, since its regulatory component (visible only by radiolabeling with 32P) has a greater MW (55,000) than any of the MPC subunits. The phosphorylation of the MPC subunits by the copurifying kinase, and the two- to threefold stimulation of the PK-A by a synthetic MPC substrate, is suggestive of a functional interaction between the two systems. Furthermore, we show that several fractionation techniques resolving proteins according to different molecular properties only partially separated the PK-A from the MPC. PK-A copurified with the complex from rabbit liver following the same purification procedure (ll), or from bovine pituitaries using a different isolation method (14), indicating that the copurification of the two enzymes is not species or tissue specific, or dependent on a single method of purification. This may indicate that in vivo the two protein species are tightly associated. Recent studies have indicated that MPC is associated with at least two other factors in a larger 26 S complex. This macromolecular complex degrades ubiquitinated proteins, with MPC postulated to be the catalytic core (34-36). Furthermore, the assembly of the three components of the 26 S complex was implied to be ATP dependent (34-36). Ganoth et al. (37) have indicated that, if ATP is required for complex formation, a protein kinase could be one of the three components of the 26 S system. The phosphorylation of MPC subunits by PK-A and/ or tyrosine kinase represents a potential mechanism for controlling its activity in response to extracellular stimuli. The identity of the MPC phosphorylation sites for PK-A requires future analysis. Studies currently under investigation in our laboratory will determine if the state of MPC phosphorylation in the bovine pituitary will affect its proteolytic activity, and/or its incorporation into a larger ATP-dependent proteolytic system, the 26 S complex. REFERENCES 1. Kopp, F., Steiner, R., Dahlmann, B., Kuehn, L., and Reinauer, (1986) B&him. Biophys. Actu 872,253-260.
H.
2. Rivett, A. J. (1989) Arch. Biochem. Biophys. 268, 1-8. 3. Wilk, S., and Orlowski, 4. Orlowski, M., and Wilk, mun. 101,814-822. 5. Orlowski,
M. (1980) J. Neurochem. S. (1981) B&hem.
35, 1172-1182.
Biophys.
Res. Com-
M., and Wilk, S. (1988) Biochem. J. 255, 750-751.
6. Arrigo, A. P., Tanaka, K., Goldberg, Nature (London) 331,192-194.
A., and Welsh, W. J. (1988)
74
PEREIRA
7. Falkenburg, P. E., Haass, C., Kloetzel, P. M., Niedel, B., Kopp, F., Kuehn, L., and Dahlmann, B. (1988) Nature (London) 331,190192. 8. Rivett, A. J. (1989) J. Biol. Chem. 264, 12,2X-12,219. 9. Wilk, S., and Orlowski, M. (1983) J. Neurochem. 40,842-849. 10. Orlowski, M., Michaud, C., and Wilk, S. (1980) Biochem. Biophys. Res. Commun. 94,1145-1153. 11. Orlowski, 9278.
M., and Michaud,
12. Ray, K., and Harris, 7545-7549. 13. Rivett,
C. (1989) Biochmistry
28, 9270-
H. (1985) Proc. Natl. Acad. Sci. USA 82,
A. J. (1985) J. Biol. Chem. 260,12,600-12,606.
14. McGuire, M. J., McCullogh, M. L., Croall, D. E., and DeMartino, G. N. (1989) Biochim. Btiphys. Acta 996,181-186. 15. Tanaka, K., Ii, K., Ichihara, A., Waxman, (1986) J. Biol. Chem. 261,15,197-15,203.
L., and Goldberg,
16. Dahlmann, B., Rutschmann, M., Kuehn, (1985) Biochem. J. 228,171-177.
A. L.
L., and Reinauer,
H.
17. Tanaka, K., Yoshimura, T., Ichihara, A., Ikai, A., Nishigai, Morimoto, Y., Sato, M., Tanaka, N., Katsube, Y., Kameyamo, and Takagi, T. (1988) J. Mol. Biol. 203,985-996.
M., K.,
18. Shenolikar, 19. Wilk, 464.
S. (1988) FASEB
J. 2,2753-2764.
S., Pearce, S., and Orlowski,
M. (1979) Life Sci. 24, 457-
20. Roskoski, R., Jr. (1983) in Methods in Enzymology (Corbin, J. D., and Hardman, J. G., Eds.), Vol. 99, pp. 3-6, Academic Press, San Diego. 21. Laemmli,
U. K. (1970) Nature
(London)
227,680-685.
22. Cheng, H. C., Kemp, B. E., Pearson, R. B., Smith, A. J., Misconi, L., Van Patten, S. M., and Walsh, D. A. (1986) J. Biol. Chem. 261,
989-992.
AND
WILK
23. Walter, U., and Greengard, P. (1983) in Methods in Enzymology (Corbin, J. D., and Hardman, J. G., Eds.), Vol. 99, pp. 154-162, Academic Press, San Diego. 24. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275. R., and Rosen, 0. M. (1976) J. Biol. Chem. 251, 25. Rangel-Aldao, 7526-7532. 26. Haass, C., and Kloetzel, P. M. (1989) Erp. CelZRes. 180.243-252. B., Multhaup, G., Beyreuther, K., and 27. Haass, C., Pesold-Hurt, Kloetzel, P. M. (1989) EMBO J. 8,2373-2379. 28. Tanaka, K., Fujiwara, T., Kumatori, A., Shin, S., Yoshimura, T., Ichihara, A., Tokunaga, F., Aruga, R., Iwanaga, S., Kakizuka, A., and Nakanishi, S. (1990) Biochemistry 29,3777-3785. 29. Sahyoun, N., Wolf, M., Besterman, J., Hsieh, T., Sander, M., Levine III, H., Chang, K., and Cuatrecasas, P. (1986) Proc. Natl. Acad. Sci. USA 83,1603-1607. 30. Sander, M., Nolan, J. M., and Hsieh, T. (1984) Proc. N&l. Acad. Sci. USA8 1,6983-6942. 31. Ramamoorthy, S., and Balasubramanian, A. (1989) Biochem. J. 258,777-783. 32. George, J. R., Haycock, J. W., Johnston, J. P., Craviso, G. L., and Waymire, J. C. (1989) J. Neurochem. 52,274-284. 33. Miyamoto, E., Petzold, G. L., Kuo, J. F., and Greengard, P. (1973) J. Biol. Chem. 248,179-189. 34. Matthews, W., Tanaka, K., James, D., Ichihara, A., and Goldberg, A. L. (1989) Proc. Natl. Acad. Sci. USA 86,2597-2601. 35. Eytan, E., Ganoth, D., Armon, T., and Hershko, A. (1989) Proc. Natl. Acad. Sci. USA 86,7751-7755. 36. Driscoll, J., and Goldberg, A. L. (1990) J. Biol. Chem. 266,47894792. 37. Ganoth, D., Leshinsky, E., Eytan, E., and Hershko, A. (1988) J. Biol. Chem. 263,12,412-12,419.
