Dec 20, 1982 - Yih-Shyun E. Cheng and Lan Bo Chen. From the Sidney Farber &ncer Institute and Department of Pathology, Harvard Medical School, Boston, ...
THEJOURNAL OF BIOLOGICAL CHEMISTRY Vol. 258, No. 12, Issue of June 25, pp. 7852-7857,1983 Printed m U.S A.
Characterization of a Phosphotyrosyl Protein Phosphatase Activity Associated with a PhosphoserylProtein Phosphatase ofM r = 95,000 from Bovine Heart* (Received for publication, December 20, 1982)
Jonathan ChernoffS and Heng-Chun Lip From the Departmentof Biochemistry, Mount Sinai School of Medicine of the City Universityof New Yo&, New York,New York 10029
Yih-Shyun E. Cheng and LanBo Chen From the Sidney Farber &ncer Institute and Department of Pathology, Harvard Medical School, Boston, Massachusetts02115
A cytosolic phosphoproteinphosphataseof M, = other retroviruses(1, 14). Normal cells also possess Tyr95,000 purifiedfrombovinecardiacmuscle,which specific protein kinase activity, although this activity is excontains a catalytic subunit of M, = 35,000, is known pressed a t much lower levels than in retrovirally transformed to be associated with a Mg2+-activatedp-nitrophenyl cells (11, 14). This Tyr-proteinkinase activity is highly conphosphatase activity. We have found that the enzyme served throughout evolution (15, 16), suggesting that phospreparation is also active toward phosphotyrosyl-IgG phorylation of Tyr residues in protein may be involved in and -casein phosphorylated bypp60"", the transform- certain essential cellular processes. ing gene product of Rous sarcoma virus. The properties In previous communications (17-20), we have reported that of this phosphotyrosylproteinphosphatase activity adivalentcation-dependent P N P phosphataseactivityis closely resemble those of the p-nitrophenyl phospha- tightly associated with a phosphoseryl (P-Ser) protein phostase activity but sharply differ from those of the phos- phatase of M , = 35,000 and its high molecular weight derivaphorylase phosphataseactivity. Comparative studies of the activities of the M , = tives have been purified from cardiac muscle (17, 21) and 95,000 phosphatase, bovine kidney alkaline phospha- several other tissues (18, 19). These two activitiesexhibit distinctly different catalytic properties. In this communicatase, and ATP. Mg-dependentphosphatasetoward tion, we show that the cardiac M , = 95,000 phosphoprotein phosphoseryl,phosphothreonyl,andphosphotyrosyl phosphatase is active toward P-Tyr proteins and that the proteins and p-nitrophenyl phosphate under various conditions have been carried out. Theresults indicate properties of this P-Tyrprotein phosphatase activityresemble that the M, = 95,000 enzyme exhibits higher activity those of the PNP phosphatase and differ from those of the P-Ser protein phosphatase. towardphosphoserylandphosphothreonylproteins than toward phosphotyrosyl proteins, while the kidney EXPERIMENTALPROCEDURES alkaline phosphatase preferentially dephosphorylates phosphotyrosylproteins. ATP. Mg-dependentphosMaterials phatase is inactive toward phosphotyrosyl proteins.
Many sarcomagenic retroviruses recently have been shown to contain onc genes coding for transformation-specific proteins which possess protein kinaseactivity (1-10). These protein kinases specifically phosphorylate tyrosine residues; a modification hitherto unknown (9, 11-13). It has been proposed that unregulated Tyr phosphorylation may be closely related to the establishment andmaintenance of malignant transformationin cells infected with RSV,' as well as
* This work wassupported by United States Public Health Service National Institutes of Health Grant HL-22962 and by American Cancer Society Grant CD92A. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $Trainee on Medical Scientist Training Grant GM-07280 from the National Institutes of Health. § To whom correspondence should be addressed. The abbreviations used are: RSV, Rous sarcoma virus; P-Tyr-, P-Ser-, P-Thr-protein, phosphotyrosyl, phosphoseryl, phosphothreonyl protein; PNP, p-nitrophenyl; PNPP, p-nitrophenyl phosphate; SDS, sodium dodecyl sulfate; EGTA, ethylene glycol his(@aminoethyl ether)-N,N,N',N'-tetraacetic acid; MES, 2-(N-morpho-
The catalyticsubunit of CAMP-dependent protein kinase was purified from bovine heart according to Ref. 22. Inhibitor 1 was purified from rabbit skeletal muscle (23). Phosphorylase kinase and phosphorylase b (2X crystallized) were purchased from Sigma. Antisera specific for pp60"~"" wereprepared by inducing tumors in newborn rabbits with RSV, Schmidt-Ruppin strain, group D, according to Ref. 24. For preparation of immunoaffinity resin specific for pp60"-"", IgG was partially purified from the antisera by 50% ammonium sulfate precipitation, followed by coupling to CNBr-activated Sepharose 4B (Pharmacia) according to Ref.13. The pp60"~""was partially purified from a RSV-transformed rat cell line, AnAn (251, by the following procedures. AnAn cells were suspended in a buffer containing 10 mM KPi, pH 7.0, 40% glycerol, 0.02%Nonidet P-40, 2 mM EDTA, 1 mM EGTA, 100 mMKC1, 2 mM 2-mercaptoethanol, and 1%aprotinin (Sigma). The cells were disrupted in a Dounce homogenizer and thesuspension was centrifuged for 30 min at 30,000 X g. The supernatant was filtered through glass wool and applied to an aminobexyl agarose (Sigma) column equilibrated in Buffer A (5 mM KP, pH 7.0, 20% glycerol, 0.05% Nonidet P-40, 1 mM EDTA, and 1 mM 2-mercaptoethanol). The column waswashed with 5 volumes of Buffer A plus 0.4 M KC1 and thepp60"-"" waseluted by 2 column volumes of Buffer A plus 1 M KCI. The enzyme solution was extensively dialyzed against Buffer A and applied to a DEAE-Sephacel (Pharmacia) column equilibrated with the same buffer. The column was extensively washed with Buffer A and the pp60"-""was
1ino)ethanesulfonic acid; F,, ATP. Mg-dependent Phosphatase; FA, protein activator of F,.
7852
7853
Phosphotyrosyl Protein Phosphatase
rylase a, and10 mM MgCl,. Low specific activity phosphorylase a was used for all studies unless otherwise indicated. Reactions were initiated by the addition of phosphatase and terminated by the addition of 5 pl of 25% trichloroacetic acid. Incubation time was adjusted such that no more than 20% of the substrate was dephosphorylated. The 32P,released was separated from 32P-proteinby a paper chromatographic method as described previously (21). The amount of M , = 95,000 enzyme used was about 20-fold greater when measuring PTyr-IgG phosphatase than phosphorylase phosphatase activity. Preparation of 32P-Proteins p-Nitrophenyl phosphatase activity was measured by the release [3ZP]Tyr-IgGwas prepared as follows. The pp60"~""in AnAn cells of p-nitrophenol from p-nitrophenyl phosphate as described previwas immunocomplexed with the antisera and collected on protein A- ously (17). The standard assay mixture (0.5 ml) contained 50 mM Sepharose beads (Pharmacia) (25). Phosphorylation of the heavy Tris-HC1, pH 8.6, 1 mM dithiothreitol, 20 mM PNPP, and 20 mM chain of IgC was carried out by incubating the protein A-Sepharose MgC1,. The reaction was initiated by the addition of enzyme, incubated at 30 "C for 30 min, and terminated by the addition of 0.5 ml beads (1.5 ml wet volume) with 20 mM Tris-HC1, pH 7.2,5 mM M&h, and 1 mCi of [Y-~'P]ATP(3000 Ci/mmol, New England Nuclear) at of 1 M Na2C03. The absorbance at 410 nm of the mixture was 20 "C for 20 min. The protein A-Sepharose beads were then washed measured spectrophotometrically by using a control lacking enzyme repeatedly with RIPA buffer (20 mM Tris-HC1, pH 7.2, 0.15 M NaC1, as a blank (the exctinction coefficient for p-nitrophenolate anion: 1.75 X 10' M" cm"). 1% Triton X-100, 0.1% SDS, and 1%deoxycholate) to remove unOne unit of phosphatase activity was defined as the amount of bound radioactivity. The 32P-labeledcomplexes were separated from protein A-Sepharose by extracting with a small volume of 0.005 M enzyme catalyzing the release of 1rmol of Pi/min for PNP phosphatase activity, and the release of 1 fmol of Pi/min for phosphoprotein HCl and dialyzed against 20 mM Tris-HC1, pH 7.2. [32P]Tyr-casein was prepared by phosphorylation of a-casein phosphatase activity. Polyacrylamide Gel Electrophoresis-Electrophoresis on a 7% pol(Sigma) with the partially purified pp60"-"". The reaction was carried out at 20 "C for 60 min in an incubation mixture containing 0.1 M yacrylamide gel was carried out by the procedure of Davis (29). MES, pH 6.5, 1 mM MnCIZ, 0.4 p M [y3'P]ATP (3000 Ci/mmol), a- Protein was stained with Coomassie brilliant blue. For localizing casein (1.0 mg/ml), and ~ ~ 6 0 (70 " " pg/ml). ~ The reaction was ter- phosphatase activity, gels weretransversely sliced into 1-mm sections minated by the addition of an equal volume of 25% trichloroacetic and each slice was placed in 100 ~1 of 50 mM Tris-HC1, pH 7.4, 10 acid. The precipitated [32P]caseinwas extensively washed with 25% mM 2-mercaptoethanol, 10 mM MgCI,, 50 mMKC1, and 10% glycerol trichloroacetic acid and suspended in asmall volume of distilled HZO. for the extraction of the enzymatic activity. PNP phosphatase activity The suspension was then adjusted to pH 7.0 by 1 N NaOH followed staining was done in 50 mM Tris-HC1, pH 8.6,lOO mM M&12,20 mM PNPP, 1 mM dithiothreitol, and 0.2 M CaC12. After incubation at by extensive dialysis against 20 mM Tris-HCI, pH 7.2. [32P]Thr-inhibitor1 and [32P]Ser-caseinwere prepared as follows. 30 "C for a suitable length of time, aband of white calcium phosphate The phosphorylation reactions was carried out at 30 "C for 30 min in appeared on the gel, givingthe location of the enzymatic activity. Polyacrylamide gel electrophoresis in the presence of SDS was an incubation mixture containing 50 mM Tris-HC1, pH 7.0, 5 mM MgCl,, 0.5 unit/ml of the catalytic subunit of CAMP-dependent carried out according to Laemmli (30). 32P-labeled proteins were protein kinase, 0.4 p~ [Y-~'P]ATP(3000 Ci/nmol), 0.5 mg/ml of located by autoradiography with the aid of Kodak Lanex intensifying inhibitor 1 or 1 mg/ml of a-casein. The reactions were terminated screens. Phsphoamino Acid An~lysis-~~P-labeledproteins were hydroandthe 32P-proteins were washed as described in the preceding lyzed in 6 N HCl for 2 h and subjected to thin layer electrophoresis paragraph. a t pH 3.5 according to the procedure of Hunter and Sefton (11). [32P]Ser-phosphorylase a oflow specific radioactivity (0.15 Ci/ Other Procedures-Protein concentration was determined by the mmol Pi) wasprepared as reported previously (25). [32P]Phosphorylase a of high specific radioactivity (3000 Ci/mmol Pi) was prepared method of Lowry et al. (31) following trichloroacetic acid precipitaby incubating crystalline phosphorylase b (1 mg/ml) with 50 mM tion. Bovine serum albumin was used as a standard. Tris-HCI, pH 8.6, 5 mM MgCl,,0.2 mM CaC12,0.4 ~ L M[y3'P]ATP RESULTS (3000 Ci/mmol), and 1.2 units of phosphorylase kinase. After incuActivation of P-Tyr-IgG and p-Nitrophenyl Phosphatase bating at 30 "C for 30 min, 5 volumes of cold, saturating (NH.),SO, solution, pH 7.0, were added to the reaction mixture. [32P]Phospho- Activities by MgC12-As shown in Fig. 1, the M , = 95,000 rylase a was collected by centrifugation a t 4 "C, dissolved in a small h volume of 20 mM Tris-HC1, pH 7.0, 10 mM 2-mercaptoethanol and reprecipitated by 5 volumes of the (NH4),S04solution. This washing mI 6 r 0 I process was repeated five times and the[32P]phosphorylasea obtained was dialyzed against 20 mM Tris-HC1, pH 7.0, 1.0 mM 2-mercaptoethanol and stored at 4 "C. Phosphoamino acid analysis of the 32P-proteinsshows that phos, Phlase a phorylation occurs exclusively at the specific amino acid residue as 4 indicated. eluted by 2 column volumes of Buffer A plus 0.2 M KCl. The enzyme solution was immediately applied to animmunoaffinity column. The column was washed with 5 volumes of Buffer A plus 1 M KC1 and the pp60u-" was eluted with 2 column volumes of Buffer A plus 0.2 M KC1 and 1.5 M KSCN. The enzyme was immediately dialyzed against Buffer A for 2 h followed by Buffer A plus 30% glycerol and stored at -20 "C. The pp60"-'" so obtained isfree from protein kinase activity which phosphorylates Ser or Thr residues.
I
Preparation of Phosphoprotein Phosphatases Phosphoprotein phosphatase of M , = 95,000 was purified from bovine cardiac muscle as previously described (20). The purification 2 procedures include (NH4),S04 fractionationand DEAE-cellulose and Sephacryl S-200 chromatography. Evidence indicates that theenzyme contains a catalytic subunitof M. = 35,000 (20). The phosphoprotein phosphatase of M, = 35,000was purified from rabbit liver by a procedure involving treatment of the enzyme with 80% ethanol as 0 previously described (19). F, and FAwere purified from bovine heart Y by procedures similar to those for purification of F, and FA from B rabbit skeletal muscle (27, 28). Bovine kidney alkaline phosphatase 2 0 10 20 30 40 0 was from Worthington. ( Mg C l p ) . m M Enzyme Assay-Phosphoprotein phosphatase assays were measured at 30 " c , in an incubation volume of25 pl, containing 50 mM FIG. 1. Effects of MgClz on the activities of the cardiac Tris-HC1, pH 7.0 or 8.6, 1 mM dithiothreitol, 0.2-0.25 nM 32P-protein phosphatase of M. = 95,000. The enzymatic activities toward (intermscontent),and 10 mM MgCI,. In experiments utilizing phosphorylase a (Phlase a) (O), P-Tyr-IgG (O), and p-nitrophenyl low specific activity phosphorylase a, phosphoprotein phosphatase phosphate (x) were measured under standard assay conditions as activity was measured in an incubation volume of25 pl, containing described under "Experimental Procedures" except that MgC12 con50 mM Tris-HC1, pH 7.0, 1 mM dithiothreitol, 10 PM [32P]phospho- centrations were varied as indicated.
-
7854
Phosphatase ProteinPhosphotyrosyl
phosphatase preparation is active toward phosphorylase a in the absence of added divalent cation, but shows little, if any, activity toward either P-Tyr-IgG or PNPP. When Mg2+ is added to the assay mixture, the enzymatic activity toward phosphorylase a is slightly inhibited, while those toward PTyr-IgG and PNPP become activatedinaconcentrationdependent manner. Both the saturation curves and the K,,, values for M$+ (about 13 mM) for these two activities are similar. pH Optima-Fig. 2 shows that the pH activity profiles of the P-Tyr-IgG and P N P phosphatase activities are similar, and thatboth activities have an optimum around pH 8.5-9.0. In contrast, phosphorylase phosphatase activity, measured either in the presence of 2 mM EDTA (Fig. 2) or 20 mM M&12 (data not shown), has an optimal pH around 7.5. It should be
noted that, in the absence of M$+, no activity toward P-TyrIgG (Fig. 2) or PNPP (data not shown)can be detected throughout the range of pH studied. Thermal Stability-As shown in Fig. 3, the phosphorylase phosphatase activity is relatively stable at 40 “C.Both the PTyr-IgG and the PNP phosphatase activities, however, are rapidly inactivated in a parallel fashion at this temperature. Within 30 min, the phosphatase becomes completely inactive toward these substrates. Effects of Phosphatase Inhibitors-Fig. 4 shows the effects of increasing Pi concentrations onphosphorylase, P-Tyr-IgG, and PNPphosphatase activities. Although all these activities areinhibited by millimolar levels of Pi,the phosphatase activities toward P-Tyr-IgG and PNPP are inhibited to a much greater extentthanthe phosphorylase phosphatase activity. Moreover, the degree of inhibition of P-Tyr-IgG and PNPP dephosphorylation is nearly identical at all levels of Pi concentration. The concentrations of Pi required for 50% inhibition of phosphatase activity is approximately 1 mM in the cases of P-Tyr-IgG and PNPP and approximately 7 mM in the case of phosphorylase a. Table I shows the effects of various compounds on the 100
....-r
B 5
I
I
6
7
a ap I
I
1
8
(P-ser) 50
0
-e-e-*1
Phlase a
9
1
0 [ P I IgG
0
PH
I
FIG. 2. Effects of pH on the activities of the cardiac phosphatase of M. = 95,000. The enzymatic activities toward phosphorylase a (Phlase a) (with 2 mM EDTA, 0, M), P-Tyr-IgG (with 2 mM EDTA, 9; 20 mM MgCl,, 0, O), and p-nitrophenyl phosphate (with 20 mM MgC12, A, A)were measured as described under “Experimental Procedures” except that 50 mM of the following buffers a t the indicated pH were used Tris-HC1 ( Q O , 9, and A) and imidazole (M, 0, and A).
0
(P-Tyr
1
I
I
5
10
[Pi], m M
FIG. 4. Effects ofPi concentration on phosphatase activities toward phosphorylase a,P-Tyr-IgG, and p-nitrophenyl phosphate. The enzymatic activities were measured as described under “Experimental Procedures,” in the presence of the indicated amounts of Pi. Enzymatic activity is given as a percentage of the activity in the absence of Pi (phosphorylase phosphatase, 0; P-Tyr-IgG phosphatase, 0;p-nitrophenyl phosphatase, X). Phlase a, phosphorylase a.
TABLEI Effect of various agents on cardiac M, = 95,000 phosphatase and kidney alkaline phosphatase activities toward phosphorylase a, PTyr-IgG, and p-nitrophenyl phosphate Aliquots of the cardiac or the kidney phosphatase were incubated with either phosphorylase a, P-Tyr-IgG, or p-nitrophenyl phosphate as described under “Experimental Procedures.” Enzymatic activity in the presence of the various reagents is given as a percentage of the activity in the absence of additions, each the average of duplicates for three determinations. Activity C a r d i a p c h P o ’ ~ ~ ~ ~ ~ t eKidney in alkalinephosphatase 25 PreincubationTime(rnin
0
50
)
FIG. 3. Thermal stability of the cardiac phosphatase of M, = 95,000. The enzyme was preincubated a t 40 “C in a solution containing 20 mM Tris-HC1, pH 7.4, 2 mM 2-mercaptoethanol, 10% glycerol, and 2 mg/ml bovine serum albumin. At the indicated time intervals, aliquots were withdrawn for the determination of enzymatic activities toward phosphorylase a (Phlasea) (in the presence of 2 mM EDTA, O), P-Tyr-IgG (in the prsence of 20 mM MgC12, 0),and pnitrophenyl phosphate (inthe presence of20mMMgCl,, X), as described under “Experimental Procedures.”
Addition
PhosPNPP phory- phorylase a 125 50 p~ ZnClz 2 mM PNPP 34 100 67 5 mM PNPP 50 mM NaF 0.7 34 98 5 mM EDTA 90 66 2 mM PPi
78
PhosPNPP
lase a % control
61149
19 2.5
1 0
137
116 24 8 60 2.4 115
56 3.7 112
7855
Phosphotyrosyl Phosphatase Protein activities of the Mr = 95,000phosphatase preparation from bovine cardiac muscle and of commercially obtained alkaline phosphatase from bovine kidney. The M , = 95,000preparation exhibits markedly different degrees of sensitivity to Zn2+, EDTA, and F- when assayed against phosphorylase a on the one hand, and P-Tyr-IgG and PNPP on the other. In general, the P-Tyr-IgG and the PNP phosphatase activities are more sensitive to inhibition by Zn2+, EDTA, and F-, whereas the phosphorylase phosphatase activity is more sensitive to inhibition by PPi. When assayed in the presence of each effector, bovine kidneyalkaline phosphatase behaves similarly toward P-Tyr-IgG and PNPP and this behavior does not resemble that of the cardiac enzyme (Talle I). For example, both the P-Tyr-IgG and the PNP phosphatase activities associated with the kidney enzymeare slightly stimulated in the presence of 50 PM Zn2+and areinhibited about 40% in the presence of 50 mM F-. By contrast, these two activities associated with the cardiac enzyme are inhibited about 30 and 99% in the presence of the same concentrations of Zn2+and F-, respectively. Table I also shows that PNPPis a potent inhibitor of the P-Tyr-IgG phosphatase activity associated either with the cardiac or the kidney phosphatase. Furthermore, the P-TyrIgG phosphatase activity is more sensitive to inhibition by PNPP than the phosphorylase phosphatase activity in the cardiac muscle enzymepreparation. Co-migrationof the P-Tyr-IgG and thep-Nitrophenyl Phosphatase Activities on Polyacrylamide Gel Electrophoresis-In previous communications, we have demonstrated that the PNP phosphatase activity co-purifies with the M , = 35,000 phosphatase (17-19)and theMr = 95,000 phosphatase which contains a Mr = 35,000 catalytic entity (20).We have examined the P-Tyr-IgG phosphatase activity in the processes of purification of the Mr= 95,000phosphatase from bovineheart and the Mr = 35,000 enzymefrom rabbit liver (32). The results indicate that the P-Tyr-phosphatase activity co-purifies with the PNP and the phosphorylase phosphatase activities throughout various separation processes including ammonium sulfate fractionation, ethanol treatment, DEAE-cellulose and gel filtration chromatographies, and thepolyacrylamide gel electrophoresis. Fig. 5 shows the results of polyacrylamide gel electrophoresis of a typical Mr = 95,000phosphatase preparation. The data indicate that the enzymatic activities are separated into a single active peak of low mobility and doublet active peaks of high mobility. The enzymatic activity toward P-Tyr-IS; coincides with those toward
PNPP and phosphorylase a in either the low mobility or the high mobility doublet bands. The activity profile shown in Fig. 5 reflects the fact that the highly purified Mr = 95,000 phosphatase tends to undergo partial dissociation on polyacrylamide gel electrophoresis (20). As previously reported (201, when the proteins in the low mobility band are extracted from the gel and re-electrophoresed on polyacrylamide inthe presence of SDS, two protein bands, corresponding to Mr= 63,000 and 35,000,are observed. Similar experiments on the high mobility doublet active bands result in a single protein band of M , = 35,000on SDS-gel electrophoresis. It has been postulated that the Mr = 95,000phosphatase consists of a catalytic subunit of M , = 35,000and anoncatalytic subunit of Mr = 63,000 (20). Regardless of the precise subunit composition of this enzyme, the present results clearly demonstrate that the activity toward P-Tyr-IgG is tightly associated with those toward PNPP and phosphorylase a. Substrate Specificity-In order to gain more understanding
B m
-
:QB
r2 P0 E
-2 c
-
-0.4 ’
’ ;
n
L
P %
sQ
P
2
0 c ” n
0 0
20
40
Slice
i
60
Number
u
FIG. 5. Polyacrylamide gel electrophoresis of the
M, =
95,000 phosphatase. Native gel electrophoresis was carried outas
onto each of three 7% polyfollows. 10 pgofenzymewereloaded acrylamide gels. Following electrophoresis, one was stained for protein (A), a second was stained for alkaline phosphatase (B), and a third was sliced and assayed for P-Tyr-I& phosphatase activity as described under “Experimental Procedures.” The dye front is indicated by the wires and the arrow.
TABLEI1 Substrate specificities of various phosphatases The activities of thebovine cardiacM,= 95,000phosphatase (1mg/ml), ATP-Mg-dependent phosphatase (0.08 mg/ml)andkidneyalkalinephosphatase (2.5 mg/ml)weremeasured at pH 7.0 or 8.6 as describedunder “Experimental Procedures” with various 32P-proteins andp-nitrophenyl phosphate at the indicated concentrations. The M, = 95,000 enzyme was assayed in the presence of 10 mMMgC12 or 0.5 mMMnC12 as indicated. F, was activated bypreincubatingwithsaturatingconcentrationsof FA, 0.1 mM ATP, and 0.5 mM MgC12,prior to measuring its activity. The kidney alkaline phosphatase was measured in the presence of 10 mM MgCI,. The amount of enmme used was adiustedso that no more than 20% of the substrate was deDhosDhorvlated. Activity
MgCL (10 mM) pH 7.0
[32P]Tyr-IgG 48(0.56nM) 725 154 47 [32P]Tyr-casein(0.40 690 nM) 1,870 570 164 [32P]Ser-casein 12 228 (0.44nM) 40819.400 26,800 6,1004,300 [32P]Phosphorylasea (0.56nM) 1.5 1,104 1,872 26,837 43,071 5,799 5,920 32P-Inhibitor1 (1.121.6 nM) 7.42,644 1,468 42,785 119,006 24,173 1,573 PNPP (4mM) 47 28 0
pH 8.6
Bovine kidney alkaline phosphatase
F,
M,= 95,000 enzyme
Substrate
MnCh (0.5 mM) pH 7.0
pH 8.6 unita/ml
PH 7.0
pH 8.6
0 0
130 29 0
pH 7.0
pH 8.6
150 83 6 4
127
0
1.1 0
0.2
7856
Phosphatase ProteinPhosphotyrosyl
concerning the specificity of the phosphatase, 32P-proteins specifically labeled at Tyr,Ser, or Thr residues were used as substrates to study the enzymatic activity in various conditions. The concentrationsof 32P-proteinsused are all in the nanomolar range because of limited availability of [32P]Tyrproteins. For comparison, the activities of F, purified from bovine heart and the commercial alkaline phosphatase from bovine kidney were also examined. As shown in Table 11, the effects of pH on the activity of the M , = 95,000 enzyme toward [32P]Tyr-proteinsis dependent on the divalent cation species present in the reaction mixture. In the presence of 0.5 mM Mn2+, the enzyme exhibits much higher activity at pH 7.0 than at 8.6. When 10 mM M P is substituted for Mn2+, the opposite effects of pH on the enzyme activity are observed. The dataalso indicate that Mn2+is a more effective activator than Mg2+ and that [32P]Tyr-casein is better a substrate than [32P]Tyr-IgG.It should be noted that, in the absence of added divalent cation, the M , = 95,000 enzyme preparation shows no detectable activity toward [32P]Tyr-proteins orP N P P a t either pH 7.0 or 8.6 (data not shown). The enzyme, however, dephosphorylates [32P]Ser-caseinand [32P]phosphorylasea at similar rates when measured either in the absence or presence of M$+. Thus, the P-Tyr-proteinphosphatase activity in the M , = 95,000 preparation is indeed distinctly different from the P-Ser-protein phosphataseactivity. When the rates of dephosphorylation of P-Ser(Thr)-proteins and P-Tyr-proteinsby the M , = 95,000 enzyme preparation are compared, the data indicate that the P-Ser(Thr)proteinphosphataseactivityis much higher thanP-Tyrprotein phosphatase activity in all conditions examined (Table 11). The dataalso show that F,, activated by preincubating with ATP. Mg2+ and FA, dephosphorylates both P-Ser- and P-Thr-proteins efficiently. It is inactive, however, toward P N P P or P-Tyr-proteins. Whenmeasured in the presence of M e , the kidney alkaline phosphatase preferentially dephosphorylates P-Tyr-proteins. Furthermore, theenzyme is much more active toward P-Tyr-proteins at pH 7.0 than at 8.6, while it hydrolyzes P N P P faster a t alkaline conditions. These data are consistent with the finding of Swarup et al. (33). When the P-Tyr-protein phosphatase activities of the M , = 95,000 preparation and the kidney alkaline phosphatase are compared, the data indicate that these two enzymes preferentiallydephosphorylate[32P]Tyr-casein and [32P]Tyr-IgG, respectively.
pp60"-"". In this paper, we present the detailed characterization of the P-Tyr-protein phosphateactivity in aM , = 95,000 phosphatase containing a Mr = 35,000 catalytic subunit (20). The results clearly indicate that theproperties of this activity are similar to those of the PNPphosphatase activity but are distinctly different from those of the phosphorylase phosphatase activity. The P-Tyr-protein and P N P phosphatase activities exhibit similar divalent cation specificities, pH activity profiles, thermal stabilities, and sensitivities to inhibition by Pi and other effectors, and mobilities on polyacrylamide gels. The data suggest that this PNP phosphatase activity represents anactivity toward P-Tyr-proteins. Swarup et al. (33) have examined the ability of various membrane-bound alkaline phosphatases to dephosphorylate histones which have been phosphorylated a t serine or tyrosine by cyclic AMP-dependent protein kinase and an epidermal growth factor-stimulatedTyr-protein kinase, respectively. They found that these enzymes preferentially dephosphorylate [32P]Tyr-histones as well as [32P]Tyr-membrane proteins derived from A-431 cells. The membrane-boundalkaline phosphatases were found to be sensitive to inhibition by EDTA, but were insensitive to F-. Our findings concerning the properties of the activity of bovine kidney alkaline phosphatase toward various phosphoproteins are consistent with these data. Furthermore,our findings support thenotion that various alkaline phosphatases, including cytosolic and membrane-bound enzymes, could function as P-Tyr-proteinphosphatases in cells. It is interesting to note that F,, which exhibits little PNP phosphatase activity, is inactive toward P-Tyr-proteins. Swarup et ~ l (33) . has also reported that a PSer-protein phosphatase preparation from rabbit muscle shows some activity toward [32P]Tyr-histones. Whether this activity is similar to the P-Tyr-protein phosphatase described here is not clear. Brautigan and co-workers have reported that a P-Tyrprotein phosphatase activity exists in membrane vesicles derived from A-431 cells (35) and RSV-transformed rat cells (36). This activity was slightly stimulated by EDTA and F-, but was strongly inhibited by micromolar Zn2+.An EDTAinsensitive, Zn2+-sensitive P-Tyr-protein phosphatase activity has also been found in crude extracts of rat liver and muscle (37). The properties of these enzymaticactivities appear to be different from those of P-Tyr-protein phosphatases described in this report. The present studies indicate that the P-Tyr-protein phosphatase activity is much lower than the P-Ser-proteinphosDISCUSSION phatase activity inthe M , = 95,000 enzyme preparation, when Previous studies in this laboratory indicated that a P-Ser- the enzymatic activities are measured with phosphoproteins protein phosphatase of M , = 35,000 and its higher molecular innanomolar concentrations (Table 11). Furthermore, the phosphaweight forms purifiedfrom various tissues areassociated with ratio of the P-Tyr-proteinphosphatase to the PNP a MF-activated, sulfhydryl compound-stimulated P N P tase activity of the M , = 95,000 enzyme preparation is lower phosphatase activity (17-21). Although these two enzymatic than that of the kidney alkaline phosphatase. These data activities co-purify throughvariousseparation procedures, seem to indicate that the P-Tyr-proteinphosphatase activity they exhibit distinct catalyticproperties. Thus, they may in the M , = 95,000 preparation represents a minor activity in represent two different proteins of similar molecular weights animal tissues. However, since the physiological substrate(4 and ionic properties, or these two activities may reside in two for pp60"" or other Tyr-protein kinases is not known, the different forms of the same polypeptide chain. Although this possibility that this phosphatase activity, as detected using [32P]Tyr-lgG and [32P]Tyr-caseinas substrates,might particP N P phosphatase activity has an optimum around pH 8.69.0, the properties of this enzyme are distinctly different from ipate in the dephosphorylation of certain biologically importhose of the conventional alkalinephosphatases of membrane tant P-Tyr-proteins cannot be ruled out. It is also possible that it represents a latent form of the enzyme which requires origin (17). During the course of studying the substrate specificity of certain unknown metabolite(s) for activation. The observathe various P-Ser-protein phosphatases isolated from cardiac tion that the cardiac and the kidney phosphatases preferenmuscle and other tissues, we have found that preparationsof tially dephosphorylate [32P]Tyr-casein and [32P]Tyr-lgG, rethe M , = 35,000 phosphatase and itshigher molecular weight spectively (Table 11), indicates that in animal tissues there forms, as well as a M$+-activated phosphataseof M , = 45,000 exist different P-Tyr-proteinphosphatases with distinct sub(32, 34), areactive toward P-Tyr-proteins phosphorylated by strate specificities. They may play different roles in the reg-
7857
PhosphotyrosylProteinPhosphatase ulation of the dephosphorylation of cellular P-Tyr-proteins. The fact that a P-Tyr-protein phosphatase activity closely associates with a P-Ser-protein phosphatase activity is intriguing. Many P-Tyr-proteins, including pp60"" itself, have been shown to contain in addition P-Ser and/or P-Thr residues (1,38). It is tempting to speculate that the close association of these two different phosphatase activities may represent adevice for the coordinate dephosphorylation of P-Tyr and P-Ser and/or P-Thr residues in a given protein. Acknowledgment-We wish to thank Wanda W. S. Chan for her excellent technical assistance. REFERENCES 1. Hunter, T., Sefton, B. S., and Cooper, J. A. (1981) Cold Spring Harbor Conf Cell Prolif. 8, 1189-1202 2. Collett, M. S., and Erikson, R. L. (1978) Proc. Natl. Acad. Sci. U. S. A. 7 5 , 2021-2024 3. Levinson, A. D., Oppermann, H., Levintow, L., Varmus, H. E., and Bishop, J. M. (1978) Cell 15,561-572 4. Kawai, S., Yoshida, M., Sewaga, K., Sugiyama, H., Ishizaki, R., and Toyoshima, K. (1980) Proc. Natl. Acad. Sei. U. S. A. 77, 6199-6203 5. Feldman, R. A., Hanafusa, T., and Hanafusa, H. (1980) Cell 2 2 , 757-765 6. Barbacid, M., Beeman, K., and Devare, S. G. (1980) Proc. Natl. Acad. Sci. U. S. A. 77,5158-5163 7. Neil, J . C., Ghysdael, J., and Vogt, P. K. (1981) Virology 109, 223-228 8. Pawson, T., Guyden, J., Kung, T.-H., Radke, K., Gilmore, T., and Martin, G. S. (1980) Cell 2 2 , 767-775 9. Witte, 0. M., Dasgupta, A., and Baltimore, D. (1980) Nature (Lord.) 283,826-831 10. Van de Ven, W. J. M., Reynolds, F. H., Jr., and Stephenson, J. R. (1980) Virology 101, 185-197 11. Hunter, T., and Sefton, B. M. (1980) Proc. Natt. Acad. Sei. U. S. A. 77, 1311-1315 12. Collett, M. S., Purchio, A. F., and Erikson, R. L. (1980) Nature (Lord.)285,167-169 13. Levinson, A. D., Oppermann, H., Varmus, H. E., and Bishop, J. M. (1980) J . Biol. Chem. 255, 11973-11980 14. Sefton, B. M., Hunter, T., Beemon, K., and Eckhart, W. (1980) Cell 20,807-816
15. Oppermann, H., Levinson, A. D., Varmus, H. E., Levintow, L., and Bishop, J. M. (1979) Proc. Natl. Acad. Sci. U. S. A. 76, 1804-1808 16. Bishop, J . M., Courtneidge, S. A., Levinson, A. D., Oppermann, H., Quintrell, N., Sheiness, D. K., Weiss, S. R., and Varmus, H. E. (1980) Cold Spring Harbor Symp. Quunt. Biol. 44, 919930 17. Li,H.-C., Hsiao, K.-J., and Sampathkumar, S. (1979) J. Biol. Chem. 254,3368-3374 18. Li, H.-C. (1979) Eur. J. Biochem. 102, 363-374 19. Li, H.-C., and Chan, W. W. S. (1981) Arch. Biochem. Bzophys. 207,270-281 20. Li, H.-C. (1981) Cold Spring Harbor Conf. Cell Prolif. 8,441-457 21. Li, H.-C., Hsiao, K.-J., and Chan, W.W. S. (1978) Eur. J. Biochem. 84, 215-225 22. Erlichman, J., Hirsch, A. H., and Rosen, 0. M. (1971) Proc. Natl. Acad. Sci. U. S. A. 68, 731-735 23. Nimmo, G. A., and Cohen, P. (1978) Eur. J. Biochem. 87, 341351 24. Brugge, J. S., and Erikson, R. L. (1977) Nature (Lord.) 2 6 9 , 346-348 25. Cheng, Y. S. E., and Chen, L. B. (1981) Proc. Natl. Acad. Sci. U. S. A. 7 8 , 2388-2392 26. Torres, H. N., and Chelala, C. A. (1961) Biochim. Biophys. Acta 481,140-151 27. Yam. S.-D.. Vandenheede. J. R.. Goris. J.. and Merlevede. W. (1980) J . biol. Chem. 255,11759-11767 28. Vandenheede, J. R., Yang, S.-D., Goris, J., and Merlevede, W. (1980) J . Biol. Chem. 255, 11768-11774 29. Davis, B. J. (1964) Ann. N. Y. Acad. Sci. 121,404-427 30. Laemmli, U. K. (1970) Nature (Lord.)227, 680-685 31. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265-275 32. Li, H.-C. (1982) Curr. Top. Cell Regul. 21, 129-174 33. Swarm, G.. Cohen.. S.,. and Garbers. D. L. (1981) J. Biol. Chem. 256,8197-8201 34. Li. H.-C.. Tabarini. D.. Chene. Y.-S.. and Chen. L. B. (1981) Fed. Am. Soc. Exp. Biol. 40, 1539 35. Brautigan, D. L., Bornstein, P., and Gallis, B. (1981) J . Biol. Chem. 256,6519-6522 36. Gallis, B., Bornstein, P., and Brautigan, D. L. (1981) Proc. Natl. Acad. Sci. U. S. A. 78,6689-6693 37. Foulkes, J . G., Howard, R. F., and Ziemiecki, A. (1981) FEBS Lett. 130, 197-200 38. Cooper, J. A,, and Hunter, J. (1981) Mol. Cell Bwl. 1, 165-178 '
