HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Marin, O. Articles by Pinna, L. A. Search for Related Content PubMed PubMed Citation Articles by Marin, O. Articles by Pinna, L. A. Social Bookmarking                      What's this?

J Biol Chem, Vol. 274, Issue 41, 29260-29265, October 8, 1999

Tyrosine Versus Serine/Threonine Phosphorylation by Protein Kinase Casein Kinase-2
A STUDY WITH PEPTIDE SUBSTRATES DERIVED FROM IMMUNOPHILIN Fpr3^*

Oriano Marin, Flavio Meggio, Stefania Sarno, Luca Cesaro, Mario A. Pagano, and Lorenzo A. Pinna

From the Dipartimento di Chimica Biologica and Centro di Studio delle Biomembrane del C.N.R., Università di Padova, viale G. Colombo 3, 35121 Padova, Italy

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Protein kinase casein kinase-2 (CK2) is a spontaneously active, ubiquitous, and pleiotropic enzyme that phosphorylates seryl/threonyl residues specified by multiple negatively charged side chains, the one at position n + 3 being of crucial importance (minimum consensus S/T-x-x-E/D/S(P)/T(P). Recently CK2 has been reported to catalyze phosphorylation of the yeast nucleolar immunophilin Fpr3 at a tyrosyl residue (Tyr¹⁸⁴) fulfilling the consensus sequence of Ser/Thr substrates (Wilson, L.K., Dhillon, N., Thorner, J., and Martin, G.S. (1997) J. Biol. Chem. 272, 12961-12967). Here we show that, by contrast to other tyrosyl peptides fulfilling the consensus sequence for CK2, a peptide reproducing the sequence around Fpr3 Tyr¹⁸⁴ (DEDADIY¹⁸⁴DEEDYDL) is phosphorylated by CK2, albeit with much higher K_m (384 versus 4.3 µM) and lower V_max (8.4 versus 1,132 nmol·min¹·mg¹) than its derivative with Tyr¹⁸⁴ replaced by serine. The replacement of Asp at position n + 1 with alanine and, to a lesser extent, of Ile at n  1 with Asp are especially detrimental to tyrosine phosphorylation as compared with serine phosphorylation, which is actually stimulated by the Ile to Asp modification. In contrast the replacement of Glu at n + 3 with alanine almost suppresses serine phosphorylation but not tyrosine phosphorylation. It can be concluded that CK2 is capable to phosphorylate, under special circumstances, tyrosyl residues, which are specified by structural features partially different from those that optimize Ser/Thr phosphorylation.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Protein kinase CK2¹ (also termed "casein kinase-2 or II") is a ubiquitous Ser/Thr-specific protein kinase, essential for cell viability, independent of second messengers or phosphorylation for activation, generally composed of two catalytic ( and/or ') and two noncatalytic  subunits and implicated in cell proliferation and transformation (for review, see Refs. 1-5). The elucidation of the crystal structure of CK2  subunit from Zea mays, 70% identical to its human counterpart (6), has disclosed the structural features that account for some unusual properties of this kinase, notably high basal activity and ability to use GTP, besides ATP as phosphate donor.

Another striking feature of CK2 is its terrific pleiotropy. The list of its substrates includes more than 160 proteins, many of which are implicated in signal transduction, gene expression, protein synthesis, and cell division and proliferation. Consistent with this wide spectrum of physiological targets is the subcellular localization of CK2, which is found in nearly all the subcellular compartments, including the outer side of the plasma membrane (7) and with the possible exception of the Golgi apparatus (8), although it is especially abundant in nuclei (9) and in the soluble cytosol (2). A common denominator of CK2 substrates is the presence of highly acidic phosphoacceptor sites in them, almost invariably fulfilling the consensus S/T-x-x-E/D, but generally including several additional negatively charged side chains (either carboxylic or phosphorylated residues) in addition to or, in few instances, instead of the crucial one at position n + 3 (10). The requirement of these acidic clusters for optimal phosphorylation efficiency has been clearly outlined by systematic studies with peptide substrates (11-14), and a number of unique basic residues responsible for the binding of individual specificity determinants (with special reference to those at positions n + 3 and n + 1) have been identified by site-directed mutagenesis of CK2 (15-17). In the course of these studies on CK2 specificity, some clues were provided that tyrosyl residues might bind to the active site instead of serine and threonine; thus random polymers including tyrosine and glutamic acid are much more powerful inhibitors than just poly-Glu (18) and pseudosubstrate peptides in which tyrosine has been replaced for serine inhibit CK2 more efficiently than their counterparts where serine had been replaced by alanine (1). Consequently the peptide EEEEEYEEEEEEE is a powerful inhibitor (19) whose binding to CK2 is closely reminiscent to that of a canonical substrate (20). Neither in the case of this peptide nor in that of poly(Glu₄Tyr), however, could any significant phosphorylation of tyrosine be detected, supporting the view that CK2 is a bona fide Ser/Thr-specific protein kinase unable to catalyze phosphate transfer to tyrosyl resides once these are bound in its active site.

Recently, however, a yeast nucleolar protein, immunophilin Fpr3, has been found to be tyrosine-phosphorylated (21), although in Saccharomyces cerevisiae bona fide tyrosine kinases are lacking (22); the kinase responsible for this event was shown to be CK2 (23). The tyrosyl residue phosphorylated in Fpr3 either in vivo or in vitro by CK2 (Tyr¹⁸⁴) is embedded in a very acidic cluster (DEDEDADIYDSEDYD) and fulfills the optimal sequence for CK2 phosphorylation for including in addition to the crucial acidic determinant at n + 3 several other acidic residues nearby. Consequently a peptide substrate reproducing this sequence was also phosphorylated by CK2 at Tyr¹⁸⁴. Such a phosphorylation was improved by previous phosphorylation of the seryl residue at position n + 2 or by its replacement with a glutamic acid, suggesting that in this special case the determinant at position n + 2 might play a relevant role. The relevance of the other residues nearby, however, was not assessed nor was the phosphorylation of the tyrosyl peptide compared with that of canonical seryl peptide substrates of CK2. Here we show that the phosphorylation efficiency of the Fpr3 tyrosyl peptide is very low compared with that of its seryl counterpart, both in terms of V_max and K_m and depending on structural features that are partially distinct from those determining the phosphorylation of canonical substrates.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- DEAE-cellulose filters (NA45) were from Schleicher & Schuell. P81 phosphocellulose papers were from Whatman. [-³²P]ATP (3,000 Ci/mmol) was purchased from Amersham Pharmacia Biotech. All other reagents were from Sigma. Native CK2 was purified from rat liver cytosol as described previously (24). Recombinant forms of CK2 either wild type or its mutants defective in substrate recognition were obtained as described elsewhere (16).

Synthetic Peptides-- The tyrosyl peptides DEDADIYDEEDYDL, DEDADIYDS(p)EDYDL, DEDADIYAEEDYDL, DEDADIYDAEDYDL, DEDADIYDEADYDL, DEDADDYDEEDYDL, DEDADIFDEEDYDL, and RRRADDYDDDDD as well as their seryl analogues DEDADISDEEDYDL, DEDADISAEEDYDL, DEDADISDEADYDL, and DEDADDSDEEDYDL were synthesized by an automated peptide synthesizer ABI 431-A (Applied Biosystems) on 4-hydroxymethyl-phenoxymethylcopolystirene-1% divinylbenzene resin (1.04 mmol/g). The synthesis of the peptides EEEEEYEEEEEEE and PEGDYEEELE was described previously (19, 25). The synthesizer was equipped for using Fmoc chemistry (26) and 2-(1H-benzotriazole-1-yl)-1,1,1,3,3 tetramethyluronium hexafluorophosphate/N-hydroxybenzotriazole protocol (27) in a 0.05 mmol scale.

The incorporation of phosphate into the Ser residue has been performed according to a building block approach using Fmoc-Ser(PO(benzyl))OH (from Novabiochem) during the synthesis.

Cleavage of peptides from the resin and side chain deprotection was carried out by treatment of the peptidyl resin with trifluoroacetic acid/anisole/ethanedithiol/ethylmethylsulfide (95:3:1:1). In the presence of the benzyl protecting group on phosphoserine a mixture of trifluoroacetic acid /triisopropylsilane/water (95:2.5:2.5) was preferred.

The crude peptides were purified by high-performance liquid chromatography (HPLC) on a preparative reverse phase column Prep Nova-Pak HR C18, 6 µm, 25 × 10 mm (Waters). The analytical HPLC and matrix-assisted laser desorption ionization/time of flight mass spectrometry analysis of the purified peptides showed a correct sequence and a 95% purity.

Phosphorylation Assay-- Synthetic peptide substrates (0.5 mM) were phosphorylated by incubation in a medium (25 µl final volume) containing 50 mM Tris-HCl buffer, pH 7.5, 10 mM MgCl₂, 50 mM KCl, and 100 µM [-³²P]ATP (specific radioactivity 500-1,000 cpm/pmol) unless otherwise indicated. The reaction was started with the addition of the enzyme (0.23 µg of either native or recombinant CK2) and stopped by the addition of 10 µl of 0.25 M EDTA and cooling in ice. Phosphorylation of Fpr3-derived synthetic peptides was evaluated according to the procedure described by Wilson et al. (23) by using DEAE-cellulose filters. Phosphorylation of RRRADDSDDDDD and of its tyrosyl derivative was quantitated by the phosphocellulose paper procedure (28). Control experiments were also performed with both types of peptides by determining phosphate incorporation after conversion of [-³²P]ATP into P_i and phosphomolybdic complex extraction as described previously (29).

HPLC Separation of Tyrosine-phosphorylated Fpr3 Peptide-- The peptide DEDADIYDEEDYDL, previously phosphorylated by CK2 as described above, was loaded onto a SuperPac Pep-S (Amersham Pharmacia Biotech) C2/18 (5 µm) column (4.6 × 250 mm). Elution was carried out at a flow rate of 1 ml/min with a gradient of acetonitrile in 0.08% trifluoroacetic acid from 0 to 40% for 30 min and then from 40 to 100% for 10 min. The effluent was monitored by absorbance measurements at 220 nm. The radioactivity of the fractions collected (1 ml) was determined by counting in a liquid scintillation apparatus (Canberra-Packard).

Molecular Modeling-- Calculations were made on an Indy R4000 (Silicon Graphics) workstation. For human CK2 subunit modeling, the crystal structure of Zea mays CK2 was used (Protein Data Bank number 1A6O). The two sequences were aligned using the Insight II package (Molecular Simulation Inc.) with the protein homology tools. For elaboration of the model of human CK2, the program Modeler (30) was used. Five structures were obtained from Modeler, and the one with the highest structural homology with respect to maize CK2 was used further. The backbone of the central part of the Fpr3 peptide (DIYDEE) has been shaped using as a template the equivalent sequence of protein kinase inhibitor bound to the catalytic subunit of PKA (Protein Data Bank number 1atp). The peptide was manually placed in CK2 catalytic pocket avoiding physical interference with residues of the kinase. The obtained structure was minimized using the Discover module of Insight II package until the energy reached a minimum.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The time courses for CK2-catalyzed phosphorylation of the Fpr3 peptide with Glu at position n + 2 relative to Tyr¹⁸⁴ instead of serine and of its derivative where Tyr¹⁸⁴ was replaced by serine are compared in Fig. 1A. It can be seen that the phosphorylation of the tyrosyl peptide is negligible as compared with that of its seryl counterpart, especially after short incubation times. By prolonging incubation the phosphorylation of the seryl peptide reaches a plateau when ATP is nearly exhausted, whereas the phosphorylation of the tyrosyl peptide slowly increases up to 2 h of incubation. By increasing ATP concentration (1 mM) over that of the peptides (0.5 mM) a stoichiometric phosphorylation of the seryl peptide was attained after 30 min of incubation, whereas the phosphorylation of the tyrosyl peptide was still far from stoichiometric (around 0.03 mol/mol) after 2 h (not shown). Albeit slow, the phosphorylation of the Fpr3 tyrosyl peptide was quite significant especially if compared with that of another tyrosyl peptide featured after the optimal peptide substrate routinely used for assaying CK2 (Fig. 1B). In this case, no detectable phosphorylation of the tyrosyl peptide could be observed, although that of its seryl counterpart was even faster than that of the seryl derivative of the Fpr3 peptide. The stoichiometry of the Fpr3 tyrosyl peptide phosphorylation was confirmed by HPLC where the phosphoderivative can be resolved from the nonphosphorylated peptide (Fig. 2), thus allowing their relative quantitation (~3% versus 97%).

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Time course phosphorylation of seryl peptides and their tyrosyl counterparts by CK2. Phosphorylation was performed for the times indicated and evaluated as described under "Experimental Procedures." The concentrations of ATP and of peptide substrates were 0.2 mM and 0.5 mM, respectively.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   HPLC separation of tyrosine-phosphorylated and unphosphorylated forms of the Fpr3-derived peptide DEDADIYDEEDYDL. The peptide was phosphorylated by CK2 for 90 min as described under "Experimental Procedures" and directly subjected to HPLC analysis. The absorbance at 220 nm (A) and the radioactivity incorporated (B) are shown. The arrows denote the position of the phosphorylated (left) and unphosphorylated (right) peptide, respectively.

The suspicion that tyrosine phosphorylation in the Fpr3 peptide might rely on structural features different from those determining serine phosphorylation was corroborated by the finding that two other tyrosyl peptides, in addition to RRRADDYDDDDD, namely EEEEEYEEEEEEE and PEGDYEEELE, are totally unaffected by CK2 despite the fact that they fulfill optimal conditions for Ser/Thr phosphorylation (see Table I).

To shed light on the uncommon phosphorylation of tyrosine in the Frp3 peptide a number of derivatives with individual substitutions were synthesized and compared with the parent peptide. This latter includes a glutamic acid at position n + 3, which is the main determinant for CK2 recognition (13) and an aspartic acid at position n + 1 where also an acidic residue is very important (17). Position n + 2 is occupied in the wild type peptide by a serine, whose primary phosphorylation by CK2 is believed to improve subsequent phosphorylation of tyrosine, because its replacement by alanine, but not by glutamic acid, was detrimental (23); thus it appeared that a negatively charged side chain at n + 2, a position rather unimportant for serine phosphorylation (13), might be crucial for tyrosine phoshorylation. In contrast, position n  1 in the Fpr3 site is occupied by isoleucine, a rare feature in CK2 sites, where an acidic residue was shown to be preferred and is actually found in about 50% of the natural substrates (10). We therefore decided to check how much the substitution of this Ile by Asp, of Asp (+1) by Ala, of Glu (+2) by Ala or phosphoserine, and of Glu (+3) by Ala might affect phosphorylation by CK2. As shown by the time courses in Fig. 3, the most detrimental substitution (a part of Tyr¹⁸⁴, confirming that this one and not the other tyrosine, Tyr¹⁸⁹ undergoes phosphorylation) is that of alanine instead of glutamic acid at n + 1, followed by aspartic acid instead of isoleucine at n  1, whereas the replacement of the most important acidic determinant in canonical CK2 substrates, i.e. Glu at n + 3, by alanine is in this case much better tolerated. In contrast to a previous report (23), the replacement of Glu at n + 2 by alanine has no effect, whereas a phosphoserine in this position significantly improves phosphorylation (Fig. 3). As shown in Table I, the differences in phosphorylation outlined by the time courses of Fig. 3 are variably accounted for by changes in V_max and/or K_m. Especially remarkable is the drop in phosphorylation efficiency due to both a dramatic decrease of V_max and an increase of K_m promoted by the Ala for Asp substitution at n + 1. In comparison, the effect of a similar substitution at the "crucial" n + 3 position, due to just a 2-fold increase in K_m, is negligible. The detrimental effect of replacing Ile at n  1 with Asp (see Fig. 3), appears to be be exclusively due to a drop in V_max (Table I).

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Effect of amino acid substitutions on the phosphorylation of the Fpr3 tyrosyl peptide by CK2. All peptides tested were 0.5 mM. Phosphorylation conditions and evaluation of the phosphate incorporated were performed as described under "Experimental Procedures." Substitutions relative to the reference peptide (DEDADIYDEEDYDL) are bold type. The phosphorylated tyrosine is underlined.

Table I also highlights the striking superiority of the seryl derivative of the Fpr3 peptide over the tyrosyl peptide both in terms of V_max (>130-fold higher) and of K_m (almost 90-fold lower), and it corroborates the view that the specificity determinants are different with the two peptides. In fact the replacement of Ile at n  1 by Asp, far from being detrimental, is actually improving the phosphorylation efficiency of the seryl derivative and the replacement at position n + 3 is more detrimental than that at n + 1.

The particularly crucial role of the acidic determinant at position n + 1 as opposed to that at position n + 3 in determining tyrosine phosphorylation was also confirmed by experiments with CK2 mutants, which are defective in the recognition of individual determinants. As shown in Fig. 4, mutant K198A with reduced ability to bind the determinant at position +1 (17) is much more defective toward the tyrosyl peptide than it is toward its seryl derivative. The opposite applies to another mutant, K74A/K77A, which is unable to recognize the determinant at n + 3 (16).

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Phosphorylation of tyrosyl versus seryl Fpr3-derived peptides by wild type CK2 and CK2 mutants defective in substrate recognition. 0.23 µg of either wild type or mutated CK2 were incubated at 37 °C in the presence of 0.12 mM DEDADISDEEDYDL (solid bars) or 1.2 mM DEDADIYDEEDYDL (empty bars) Fpr3 derivatives for 5 and 40 min, respectively, and 32P incorporated was determined as described under "Experimental Procedures." CK2 activity is expressed as percentage of the activity measured with CK2 wild type, which was 271 and 38 pmol for the seryl and the tyrosyl peptide, respectively.

To gain further information about the mode of binding of Tyr¹⁸⁴-containing peptides to CK2 active site a modelization analysis has been performed using as a template a structure of human CK2 derived from 70% identical maize CK2 crystallized in the absence of phosphoacceptor substrate (6). The central part of the parent Fpr3 peptide (-DIY¹⁸⁴DEE-) shaped after the pseudo-substrate peptide bound to the catalytic subunit of PKA (31) has been placed into the active site of CK2. After energy minimization, a model was obtained (Fig. 5) that is largely consistent with experimental observations. It should be noted in particular that the tyrosyl residue protrudes into the active site with its hydroxyl group in the proximity of two crucial elements of catalysis, namely the catalytic base Asp¹⁵⁶ (homologous to PKA Asp¹⁶⁶) and Lys¹⁵⁸, homologous to PKA Lys¹⁶⁸ believed to bridge the phosphate to the peptide during the catalytic event (32). The distances between the phosphoacceptor hydroxyl group of tyrosine and either Asp¹⁵⁶ (2.67 Å) or Lys¹⁵⁸ (2.89 Å) are comparable with those found in the complex between PKA and a seryl peptide substrate (2.7 and 3.0 Å, respectively) (32). The only evident interaction with peptide substrate side chains is the one between the carboxylate of aspartate at position n + 1 and positively charged groups in the loop p + 1 (Lys¹⁹⁵ and Lys¹⁹⁸). In contrast, the side chain of glutamate at position n + 3 is far away from the 74-77 basic cluster, which is responsible for tight interaction with the n + 3 determinant of seryl peptides (17). This is also in agreement with the relatively modest effect of replacing Glu (n + 3) as compared with the dramatic drop of efficiency promoted by Asp (n + 1) substitution (see Table I). In our model, the side chain of Ile (n  1) protrudes outside the surface of the kinase. This, however, is likely to be an artifact of modelization where the water surrounding CK2, expected to repel the hydrophobic side chain of isoleucine, is not considered.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Modelization of Fpr3-derived 181-188 tyrosyl peptide bound to the catalytic site of human CK2. Drawn from human CK2 modeling based on the crystal structure of maize CK2 (see "Experimental Procedures"). The peptide is shown in green, and the CK2 elements in light blue. The peptide residues are indicated according to their position relative to the phosphorylatable tyrosine (Tyr184 of Fpr3). Distances between key residues at active site and peptide residues were as follows: Asp156 to substrate Tyr-OH, 2.67 Å; Lys158 to substrate Tyr-OH, 2.89 Å; Lys198 to side chain carboxylate of substrate Asp (n + 1), 4.11 Å; Arg195 to side chain carboxylate of substrate Asp (n + 1), 2.46 Å.

It should be finally outlined that the Tyr versus Ser/Thr specificity of CK2 appears not to be affected by a number of effectors and/or experimental conditions, which are known to impinge on CK2 activity. In particular CK2 catalytic activity toward the Frp3 tyrosyl and seryl peptides is equally stimulated by association with the  subunit (about 3-fold) and by addition of 336 nM polylysine to the holoenzyme (about 2-fold) and displays similar activation curves by either Mg²⁺ or Mn²⁺. With both peptides, GTP can replace ATP as phosphate donor and the K_m with respect to ATP (33 µM) is not modified by replacing one peptide with the other (data not shown).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Nearly all members of the huge eukaryotic protein kinase family fall into two categories, Ser/Thr- and Tyr-specific enzymes, respectively (33). In most instances the borderline between these two groups, empirically drawn using substrates of the two sorts, is quite sharp and clear-cut, being also based on differences in primary structure motifs (34). In a few instances, however, protein kinases may display a dual specificity in that they are able to phosphorylate both Ser/Thr and Tyr residues (35). The only known sensu stricto dual specificity protein kinases are probably those of the mitogen-activated protein kinase kinase family, which are highly dedicated kinases that catalyze the simultaneous phosphorylation of both Thr and Tyr in a conserved sequence, T-x-Y present in the activation loop of mitogen-activated protein kinases. In several instances, however, bona fide Ser/Thr protein kinases have been reported to display tyrosine kinase activity as well under certain circumstances, detectable either as autophosphorylation or toward trans substrates. In general the latter is negligible compared with that observed toward canonical Ser/Thr sites. The list of these Ser/Thr kinases exhibiting low Tyr kinase activity as well includes among others phosphorylase kinase (36), calcium-calmodulin-dependent protein kinase II and Spk1 (37), CK1 (38, 39), CK2 (40), and even PKA (37). Conversely it should be noted that also a classical tyrosine kinase like Src can phosphorylate aliphatic as well as aromatic alcohols in a stretch of peptide (41). It is likely therefore that in several instances this hardly detectable "dual" specificity of bona fide Ser/Thr protein kinases is a mechanistic curiosity devoid of practical consequences. That this may not always be the case, however, is supported by the observation that in lower eukaryotes, like yeast, where bona fide tyrosine kinases are absent, tyrosine-specific protein phosphatases are nevertheless present (22), and tyrosine phosphorylation of proteins other than kinases of the mitogen-activated protein kinase family has been reported to take place. It is tempting therefore to assume that tyrosine phosphorylation by bona fide Ser/Thr kinases is a physiologically relevant event in these cases.

A notable example of this situation is provided by immunophilin Fpr3, an abundant nucleolar protein, a member of the FK506-binding subfamily of immunophilins (42), which is a substrate of S. cerevisiae protein-tyrosine phosphatase Ptp1. Phosphotyrosine dephosphorylation by Ptp1 prevents nucleolar accumulation of Fpr3 in vivo (21). The tyrosyl residue phosphorylated in Fpr3 was identified as Tyr¹⁸⁴, residing in a highly acidic sequence fulfilling optimal conditions for CK2 catalyzed phosphorylation, and indeed CK2 was shown to be responsible for Tyr phosphorylation of Fpr3 and to be able to phosphorylate in vitro a tyrosyl peptide reproducing the sequence around Tyr¹⁸⁴ (23). This observation would include CK2 in the growing list of sensu lato dual specificity protein kinases, despite the previous failure to demonstrate any activity of CK2 toward tyrosyl peptides fulfilling its consensus sequence. This prompted us to undertake a study of the structural features underlying tyrosine phosphorylation by CK2 and to compare this activity with that toward canonical Ser/Thr substrates. The main outcomes of this investigation are the following.

1) The phosphorylation of Fpr3-derived tyrosyl peptide, albeit quite significant, is negligible compared with that of an identical peptide where tyrosine is replaced by serine. This shows that even in this case serine is by far preferred over tyrosine by CK2. The superiority of the seryl peptide is due to both much lower K_m and much higher V_max. The low V_max of the tyrosyl peptide suggests that its phosphorylated product leaves the active site of the kinase very slowly, a situation that could give rise to remarkable product inhibition. Also, consistent with this interpretation would be the low phosphorylation stoichiometry of the tyrosyl peptide (<5%). It is possible that in vivo the kinetics of Tyr¹⁸⁴ phosphorylation are improved either by features intrinsic to the full-length Fpr3 protein conformation or by the fast translocation of phospho-Fpr3 to separate compartment(s). In this respect a comparison between the kinetic constants of Tyr¹⁸⁴ peptides and those of full-length Fpr3 would be misleading since Tyr¹⁸⁴ is just a minor Fpr3 phosphoacceptor site (23); consequently the parameters calculated with Fpr3 would mainly reflect the phosphorylation of residues other than Tyr¹⁸⁴. It is worth noting, however, that a significant phosphorylation of Tyr¹⁸⁴ by CK2 is observable under conditions where GST-Fpr3 is about 1 µM (23), although at this concentration the phosphorylation of the best peptide substrate, whose K_m is 370 µM, would be flatly undetectable. It is likely therefore that indeed the phosphorylation efficiency of Tyr¹⁸⁴ is improved by features that are lacking in the peptides used in our study. On the other hand the reason for the discrepancy between our data and that reported by Wilson et al. (23) where a 1 to 1 phosphorylation stoichiometry of the same peptide is reported to occur after less than 2 h of incubation with recombinant human CK2, is not clear. It should be noted, however, that the stoichiometry reported in that paper is not consistent with the experimental conditions used, where ATP and the peptide are 100 and 580 µM, respectively, allowing at best a 20% phosphorylation stoichiometry.

2) CK2 activity ratio toward the tyrosyl versus the seryl derivative of the Fpr3 peptide is not significantly affected by either the subunit composition of CK2 (catalytic subunit versus holoenzyme), or the addition of polycationic stimulators, or the nature of the phosphate donor (GTP versus ATP) and of the activatory cation (Mn²⁺ versus Mg²⁺). This last outcome is worth noting, because in the case of phosphorylase kinase, Mn²⁺ instead of Mg²⁺ is required to see tyrosine-phosphorylating activity (36).

3) Ability to phosphorylate tyrosine is not a general feature of CK2 toward any kind of substrate fulfilling its consensus sequence, because the replacement of serine (or threonine) by tyrosine in several outstanding peptide substrates of CK2 different from the Fpr3-derived peptide rendered their phosphorylation undetectable.

4) Among the local structural determinants favoring tyrosine versus serine phosphorylation, the isoleucine at position n  1 and the acidic residue at position n + 1 appear to play a prominent role. Interestingly, hydrophobic residues are only sporadically found in CK2 substrates at position n  1, where instead acidic residues are quite frequent (10). Consequently the replacement of Ile by Asp at this position in the seryl derivative of the Fpr3 peptide increases its phosphorylation efficiency, but it is detrimental in the tyrosyl peptide. The molecular basis for such an opposite behavior is presently unclear. The favorable effect of an acidic residue at position n + 1, on the other hand, is a well known feature of all CK2 phosphoacceptor sites, including Ser/Thr (10, 12). In the latter case, however, the n + 1 acidic determinant is generally less important than the one at n + 3 (10, 12), whereas the opposite applies to the Fpr3 tyrosyl peptide where the replacement of the n + 3 Glu by alanine is exceptionally well tolerated. Computer-aided modelization (Fig. 5) corroborates the concept that in the case of Fpr3-derived tyrosyl peptide the acidic determinant at n + 1 is much more important than the one at n + 3 in making contacts with the enzyme. The replacement of glutamate at n + 2 with phosphoserine improves 2-3-fold the V_max and phosphorylation efficiency of the peptide, consistent with the knowledge that phosphoserine is a better determinant than Glu especially at positions n + 1 and n + 2 (14); it is quite possible therefore that, as suggested by others (23), the phosphorylation of Fpr3 in vivo is enhanced by a hierarchical effect where previous phosphorylation of Ser¹⁸⁶ will improve subsequent phosphorylation of Tyr¹⁸⁴. The replacement of Glu at n + 2 with Ala, however, has only a modest effect, similar to that caused by the same substitution in canonical seryl peptide substrates of CK2 (12). therefore, it seems unlikely that the phosphorylation of Tyr¹⁸⁴ in the Fpr3 peptide is entirely dependent on the presence of a negatively charged side chain at n + 2, as suggested by others (23). The reasons why the same peptide, DEDADIYDAEDYDL, which in our hands is nearly as good as the parent peptide with Glu instead of the underlined Ala (see Fig. 3 and Table I), was not phosphorylated at all (23) is unclear to us.

5) The different mode of binding of tyrosyl versus seryl peptide substrates was also corroborated by experiments with CK2 mutants with a reduced ability to recognize individual determinants. Especially remarkable is the behavior of the mutant K198A, which is much more defective toward the Fpr3 tyrosyl peptide as compared with its seryl derivative.

In conclusion, CK2 can be legitimately included in the list of sensu lato dual specificity protein kinases. Clearly, however, its ability to phosphorylate tyrosine in peptide substrates is very modest as compared with its canonical Ser/Thr kinase activity and, more important, it depends on structural features that are partially different from those determining Ser phosphorylation. Among these the most striking are the absolute requirement for the acidic determinant at position n + 1 as opposed to the tolerance toward the replacement of the one at position n + 3, and the preference for a hydrophobic over an acidic residue at position n  1.

	FOOTNOTES

^* This work was supported by Grants from Consiglio Nazionale delle Richerche (97.03614.PS14 and T.P on Biotechnology) (to L. A. P.), from Ministero per l'Universitá e per la Ricerca Scientifica e Techologica (PRIN 1997) (to L. A. P. and F. M.), from Associazone Italiana per la Ricerca sul Canero, Italian Ministry of Health (Progetto AIDS, Istituto Superiore di Sanità), the European Commission (Biomed-2 BMH4-C&96-0047), and from the Armenise-Harvard Foundation.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Dipartimento di Chimica Biologica, viale G. Colombo 3, 35121 Padova, Italy. Tel.: 39-49-8276108; Fax: 39-49-8073310; E-mail: pinna@civ.bio.unipd.it.

	ABBREVIATIONS

The abbreviations used are: CK2, casein kinase 2; Fmoc, N-(9-fluorenyl)methoxycarbonyl; HPLC, high-performance liquid chromatography; PKA, protein kinase A.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				G. Panasyuk, I. Nemazanyy, A. Zhyvoloup, M. Bretner, D. W. Litchfield, V. Filonenko, and I. T. Gout Nuclear Export of S6K1 II Is Regulated by Protein Kinase CK2 Phosphorylation at Ser-17 J. Biol. Chem., October 20, 2006; 281(42): 31188 - 31201. [Abstract] [Full Text] [PDF]

				S. Sarno, P. Ghisellini, and L. A. Pinna Unique Activation Mechanism of Protein Kinase CK2. THE N-TERMINAL SEGMENT IS ESSENTIAL FOR CONSTITUTIVE ACTIVITY OF THE CATALYTIC SUBUNIT BUT NOT OF THE HOLOENZYME J. Biol. Chem., June 14, 2002; 277(25): 22509 - 22514. [Abstract] [Full Text] [PDF]

				C. Eichwald, F. Vascotto, E. Fabbretti, and O. R. Burrone Rotavirus NSP5: Mapping Phosphorylation Sites and Kinase Activation and Viroplasm Localization Domains J. Virol., March 7, 2002; 76(7): 3461 - 3470. [Abstract] [Full Text] [PDF]

				D. Chaillot, N. Declerck, K. Niefind, D. Schomburg, T. Chardot, and J.C. Meunier Mutation of recombinant catalytic subunit {alpha} of the protein kinase CK2 that affects catalytic efficiency and specificity Protein Eng. Des. Sel., April 1, 2000; 13(4): 291 - 298. [Abstract] [Full Text] [PDF]

				R. L. Trott, M. Kalive, Z. Paroush, and A. P. Bidwai Drosophila melanogaster Casein Kinase II Interacts with and Phosphorylates the Basic Helix-Loop-Helix Proteins m5, m7, and m8 Derived from the Enhancer of split Complex J. Biol. Chem., January 12, 2001; 276(3): 2159 - 2167. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Marin, O. Articles by Pinna, L. A. Search for Related Content PubMed PubMed Citation Articles by Marin, O. Articles by Pinna, L. A. Social Bookmarking