- Email: [email protected]
Protein kinase C pseudosubstrate prototope: Structure-function relationships
Cellular Signalling VoL 2, No. 2, pp. 187-190, 1990.
0698--6568t'90 $3.00 + .00 ~ 1990 l~rgamon Press pie
Printed in Great Britain.
PROTEIN KINASE C PSEUDOSUBSTRATE PROTOTOPE: S T R U C T U R E F U N C T I O N RELATIONSHIPS COLIN HOUSE and BRUCE E. KEMP* St. Vincent's Institute of Medical Research, 41 Victoria Parade, Fitzroy, Victoria, 3065, Australia (Received 22 September 1989; and accepted 1 November 1989) Abstract--A structure-function study of the protein kinase C (PK-C) pseudosubstrate sequence (RtgFARKGALRQKNV3~) has been undertaken. The role of specific residues was investigated using an alanine substitution scan. Arg-22 was the most important determinant in the inhibitor sequence, since substitution of this residue by alanine gave a 600-fold increase in the ICs0 value to 81 +9#M. Substitutions of other basic residue also increased the ICs0, 5-, 11- and 24-fold for the Ala-19, Ala-23 and Ala-27 substitutions, respectively. The importance of basic residues in determining the potency of the pseudosubstrate peptide reflects the requirements for these residues in peptide substrate phosphorylation. The residues Gly-24, Leu-26 and Gin-28 were also important for pseudosubstrate inhibitor potency. The large difference in the ICs0 value for the [A2r]PK-C(19-31) peptide makes it a valuable control in studies employing the pseudosubstrate peptide to explore functional roles of PK-C.
Key words: Protein kinase C, pseudosubstrate, peptide, inhibitor. INTRODUCTION
idendified from eDNA sequences for which natural substrates are not yet identified [7]. It was therefore of interest to understand the structure-function relationships that determine the potency of the pseudosubstrate peptide inhibitors.
PROa~IN kinase C (PK-C) is an important regulatory enzyme involved in multiple cellular responses [1]. The protein substrates phosphorylated by PK-C contain basic residues in proximity to the phosphorylated residue [2] that have been shown to act as important specificity determinants. Recently we reported that the enzyme's regulatory domain also contains a basic residue-rich region between residues 19 and 36, which is thought to inhibit the enzyme by binding as a pseudosubstrate to the active site [3]. The pseudosubstrate region is amino terminal to the zinc finger-like sequence which has recently been shown to act as the phorbol ester binding site [4]. Synthetic peptides corresponding to the pseudosubstrate region are particularly potent inhibitors of PK-C and are being used as inhibitors to investigate the role of PK-C in a variety of physiological processes [5, 6]. Serine containing analogues of the pseudosubstrate sequence have also been employed as substrates for members of the PK-C family
MATERIALS AND METHODS [y-P32]ATP was purchased from New England Nuclear and all other reagents and methods were as described previously [2]. The enzyme was purified from rat brain and stored in buffer containing glycerol and Triton X-100; peptides were synthesized as the carboxyl terminal amides using the Merrifield procedure on an Applied Biosystems 430A automated peptide synthesizer; and peptide purification was by reversed phase chromatography on a Vydac C18 column (2.5 x 70 era), using an acetonitrile gradient in the presence of 0.1% (v/v) trifluoroacetic acid. All peptides were quantitated by amino acid analysis following acid hydrolysis as described previously [2]. Phosphorylation assays were performed in a volume of 40/d in 20mM Tris-HC1, pH7.5, 12.5mM MgCI2, 0.5raM EGTA, 0.75mM CaCI2, 250/aM [~-~2p]ATP (500 cpm/pmol), 5/aM [Ala9.~° LystL~2]GS(1-12) (synthetic peptide substrate from glycogen synthase [2]), 0.5/ag/ml phosphatidylserinc
*Author to whom correspondenceshould be addressed. 187
188
C. House and B. E. KEMP
TABLE I. EFFECT OF PEPTIDE LENGTH ON KINETIC PROPERTIES
RIgF2°ARKGA25LRQKN3°VHEVKN36 Peptide sequence PKC (19-27)* [A2s] PKC (19-28) PKC (19-31) PKC (19-36)* [K 17] PKC (16-31) [yH, KI7] PKC (11-31)
IC50 (/~M) 6.2__+0.9 2.98 + 0.07 0.13__+0.02 0.18 + 0.02 0.075__+0.005 1.40-0.11
The synthetic peptide residues are numbered according to the protein sequence reported by Knopf e t al.[14]. The length of the peptides are indicated in parentheses and amino acid substitutions are given in square brackets. The kinetic parameters were determined as described under Materials and Methods with the synthetic peptide substrate [A9'~°,Kn'12] GS (1-12) used at its Km (5 #M) concentration. Each experiment contained six concentrations of inhibitor peptide ranging from approximately 0.3- to 2.0-fold the preliminary estimate of the ICs0; inhibition ranged from approximately 30 to 70%. Values given are mean + SEM for three independent experiments. Data for previously reported peptides [3] is indicated by an asterisk.
and 0.35/~g of rat brain PK-C. Reactions were terminated by the addition of 30/zl of reaction mixture onto phosphocellulose paper, and then washed in 75 mM phosphoric acid before drying and counting (2]. RESULTS AND DISCUSSION It has been found that peptide length may have an important influence on the potency of peptide inhibitors or the kinetics of peptide substrate phosphorylation for several different protein kinases [8]. In the present study we have investigated the influence of shortening and extending the pseudosubstrate sequence. The shortened peptide [A~] PKC (19-28) had an increased ICso, to approximately 3/~M (Table 1), indicating that reducing the peptide length had a deleterious effect on inhibitor potency. In contrast, the peptide [K ~7] PKC (16-31) was approximately twice as potent as PKC (19-31), indicating that the amino term-
inal extension SKV improved the interaction of the pseudosubstrate inhibitor with the enzyme. Lysine was included at residue 17 in place of the threonine to provide an additional positive charge on the same side of the peptide backbone as Arg-22 (see below). On the other hand, extending the peptide to 11 with the glutamic acid containing sequence YEGEESKV, corresponding to residues 11-18 with tyrosine at position 11, resulted in an increased IC50 value of the peptide from 0.075 to 1.4/~M. Tyrosine was included at position 11 for radioiodination in the event of this peptide having a substantially lower IC50. This result contrasts with the cAMP-dependent protein kinase where the amino terminal extension beyond the arginine cluster of the protein kinase inhibitor confers a substantial increase in potency [9]. Our results demonstrate that the optimal length for inhibition by the pseudosubstrate peptide is between 13 and 16 residues in the region between residues 16 and 31. While these findings are important for the design of peptide inhibitors of PK-C it is not yet known to what extent they reflect the properties and function of the pseudosubstrate region in the parent protein. The contribution of individual residues to the potency of the pseudosubstrate peptide PK-C (19-31) was investigated by making a series of substitutions (Table2). Substrate specificity studies on PK-C in several laboratories [2, 1012] have provided evidence that basic residues on either side of the phosphate acceptor site can act as specificity determinants. Substitution of the basic residues in the pseudosubstrate peptide PKC (19-31) all decreased potency, however, there were large differences in the magnitude. Replacement of Arg-19 with alanine caused a 4.7-fold increase in the IC50 to 0.6/zM (Table 2). Previously it was found that replacing Arg-22 with alanine increased the IC50 to 3.1/~M [3]. However, replacement of Arg-22 with alanine caused an approximate 600-fold increase in the IC50 to 81/~M. This residue is separated by two amino acids from Arg-25, the equivalent of the phosphate acceptor site. Previously the importance of amino terminal basic residues had been demonstrated with synthetic
P K - C pseudosubstrate prototope
TABLE 2.
EFFECT OF AMINO ACID SUBSTITUTIONS ON KINETIC PROPERTIES
RI9F2°ARKGA25LRQKN3°VHEVKN 36 Peptide sequence
IC50 (#M)
PKC (19-31) [A~9]PKC (19-31) [A22] PKC (19-31) [A23] PKC (19-31) [V24] PKC (19-31) [A26] PKC (19-31) [A27] PKC (19-31)
0.13+0.02 0.60+0.07 81 +9 1.47 + 0.09 1.7+0.4 2.23+0.29 3.1_+0.6
The synthetic peptide residues are numbered according to the protein sequence reported by Knopf et al. [14] as described in Table 1. The length of the peptides are indicated in parentheses and amino acid substitutions are given in square brackets. The kinetic parameters were determined as described under Materials and Methods and values given are mean _ SEM for three independent experiments. peptides, however, not all phosphorylation site sequences PK-C are arranged in this way. The lysine residue at position 23 influenced the IC50, since substitution with alanine resulted in a 10fold increase in the IC50. Lysine residues are particularly prominent in the local phosphorylation site sequence of the recently described M A R C K S protein [12] which is a potent substrate for PK-C. The glycine residue at position 24 was substituted with the bulkier residue valine because glycine residues in the cAMPdependent protein kinase inhibitor adjacent to the critical arginine residues have an influence on the potency [13]. Substitution at Gly-24 caused a 14-fold increase in the IC5o equivalent to changing one of the basic residues in the pseudosubstrate peptide. The detrimental effect of valine in this position suggests that flexibility of the peptide may be important either in the correct orientation of the critical basic residues or in accommodating Ala-25 in the active site. Replacement of Gln-28 with alanine caused a decrease in potency (Table 2) when substitution with a smaller residue may have been expected to increase potency as seen for Gly-24. The results obtained in the present study demonstrate that the factors influencing the potency of
189
PK-C pseudosubstrate inhibitor are similar to those that determine substrate specificity for peptide phosphorylation [2]. The potency of the pseudosubstrate peptide PK-C (19-31) results from the contribution of multiple residues. All substitutions had an effect with only the [A tg] PK-C (19-31) analogue having less than a 10fold increase in IC50 over the parent peptide. This is in marked contrast to other protein kinases such as the myosin light chain kinase and the cAMP-dependent protein kinase where some residues can be substituted without modifying the kinetic properties. In view of the importance of the basic residues in the pseudosubstrate region it can be expected that changing these to alanine by point mutations would render the enzyme constitutively actively. The very large difference in the IC5o of the peptide PK-C (19-31) and the Ala-22 analogue suggests that these peptides could be used effectively as an inhibitor and analogue control of PK-C, respectively. Acknowledgements--This work was supported by grants from the Victorian Anti-Cancer Council and the NH and MRC.
REFERENCES I. Nishizuka Y. (1988) Nature 334, 661-665. 2. House C., Wettenhall R. E. H. and Kemp B. E. (1987) J. biol. Chem. 262, 772-777. 3. House C. and Kemp B. E. (1987) Science, 238, 1726--1728. 4. Ono Y., Fujii T., Igarshi K., Kuno T., Tanaka C., Kikkawa U. and Nishizuka Y. (1989) Proc. natn. Acad. Sci. U.S.A. 86, 4868-4871. 5. Bosma M. W. and Hille B. (1989) Proc. natn. Acad. Sci. U.S.A. 86, 2843-2947. 6. Maliow R., Schulman H. and Tsien R. W. (1989) Science 245, 862-866. 7. Schapp D., Parker P. J., Bristol A., Kriz R. and Knopf J. (1989) FEBS Lett., 243, 351-357. 8. Kemp B. E. (1988) Molecular Aspects of Cellular Regulation, Voi. 5 (P. Cohen and C. Klee, Eds). Elsevier, Amsterdam, 195-224. 9. Kemp B. E., Cheng H. C. and Walsh D. A. (1988) Meth. Enzym. 159, 173-183. 10. Woodgett J. R. and Hunter T. (1986), Fur. J. Biochem. 161, 177-184. I I. Ferrari S., Marchiori F., Borin G. and Pinna L. A. (1985) FEBS Lett. 184, 72-77.
190
C. House and B. E. KEMP
12. Graft J. M., Stumpo D. O. and Black,shear P. J. (1989) J. biol. Chem. 264, 11,912-11,919. 13. Glass D. B,, Chen H. C., Mende-Mueller L., Reed J. and Walsh D. A. (1989) J. biol. Chem.
264, 8802-8810.
14. Knopf J. L., Lee M. H., Sultzman L. A., Kliz R. W., Loomis C. R., Hewick R. M. and Bell R. M. (1986) Cell 46, 491-502.
0698--6568t'90 $3.00 + .00 ~ 1990 l~rgamon Press pie
Printed in Great Britain.
PROTEIN KINASE C PSEUDOSUBSTRATE PROTOTOPE: S T R U C T U R E F U N C T I O N RELATIONSHIPS COLIN HOUSE and BRUCE E. KEMP* St. Vincent's Institute of Medical Research, 41 Victoria Parade, Fitzroy, Victoria, 3065, Australia (Received 22 September 1989; and accepted 1 November 1989) Abstract--A structure-function study of the protein kinase C (PK-C) pseudosubstrate sequence (RtgFARKGALRQKNV3~) has been undertaken. The role of specific residues was investigated using an alanine substitution scan. Arg-22 was the most important determinant in the inhibitor sequence, since substitution of this residue by alanine gave a 600-fold increase in the ICs0 value to 81 +9#M. Substitutions of other basic residue also increased the ICs0, 5-, 11- and 24-fold for the Ala-19, Ala-23 and Ala-27 substitutions, respectively. The importance of basic residues in determining the potency of the pseudosubstrate peptide reflects the requirements for these residues in peptide substrate phosphorylation. The residues Gly-24, Leu-26 and Gin-28 were also important for pseudosubstrate inhibitor potency. The large difference in the ICs0 value for the [A2r]PK-C(19-31) peptide makes it a valuable control in studies employing the pseudosubstrate peptide to explore functional roles of PK-C.
Key words: Protein kinase C, pseudosubstrate, peptide, inhibitor. INTRODUCTION
idendified from eDNA sequences for which natural substrates are not yet identified [7]. It was therefore of interest to understand the structure-function relationships that determine the potency of the pseudosubstrate peptide inhibitors.
PROa~IN kinase C (PK-C) is an important regulatory enzyme involved in multiple cellular responses [1]. The protein substrates phosphorylated by PK-C contain basic residues in proximity to the phosphorylated residue [2] that have been shown to act as important specificity determinants. Recently we reported that the enzyme's regulatory domain also contains a basic residue-rich region between residues 19 and 36, which is thought to inhibit the enzyme by binding as a pseudosubstrate to the active site [3]. The pseudosubstrate region is amino terminal to the zinc finger-like sequence which has recently been shown to act as the phorbol ester binding site [4]. Synthetic peptides corresponding to the pseudosubstrate region are particularly potent inhibitors of PK-C and are being used as inhibitors to investigate the role of PK-C in a variety of physiological processes [5, 6]. Serine containing analogues of the pseudosubstrate sequence have also been employed as substrates for members of the PK-C family
MATERIALS AND METHODS [y-P32]ATP was purchased from New England Nuclear and all other reagents and methods were as described previously [2]. The enzyme was purified from rat brain and stored in buffer containing glycerol and Triton X-100; peptides were synthesized as the carboxyl terminal amides using the Merrifield procedure on an Applied Biosystems 430A automated peptide synthesizer; and peptide purification was by reversed phase chromatography on a Vydac C18 column (2.5 x 70 era), using an acetonitrile gradient in the presence of 0.1% (v/v) trifluoroacetic acid. All peptides were quantitated by amino acid analysis following acid hydrolysis as described previously [2]. Phosphorylation assays were performed in a volume of 40/d in 20mM Tris-HC1, pH7.5, 12.5mM MgCI2, 0.5raM EGTA, 0.75mM CaCI2, 250/aM [~-~2p]ATP (500 cpm/pmol), 5/aM [Ala9.~° LystL~2]GS(1-12) (synthetic peptide substrate from glycogen synthase [2]), 0.5/ag/ml phosphatidylserinc
*Author to whom correspondenceshould be addressed. 187
188
C. House and B. E. KEMP
TABLE I. EFFECT OF PEPTIDE LENGTH ON KINETIC PROPERTIES
RIgF2°ARKGA25LRQKN3°VHEVKN36 Peptide sequence PKC (19-27)* [A2s] PKC (19-28) PKC (19-31) PKC (19-36)* [K 17] PKC (16-31) [yH, KI7] PKC (11-31)
IC50 (/~M) 6.2__+0.9 2.98 + 0.07 0.13__+0.02 0.18 + 0.02 0.075__+0.005 1.40-0.11
The synthetic peptide residues are numbered according to the protein sequence reported by Knopf e t al.[14]. The length of the peptides are indicated in parentheses and amino acid substitutions are given in square brackets. The kinetic parameters were determined as described under Materials and Methods with the synthetic peptide substrate [A9'~°,Kn'12] GS (1-12) used at its Km (5 #M) concentration. Each experiment contained six concentrations of inhibitor peptide ranging from approximately 0.3- to 2.0-fold the preliminary estimate of the ICs0; inhibition ranged from approximately 30 to 70%. Values given are mean + SEM for three independent experiments. Data for previously reported peptides [3] is indicated by an asterisk.
and 0.35/~g of rat brain PK-C. Reactions were terminated by the addition of 30/zl of reaction mixture onto phosphocellulose paper, and then washed in 75 mM phosphoric acid before drying and counting (2]. RESULTS AND DISCUSSION It has been found that peptide length may have an important influence on the potency of peptide inhibitors or the kinetics of peptide substrate phosphorylation for several different protein kinases [8]. In the present study we have investigated the influence of shortening and extending the pseudosubstrate sequence. The shortened peptide [A~] PKC (19-28) had an increased ICso, to approximately 3/~M (Table 1), indicating that reducing the peptide length had a deleterious effect on inhibitor potency. In contrast, the peptide [K ~7] PKC (16-31) was approximately twice as potent as PKC (19-31), indicating that the amino term-
inal extension SKV improved the interaction of the pseudosubstrate inhibitor with the enzyme. Lysine was included at residue 17 in place of the threonine to provide an additional positive charge on the same side of the peptide backbone as Arg-22 (see below). On the other hand, extending the peptide to 11 with the glutamic acid containing sequence YEGEESKV, corresponding to residues 11-18 with tyrosine at position 11, resulted in an increased IC50 value of the peptide from 0.075 to 1.4/~M. Tyrosine was included at position 11 for radioiodination in the event of this peptide having a substantially lower IC50. This result contrasts with the cAMP-dependent protein kinase where the amino terminal extension beyond the arginine cluster of the protein kinase inhibitor confers a substantial increase in potency [9]. Our results demonstrate that the optimal length for inhibition by the pseudosubstrate peptide is between 13 and 16 residues in the region between residues 16 and 31. While these findings are important for the design of peptide inhibitors of PK-C it is not yet known to what extent they reflect the properties and function of the pseudosubstrate region in the parent protein. The contribution of individual residues to the potency of the pseudosubstrate peptide PK-C (19-31) was investigated by making a series of substitutions (Table2). Substrate specificity studies on PK-C in several laboratories [2, 1012] have provided evidence that basic residues on either side of the phosphate acceptor site can act as specificity determinants. Substitution of the basic residues in the pseudosubstrate peptide PKC (19-31) all decreased potency, however, there were large differences in the magnitude. Replacement of Arg-19 with alanine caused a 4.7-fold increase in the IC50 to 0.6/zM (Table 2). Previously it was found that replacing Arg-22 with alanine increased the IC50 to 3.1/~M [3]. However, replacement of Arg-22 with alanine caused an approximate 600-fold increase in the IC50 to 81/~M. This residue is separated by two amino acids from Arg-25, the equivalent of the phosphate acceptor site. Previously the importance of amino terminal basic residues had been demonstrated with synthetic
P K - C pseudosubstrate prototope
TABLE 2.
EFFECT OF AMINO ACID SUBSTITUTIONS ON KINETIC PROPERTIES
RI9F2°ARKGA25LRQKN3°VHEVKN 36 Peptide sequence
IC50 (#M)
PKC (19-31) [A~9]PKC (19-31) [A22] PKC (19-31) [A23] PKC (19-31) [V24] PKC (19-31) [A26] PKC (19-31) [A27] PKC (19-31)
0.13+0.02 0.60+0.07 81 +9 1.47 + 0.09 1.7+0.4 2.23+0.29 3.1_+0.6
The synthetic peptide residues are numbered according to the protein sequence reported by Knopf et al. [14] as described in Table 1. The length of the peptides are indicated in parentheses and amino acid substitutions are given in square brackets. The kinetic parameters were determined as described under Materials and Methods and values given are mean _ SEM for three independent experiments. peptides, however, not all phosphorylation site sequences PK-C are arranged in this way. The lysine residue at position 23 influenced the IC50, since substitution with alanine resulted in a 10fold increase in the IC50. Lysine residues are particularly prominent in the local phosphorylation site sequence of the recently described M A R C K S protein [12] which is a potent substrate for PK-C. The glycine residue at position 24 was substituted with the bulkier residue valine because glycine residues in the cAMPdependent protein kinase inhibitor adjacent to the critical arginine residues have an influence on the potency [13]. Substitution at Gly-24 caused a 14-fold increase in the IC5o equivalent to changing one of the basic residues in the pseudosubstrate peptide. The detrimental effect of valine in this position suggests that flexibility of the peptide may be important either in the correct orientation of the critical basic residues or in accommodating Ala-25 in the active site. Replacement of Gln-28 with alanine caused a decrease in potency (Table 2) when substitution with a smaller residue may have been expected to increase potency as seen for Gly-24. The results obtained in the present study demonstrate that the factors influencing the potency of
189
PK-C pseudosubstrate inhibitor are similar to those that determine substrate specificity for peptide phosphorylation [2]. The potency of the pseudosubstrate peptide PK-C (19-31) results from the contribution of multiple residues. All substitutions had an effect with only the [A tg] PK-C (19-31) analogue having less than a 10fold increase in IC50 over the parent peptide. This is in marked contrast to other protein kinases such as the myosin light chain kinase and the cAMP-dependent protein kinase where some residues can be substituted without modifying the kinetic properties. In view of the importance of the basic residues in the pseudosubstrate region it can be expected that changing these to alanine by point mutations would render the enzyme constitutively actively. The very large difference in the IC5o of the peptide PK-C (19-31) and the Ala-22 analogue suggests that these peptides could be used effectively as an inhibitor and analogue control of PK-C, respectively. Acknowledgements--This work was supported by grants from the Victorian Anti-Cancer Council and the NH and MRC.
REFERENCES I. Nishizuka Y. (1988) Nature 334, 661-665. 2. House C., Wettenhall R. E. H. and Kemp B. E. (1987) J. biol. Chem. 262, 772-777. 3. House C. and Kemp B. E. (1987) Science, 238, 1726--1728. 4. Ono Y., Fujii T., Igarshi K., Kuno T., Tanaka C., Kikkawa U. and Nishizuka Y. (1989) Proc. natn. Acad. Sci. U.S.A. 86, 4868-4871. 5. Bosma M. W. and Hille B. (1989) Proc. natn. Acad. Sci. U.S.A. 86, 2843-2947. 6. Maliow R., Schulman H. and Tsien R. W. (1989) Science 245, 862-866. 7. Schapp D., Parker P. J., Bristol A., Kriz R. and Knopf J. (1989) FEBS Lett., 243, 351-357. 8. Kemp B. E. (1988) Molecular Aspects of Cellular Regulation, Voi. 5 (P. Cohen and C. Klee, Eds). Elsevier, Amsterdam, 195-224. 9. Kemp B. E., Cheng H. C. and Walsh D. A. (1988) Meth. Enzym. 159, 173-183. 10. Woodgett J. R. and Hunter T. (1986), Fur. J. Biochem. 161, 177-184. I I. Ferrari S., Marchiori F., Borin G. and Pinna L. A. (1985) FEBS Lett. 184, 72-77.
190
C. House and B. E. KEMP
12. Graft J. M., Stumpo D. O. and Black,shear P. J. (1989) J. biol. Chem. 264, 11,912-11,919. 13. Glass D. B,, Chen H. C., Mende-Mueller L., Reed J. and Walsh D. A. (1989) J. biol. Chem.
264, 8802-8810.
14. Knopf J. L., Lee M. H., Sultzman L. A., Kliz R. W., Loomis C. R., Hewick R. M. and Bell R. M. (1986) Cell 46, 491-502.