199

Biochimica et Biophysica Acta, 1051 (1990) 199-202 Elsevier

BBA Report

BBAMCR 10250

An investigation of the substrate specificity of protein phosphatase 2C using synthetic peptide substrates; comparison with protein phosphatase 2A A r i a n n a D o n e l l a D e a n a 1, Clare H. M a c G o w a n 2, Philip C o h e n 2 F e r n a n d o M a r c h i o r i 37 H e l m u t E. M e y e r 4 and L o r e n z o A. P i n n a 1 I Dipartimento di Chimica Biologica, Universitd di Padova, Padova (Italy), 2 Department of Biochemistry, University of Dundee, Dundee (U.K.) 3 Dipartimento di Chimica Organica, Universitd di Padova and Centro di Studio dei Biopolimeri del CNR, Padova (Italy) and 4 Institut fur Physiologische Chemie, Ruhr- Universitiit Bochum, Bochum (F.R.G.)

(Received 18 September 1989)

Key words: Protein phosphatase; Protein phosphorylation; Synthetic peptides; Cell regulation; Substrate specificity; Enzymology

The synthetic phosphopeptide RRATpVA was found to be the most effective substrate for protein phosphatase 2C (PP2C) so far identified. Replacement of phosphothreonine by phosphoserine decreased activity over 20-fold and a striking preference for phosphothreonine was also observed with two other substrates (RRSpTpVA and casein) that were phosphorylated on both serine and threonine. Replacement of the C-terminal valine in RRATpVA by proline abolished dephosphorylation, while exchanging the N-terminal alanine by proline had no effect. The preference for phosphothreonine and the effect of proline are similar to protein phosphatase 2A (PP2A). However, the peptide RRREEETpEEEAA, an excellent substrate for PP2A, was not dephosphorylated by PP2C, and substitution of the C-terminal valine in RRATpVA by glutamic acid reduced the rate of dephosphorylation by PP2C over 10-fold, without affecting dephosphorylation by PP2A. Addition of two extra N-terminal arginine residues to RRASpVA increased PP2A catalysed dephosphorylation 4- to 5-fold, without altering dephosphorylation by PP2C. These results represent the first study of the specificity of PP2C using synthetic peptides, and strenghten the view that this approach may lead to the development of more effective and specific substrates for the serine/threonine-specific protein phosphatases.

Protein phosphatase 2C (PP2C) is one of the four major types of protein phosphatase in eukaryotic cells that dephosphorylate serine and threonine residues (reviewed in Ref. 1). It can be distinguished from protein phosphatase 2A (PP2A) and protein phosphatase 2B (PP2B) by its absolute requirement for Mg 2÷ [1] and lack of inhibition by the tumour promoter okadaic acid [2], and from protein phosphatase 1 (PP1) by its preferential dephosphorylation of the a-subunit (rather than the fl-subunit) of phosphorylase kinase and lack of inhibition by the thermostable proteins inhibitor 1 and inhibitor 2 [1]. Recently, cDNA cloning of the PP2C 1 isoform [3] and partial amino acid sequencing of the PP2C 2 isoform [4] demonstrated that type 2C protein phosphatases are completely unrelated in structure to

Abbreviation: PP protein phosphatase. Correspondence: A. Donella Deana, Dipartimento di Chimica Biologica, Universitfi degli Studi di Padova, Via Trieste 75, 35132 Padova, Italy

PP1, PP2A and PP2B, which are all members of the same gene family (reviewed in Ref. 5). The molar concentration of PP2C in vivo is similar to that of PP1 and PP2A [6], and like these other phosphatases it is capable of dephosphorylating a number of proteins in vitro. However, its specific activity is low towards all phosphoproteins so far examined, suggesting that the physiological substrates of PP2C may have yet to be identified [6]. In order to gain further information about the substrate specificity of PP2C, we have therefore examined its ability to dephosphorylate synthetic peptide substrates that were recently used to study the substrate specificities of PP1 and PP2A [7,8]. The rates of dephosphorylation of some synthetic peptides by PP2C are summarized in Tables I and II. PP2C showed a striking preference for the peptide RRATpVA over RRASpVA, as observed previously with PP2A. The strong preference for phosphothreonine was also observed with the peptide RRSpTpVA which was phosphorylated by PK-A on both the serine and threonine residue (Table III) and with mixed casein phosphorylated on both

0167-4889/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

200 serine and threonine residues by casein kinase 2 (Table III, see Refs. 9 and 10 for the sites of phosphorylation). No small peptides phosphorylated on serine residues were dephosphorylated at significant rates, and PP2C was also unable to dephosphorylate the tyrosine residue in angiotensin (DRVYplHPF, Table I).

TABLE I

Rates of dephosphorylation of phosphopeptides and proteins by PP2C PP2C2 was purified as in Ref. 6 and the catalytic subunit of cyclic AMP-dependent protein kinase (PK-A) from bovine cardiac muscle [13] was a generous gift from Dr. B.A. Hemmings (Friedrich-Miescher Institute, Basel, Switzerland). Histone type II AS and angiotensin II (DRVYIHPF) were obtained from Sigma, mixed casein and fl-casein were prepared as in Ref. 14 and peptide substrates were synthesized as in [15-17]. The synthesis of peptides RRREEESEEEAA, RRREEETEEEAA and GRTGRRNSIHDIL was done with a CRB peptides synthesizer using Fmoc-protected and pentafluorphenylactivated amino acids. The deprotected peptides were purified using reversed-phase HPLC. The purified peptides were analyzed by PTCamino acid analysis and were sequenced with an Applied Biosystems Gas Phase sequenator equipped with a 120 PTH-on-line analyzer. 32P-labelled ~-casein and whole casein were phosphorylated by either rat liver casein kinase 2 (CK-2) [18] or PK-A [6]. After incubation for 4 h at 30°C, reactions were stopped by addition of 10% (w/v) trichloroacetic acid. Phosphoproteins were recovered by centrifugation, washed four times by resuspension in 10% (w/v) trichloroacetic acid, dissolved in 50 mM Tris/HCL (pH 7), 0.1 mM EGTA, 0.1% (v/v) 2-mercaptoethanol (buffer A) and dialysed overnight against buffer A. Histone type II AS and most peptides were phosphorylated by incubation with PK-A [16], ASEEEEE and TEEEEE by CK-2 [19] and DRVYIHPF by rat spleen protein tyrosine kinase III [20]. Incubations (4 h) were stopped by addition of 30% (v/v) acetic acid and [732p]-ATP removed by ion exchange chromatography on a 2×0.5 cm Dowex AG1X8 column (BioRad) equilibrated with 30% (v/v) acetic acid [21]. The eluate containing the phosphorylated substrates was lyophilised and dissolved in buffer A. The phosphopeptide AAA[32p]SVA was obtained by incubating the phosphopeptide RRRRAAA[32p]SVA with trypsin, at a weight ratio enzyme:substrate of 1:10, separated from residual parent peptide by high voltage paper electrophoresis [22], eluted from the paper with 10% (v/v) acetic acid, lyophilised and dissolved in buffer A. 32P-radioactivity incorporated into protein and peptide substrates was 1000-1500 cpm/pmol and the concentration of 32P-labelled substrafes in the assays (calculated from specific radioactivities) were 2 ~tM. No substantial influence of the concentration of non-phosphorylated substrate on the dephosphorylation rate could be observed over the range 3-300 ~M. Protein phosphatase 2C was assayed by incubating enzyme and 32P-labelled substrates for 10 rain at 30"C in the presence of 20 mM Mg 2÷ [6]. Reactions were stopped by addition of 10% (w/v) trichloroacetic acid and the [32p]phosphate liberated was converted into its phosphomolybdic complex, extracted with isobutyl alcohol-toluene (1:1, v/v) and quantitated as in [23]. Assays with each substrate were only linear up to 10-20% phosphate released and dephosphorylation was kept within this limit to ensure that initial rate conditions were met. Phosphorylated residues are underlined and activities are expressed relative to those obtained with mixed casein phosphorylated by PK-A. Peptides 1, 10, 13 and 14 reproduce the phosphorylation sites of pyruvate kinase, inhibitor-l, phosphorylase kinase (a-subunit) and ribosomal protein $6. Essentially identical results were obtained using phosphopeptides that had been freed from their dephosphorylated counterparts by high voltage paper electrophoresis [22].

TABLE I (continued) Substrate

Concentration (mg/ml)

RRA_SVA RRRRA_SVA RRRRAA_SVA RRRRAAASVA AAASVA RRATVA RRPTVA

1 2 3 4 5 6 7

RRAT_PA

8 9 10 11 12 13 14 15 16 17 18 19 20

RRP_TPA RRRRPTPA RRSTVA RRSTEA RRI._SISTES RRL__SSLRA PRRRRRSSRPVR GRTGRRN_SIHDIL

RRREEES_EEEAA RRREEETEEEAA A§EEEEE

T EEEEE DRVYIHPF

21

[32P]histone II AS (PK-A) [32 P]fl-casein (CK-2) mixed[ 32P]casein (CK-2) mixed[ 32P]casein (PK-A)

0.006 0.005 0.010 0.004 0.003 0.054 0.120 0.520 1.000 0.300 0.007 0.220 0.020 0.020 0.040 0.064 0.068 0.080 0.007 0.050 0.065 0.120 0.480 1.200 2.000

Phosphate content (mol/mol) 0.30 0.42 0.16 0.67 1.00 0.05 0.05 0.0l 0.01 0.05 0.22 0.01 0.80 0.42 0.61 0.42 0.38 0.27 0.27 0.02 0.05 0.45 0.30 0.18 0.02

Activity/ (q,) 8.8 8.8 25 9.3 0 253 276 0 0 0 210 18 6.4 5.7 0 0 0 0 0 0 0 58 52 58 100

The specificity of PP2C towards phosphopeptides also resembled PP2A in the effect of proline. In the peptide RRATVA, replacement of the valine residue immediately C-terminal to the phosphothreonine residue with proline abolished phosphorylation, while substitution of the alanine N-terminal to the phos-

TABLE I1

Differences in substrate specificity between PP2A and PP2C PP2A was the PCSH1 phosphatase from rabbit skeletal muscle [24], a three-subunit enzyme, also termed PP2A 1 [1], and kindly supplied by Dr. J. Goris (University of Leuven, Belgium). PP2C 2 was purified from rabbit skeletal muscle as in Ref. 6. Activities are expressed as pmol. rain- 1. rnl- 1 and further details are given in the legend to Table I. The data with PP2A are in substantial agreement with Ref..7 and 8. Peptide RRASVA RRRRASVA RRRRAASVA RRRRAAASVA RRATVA RRPTVA RRATPA RRSTVA RRSTEA RRREEETEEEAA

PP2A

PP2C

11.7 51.2 375 104 320 316 0 120

3.4 3.4 9.7 3.6 98.3 106 0 81.6 7 0

130

200

201 TABLE III Dephosphory lation of substrates containing both [ 32p] Ser and [ 32p] Thr by PP 2C

32p-labelled mixed casein phosphorylated by casein kinase 2 was incubated for 30 min in the presence and absence of PP2C. Reactions were terminated by addition of 6 M HC1, and after hydrolysis for 4 h at l l 0 ° C , phosphoserine and phosphothreonine were resolved by high voltage paper electrophoresis at pH 1.9 [22]. The spots, corresponding to the labelled phosphoamino acids were cut out and counted for 32P-radioactivity. The experimental values were corrected for hydrolytic losses of 48% (phosphoserine) and 13% (phosphothreonine), respectively [25]. Substrate

Residue

[ 32 p] Pi content (pmol)

Control

+ PP2C

RRSTV A RRSTVA

[ 32P]Ser [nP]Thr

20.0 12.2

20.2 1.3

32P-mixed casein 32P-mixed casein

[ 32P]Ser [ 32P]Thr

13.8 18.5

14.0 11.4

The specificity of PP2C can also be clearly distinguished from that of PP1, which is almost inactive towards all peptides that have so far been tested [7,12]. In summary, the specificity of PP2C towards phosphopeptides appears to be narrower than that of PP2A, although both enzymes display a striking preference for phosphothreonyl peptides over their phosphoseryl homologues. Not all the phosphothreonyl substrates dephosphorylated by PP2A are good substrates for PP2C, and the positive determinants that improve the dephosphorylation of phosphoseryl peptides by PP2A appear to be ineffective with PP2C (which was virtually inactive towards all phosphoseryl peptides tested). Whether its specificity for phosphopeptides in vitro also reflects the activity of PP2C towards endogenous targets in vivo remains an open question. Nevertheless, phosphopeptide substrates may prove useful for detecting and quantitating PP2C activity in the presence of other protein phosphatase activities.

Acknowledgements phothreonine with proline had almost no effect (Tables I and II). The same was observed with PP2A [7,11]. The peptides RRATVA (K m = 0.6 pM), RRPTVA (2.0/~M) and RRSTVA (K m = 3.0 pM) were each dephosphorylated several-fold more rapidly than casein phosphorylated by PKA (Table I), previously the most effective substrate for PP2C. These small peptides should therefore be very useful for the routine assay of PP2C. As observed with protein substrates, the dephosphorylation of small peptides by PP2C was dependent on Mg 2+ and unaffected by inhibitor 2 (0.1/~M) or okadaic acid (1.0 /~M). With the peptide RRSTVA, activity in the presence of 1 mM EDTA was less than 6% of that measured at 20 mM Mg 2÷. Despite the similarities in specificity between PP2A and PP2C, some striking differences were also observed. In particular, the peptide RRREEETpEEEAA, which is an excellent substrate for PP2A (Table II), was not dephosphorylated at all by PP2C. The failure of PP2C to dephosphorylate this peptide is most likely due to the presence of acidic residues surrounding the phosphothreonine, because in RRSpTpVA substitution of the valine with glutamic acid drastically reduced the rate of dephosphorylation by PP2C (Table I). In contrast, dephosphorylation by PP2A was unaffected (Table II). Another difference was the failure of two additional N-terminal arginine residues to improve the activity of PP2C towards the phosphoserine in RRASpVA. This feature has been shown to markedly improve dephosphorylation by PP2A (Ref. 8 and Table II). In particular, the peptide RRRRAASpVA was dephosphorylated over 30-fold more rapidly than RRASpVA by PP2A, but less than 3-fold faster than RRASpVA by PP2C (Table II).

This work was supported by the Italian Ministero della Pubblica Istruzione and Consiglio Nazionale delle Ricerche (Grant 88.00612.04 to LAP), by the Medical Research Council and Royal Society (to PC) and by an MRC studentship (to C.H.M.). We are indebted with Dr. J. Gods (University of Leuven, Belgium) for the generous gift of protein phosphatase PCS m.

References 1 Cohen, P. (1989) Ann. Rev. Biochem. 58, 453-508. 2 Bialojan, C. and Takai, A (1988) Biochem. J. 256, 283-290. 3 Tamura, S., Lynch, K.R., Lamer, J., Fox, J., Yasui, A., Kikuchi, K., Suzuki, Y. and Tsuiki, S. (1989) Proc. Natl. Acad. Sci. USA 86, 1796-1800. 4 McGowan, C.H. (1988) Ph.D. Thesis, University of Dundee. 5 Cohen, P. and Cohen, P.T.W. (1989) J. Biol. Chem., 264, 21435-21438. 6 McGowan, C.H. and Cohen, P. (1987) Eur. J. Biochem. 166, 713-722. 7 Agostinis, P., Goris, J., Waelkens, E., Pinna, L.A. Marchiori, F. and Merlevede, W. (1987) J. Biol. Chem. 262, 1060-1064. 8 Agostinis, P., Goris, J., Marchiori, F., Pinna, L.A. and Merlevede, W. Eur. J. Biochem., in press. 9 Pinna, L.A., Donella-Deana, A. and Meggio, F. (1979) Biochem. Biophys. Res. Commun. 87, 114-120. 10 Meggio, F., Donella-Deana, A and Pinna, L.A. (1980) In Enzyme Regulation and Mechanism of Action (Mildner, P. and Pies, B., eds.) pp. 23-33 Pergamon Press, New York. 11 Pinna, L.A., Agostinis, P. and Ferrari S. (1986) Adv. Prot. Phosphatases 3, 327-368. 12 McNall, S. and Fischer, E.H. (1988) J. Biol. Chem. 263, 1893-1897. 13 Beavo, J., Bechtel, P. and Krebs, E.G. (1974) Methods Enzymol. 38, 299-308. 14 Mercier, J.C., Maubois, J.L., Porhanski, S. and Ribadeau Dumas B. (1968) Bull. Soc. Chim. Biol. 50, 521-530. 15 Meggio, F., Chessa, G., Borin, G., Pinna, L.A. and Marchiori, F. (1981) Biochim. Biophys. Acta 662, 94-101.

202 16 Chessa, G., Borin, G., Marchiori, F., Meggio, F., Brunati, A.M. and Pinna, L.A. (1983) Eur. J. Biochem. 135, 609-614. 17 Marchiori, F., Meggio, F., Matin, O., Borin, G., Calderan, A., Ruzza, P. and Pinna, L.A. (1988) Biochim. Biophys. Acta 971, 332-338. 18 Donella-Deana, A., Meggio, F. and Pinna, L.A. (1981) FEBS Lett. 125, 77-82. 19 Pinna, L.A., Meggio, F., Marchiori, F. and Borin, G. (1984) FEBS Lett. 171, 211-214. 20 Brunati, A.M. and Pinna, L.A. (1988) Eur. J. Biochem. 172, 451-457.

21 Kemp, B.E., Bylund, D.B., Huang, T.S. and Krebs, E.G. (1975) Proc. Natl. Acad. Sci. USA 72, 3448-3452. 22 Donella-Deana,, A., Meggio, F. and Pinna, L.A. (1979) Biochem. J. 179, 693-696. 23 Pinna, L.A., Donella, A., Clari, G. and Moret, V. (1976) Biochem. Biophys. Res. Commun. 70, 1308-1315. 24 Waelkens, E., Goris, J. and Merlevede, W. (1987) J. Biol. Chem. 262, 1049-1059. 25 Bylund, D.B. and Huang, T.S. (1976) Anal. Biochem. 73, 477-485.