How do kinases transfer phosphoryl groups? Allan Matte†, Leslie W Tari and Louis TJ Delbaere* Understanding how phosphoryl transfer is accomplished by kinases, a ubiquitous group of enzymes, is central to many biochemical processes. Qualitative analysis of the crystal structures of enzyme–substrate complexes of kinases reveals structural features of these enzymes important to phosphoryl transfer. Recently determined crystal structures which mimic the transition state complex have added new insight into the debate as to whether kinases use associative or dissociative mechanisms of catalysis. Address: Department of Biochemistry, University of Saskatchewan, Saskatoon, Saskatchewan S7N 5E5, Canada. †Present address: Biotechnology Research Institute, 6100 Royalmount Avenue, Montreal QC H4P 2R2, Canada.
*Corresponding author. E-mail: [email protected]
Structure 15 April 1998, 6:413–419 http://biomednet.com/elecref/0969212600600413 © Current Biology Ltd ISSN 0969-2126
Kinases are a ubiquitous group of enzymes central to many biochemical processes. The X-ray crystal structures of kinase–substrate complexes have revealed structural features of these enzymes important to phosphoryl transfer, and have added new insight into the debate as to whether kinases use associative or dissociative mechanisms of catalysis. The transfer of the terminal phosphoryl group from one nucleotide to another, or to a small molecule (by enzymes which we will term ‘small molecule kinases’), or to a protein substrate (by protein kinases) is a fundamental process in many aspects of metabolism, gene regulation and signal transduction. With a wealth of experimental data, obtained using many biochemical and structural techniques, this should be among the best understood of enzyme-catalyzed reactions. In view of the enormous variety of proteins which carry out these reactions, two important questions need to be addressed. What structural features of these enzymes and their interactions with substrates are potentially important to the process of phosphoryl transfer? and how do these features relate to the mechanisms and transition states of catalysis? This review focuses primarily on the crystal structures of enzyme–substrate complexes of kinases and, in particular, what these structures tell us about the process of phosphoryl transfer. Many structural parameters may potentially contribute to the process of phosphoryl transfer, including global and local conformational changes upon
substrate binding, various modes of nucleotide binding, substrate orientation and geometry, active-site organization, metal-ion coordination and electrostatic effects (Figure 1). Here, we present an overview of some structural features which relate to the activity of these enzymes. Other reviews on phosphoryl transfer have been published previously [1–5]. Conformational changes upon substrate binding
Comparison of the crystallographic structures of many nucleotide-dependent phosphoryl transfer enzymes in the presence and absence of substrates or substrate analogs reveals both global and local changes in protein structure. This process can be viewed in one of two ways. Either binding of one or more substrates induces a change in conformation or, alternatively, binding of one or more substrates serves to trap the protein in a particular conformational state which otherwise already exists in solution. In this latter view, substrate binding simply shifts the equilibrium distribution of structures to one conformational state over another. Evidence from time-resolved fluorescence studies on adenylate kinase is in agreement with the latter of these two models . Global changes in structure usually involve hingebending motions [7–9] or rotation  of two or more domains towards one another, resulting in the closure of the active-site cleft (reviewed in [11,12]). In adenylate kinases  and phosphoglycerate kinase  there are synergistic effects on global conformational changes, with each substrate inducing a change in structure. These structural changes act to position enzyme sidechains appropriately around the substrates and to sequester the substrates so as to prevent the wasteful hydrolysis of nucleotides or other phosphoryl-containing substrates prior to catalysis. In other enzymes, such as phosphoenolpyruvate carboxykinase, large domain movements are observed upon ATP binding, although evidence for synergistic structural effects based on X-ray crystallography is so far lacking . A unique, swiveling-domain mechanism, involving apparently two conformational states, has been proposed for phosphoryl transfer in pyruvate phosphate dikinase based on the crystal structure of the native enzyme . Nucleoside diphosphate kinase, an enzyme with four or six subunits depending on the source, is unusual in that it does not show large-scale changes in structure upon substrate binding [15–17]. In addition to the global changes in structure accompanying nucleotide binding, local changes in protein structure, as well as alterations to the nucleotide conformation compared to its structure in free solution, may occur. Smaller
Structure 1998, Vol 6 No 4
Second divalent cation (Mn2+ )
Positively charged amino acid residue(s)
Structural parameters influencing phosphoryl transfer by kinases. Nucleotides may be bound with the β- and γ-phosphoryl groups in an eclipsed conformation, resulting in a sterically and energetically strained substrate ready for catalysis. Mg2+ binding may help to promote this structure, as well as positioning the γphosphoryl group with respect to the second substrate so as to create the correct geometry for phosphoryl transfer. Nucleotide-binding motifs (e.g. P loop or glycine-rich loop) also serve to orient the nucleotide in the correct position for phosphoryl transfer. Positively charged residues, including the conserved lysine of the P-loop motif or the positive terminus of an α-helix dipole, may assist in stabilizing the phosphoryl group during transfer. In the case of an associative mechanism, Mg2+ would also contribute by increasing the electrophilicity of the γ-phosphorous atom by withdrawing charge via its interactions with the oxygen atoms. In some enzymes, such as phosphoenolpyruvate carboxykinase, a second metal ion may further increase the electrophilicity of the transferred phosphoryl group, impart additional orientation effects on the substrates, and play a role in forming the transition state. Water molecules which complete the coordination sphere of the metal(s) help anchor
+ H O H– OOC + –
P loop, glycine-rich loop or other motif Purine- or pyrimidine-binding pocket
H O H
O H H
Geometrical alignment for 'in-line' transfer
Nucleotide-associated metal (Mg 2+) Structure
them to the protein, while at the same time permitting changes in metal environment during catalysis. Usually, positively charged sidechains bind and position substrates for catalysis
changes in structure are seen in Escherichia coli phosphofructokinase, where binding of ADP–Mg2+ and fructose1,6-bisphosphate induce a shear motion within a single domain . In phosphoenolpyruvate carboxykinase, significant changes in the adenine-binding region are observed upon complexation with Mg2+–ATP . Crystallographic analysis of several complexes of cAMPdependent protein kinase [12,19] has demonstrated local flexibility in the glycine-rich loop, a conserved element responsible for interacting with ATP in the protein kinase catalytic core. Role of divalent cations
Essentially all kinases require a Mg2+–nucleotide complex as one of the enzyme substrates, an exception being the first partial reaction catalyzed by nucleoside diphosphate kinase, which can proceed independently of Mg2+ . The orientation of the Mg2+ ion relative to the substrates seems to fall into one of two patterns. With bound ATP or GTP, the Mg2+ is usually coordinated to the β- and γ-phosphoryl groups, but may also be coordinated to the α-, β- and γ-phosphates . High-resolution structures have shown that between two and four water molecules make up part of the Mg2+ coordination sphere. The ligands completing the octahedral coordination may be oxygen atoms from the sidechains of threonine, serine, aspartic acid or glutamic acid residues of the protein or, in
through both local and global (domain–domain) motions. It should be noted that any single enzyme would not be expected to show all of the possible features depicted.
an unusual case with UMP/CMP kinase, only water . In phosphoenolpyruvate carboxykinase, Mg2+ coordination orients the β- and γ-phosphoryl groups in a highenergy, eclipsed conformation resulting in increased electrostatic repulsion between the phosphoryl groups which may activate ATP for catalysis [7,21]. Presumably Mg2+ also assists in orienting the γ-phosphoryl group ‘inline’ with respect to the second substrate, creating the correct geometry to complete phosphoryl transfer. In ADP- or GDP-bound structures, in which a second phosphorylated product is present, the coordination of the Mg2+ changes so that in many instances it acts to bridge the β-phosphoryl group of the ADP and the phosphoryl group of the second product. This mode of Mg2+ binding is observed in the complexes of Trypanosoma brucei phosphoglycerate kinase with 3-phosphoglycerate and ADP , phosphofructokinase with fructose 1,6-bisphosphate and ADP , and glycerol kinase with glycerol phosphate and ADP . At least two kinases, pyruvate kinase  and phosphoenolpyruvate carboxykinase , require two divalent cations for full activity. Pyruvate kinase has an additional requirement for a monovalent cation. For both enzymes, Mg2+ is a cosubstrate with the nucleotide, while the second cation, a transition metal, has a separate distinct binding site. Electron paramagnetic resonance (EPR)
Minireview Enzyme-catalyzed phosphoryl transfer Matte, Tari and Delbaere
Figure 2 Structural organization of the bimetal cluster in phosphoenolpyruvate carboxykinase. Detailed stereo view of substrate/metal binding in the active site. Mg2+ is shown in blue, Mn2+ in green and water molecules are in red. Potential hydrogen bonds and/or electrostatic interactions are shown as yellow dotted lines. Both metals possess octahedral coordination spheres which are partially completed by water molecules. Mg2+ and a lysine residue neutralize electrostatic repulsions between phosphate oxygen atoms on ATP, promoting the formation of a high-energy conformer which eclipses the oxygens of the β- and γphosphates and orients the triphosphate chain so that it is aligned towards pyruvate. Mn2+ and an arginine bridge ATP and pyruvate, neutralizing electrostatic repulsions between the anionic substrates, facilitating proper substrate alignment and activating both substrates through direct and watermediated interactions. One arginine electrostatically promotes and stabilizes pyruvate enolate formation via direct ionic interactions with the enolate oxygen, while a
second arginine stabilizes enolate formation through neutralization of the negative charge on the adjacent carboxylate group on
spectroscopy of various substrate and inhibitor complexes of pyruvate kinase indicates that Mn2+ coordinates to both ATP and oxalate, a structural analog of the enolate of pyruvate, the proposed reaction intermediate [25,26]. Two divalent cations are observed at the active site of cAMPdependent protein kinase, although only the Mg2+ ion bound to the β- and γ-phosphoryl groups appears to activate catalysis [27,28]. Recently, the crystal structure of phosphoenolpyruvate carboxykinase with both Mg2+ and Mn2+, as well as pyruvate and ATP, has been solved . This structure provides insight into how both metals participate in the process of synergistic activation of the enzyme (Figure 2). The Mg2+ cation has identical geometry to that found in phosphoenolpyruvate carboxykinase complexed with ATP– Mg2+–oxalate . Mn2+ bridges the γ-phosphoryl group of ATP and pyruvate via two water molecules. In the proposed transition state, Mn2+ displaces the waters to form an inner-sphere complex with the enolate anion. Both metals appear specific for their respective binding sites, which can be explained by their relative Lewis acidity and ionic radii. A similar mode of binding of the two metals was proposed previously using EPR and proton longitudinal relaxation rate measurements [29–31]. Binding and orientation of nucleotide substrates
It is well known that kinases may bind their nucleotide substrates, usually GTP or ATP, in a variety of ways (reviewed in ). One structural motif commonly employed in a variety of ATP- and GTP-binding proteins
pyruvate. Mn2+ acts in concert with both arginines to promote and stabilize enolate formation.
is the phosphate-binding (P-loop) motif [32,33]. In this motif, conserved glycine residues make mainchain contacts with phosphoryl groups, while the oxygen atom of a conserved serine or threonine residue makes up one ligand of the Mg2+ ion. A conserved lysine residue usually coordinates to the β- and γ-phosphoryl groups, and plays a crucial role in phosphoryl transfer. It is not clear whether this lysine, or other basic residue, ‘follows’ the transferred phosphoryl moiety during catalysis, [9,27,32] or rather forms part of a structural template for the reaction . Protein kinase catalytic cores also contain a glycine-rich motif responsible for interacting with ATP which is both different in sequence and structure compared to the P loop . In some kinases, the positive terminus of an α-helix dipole points towards the β- or γ-phosphate, thereby imparting a stabilizing effect during phosphoryl transfer [8,9,18]. There are less definitive sequence or structural patterns relating to ribose and purine or pyrimidine binding. In some kinases hydrophobic interactions sandwich the adenine ring; these interactions may be combined with one or more mainchain, or occasionally sidechain, hydrogen bonds to N1, N6 or N7 of adenine. In other proteins, such as nucleoside diphosphate kinase [15–17], phosphofructokinase , and creatine kinase , no specific polar contacts between the base and protein have been observed. In fact, it appears that mainchain contacts with the bases and the presence of a structural motif independent of sequence may be more important for adenine binding, for example, as described for protein kinases and
Structure 1998, Vol 6 No 4
D-Ala:D-Ala ligase adenine-binding sites, as well as proteins with the ATP grasp fold [37,38]. These observations help to explain why no definitive sequence pattern for adenine binding has emerged, despite the large number of sequences which have been analyzed . Another possibility is a role for structural water molecules in mediating protein–base recognition .
Double displacement (ping-pong) mechanism
Do phosphoryl transfer reactions operate via an associative or dissociative transition state?
There has been considerable debate within the literature as to whether enzyme-catalyzed phosphoryl transfer reactions operate primarily with an associative (SN2-like) or dissociative (SN1-like) transition state. A summary of the two possible enzymatic mechanisms and the associated transition states for phosphoryl transfer is presented in Figure 3. Experimental studies on phosphoryl transfer by small molecules ([40,41] and references cited therein) and model studies on the G protein p21Haras  have led to arguments in favor of a dissociative mechanism of catalysis in kinases and phosphatases. Others have made elegant arguments in favor of these enzymes operating with more of an associative character [2,3]. Various biochemical approaches have been used to tackle this question with enzymes in solution, yielding different results depending on the particular protein. Positional isotope exchange experiments on pyruvate kinase indicate that an associative transition state is operational , while kinetic studies on nucleoside diphosphate kinase  and secondary 18O isotope experiments with hexokinase  in both cases suggest a dissociative transition state. What structural evidence exists from crystal structures of enzyme–substrate complexes of kinases in favor of associative versus dissociative transition states? Many of the arguments in favor of associative [7,8,21,46–48] or dissociative  transition states for phosphoryl transfer have been based on the geometry and reaction coordinate distances between the terminal phosphoryl group and its Figure 3
(b) Associative transition state Nucleotide O
Enolate anion O
Adenosine O P O P O O
O OC O
C CH 2
Pentacoordinate transition state Dissociative transition state
Trigonal planar transition state (metaphosphate) Structure
Mechanisms and transition states for phosphoryl transfer reactions. Two catalytic mechanisms of phosphoryl transfer are commonly encountered, either single (direct) displacement or double displacement (ping-pong) mechanisms. In the single displacement mechanism, all substrates must be bound to the enzyme, forming a ternary complex, before phosphoryl transfer can take place. With the double displacement mechanism, illustrated in (a), the nucleotide initially binds and donates a phosphoryl group to the enzyme, usually at a histidinyl residue, generating a phospho–enzyme intermediate. The enzyme-bound phosphoryl group is in turn donated to a second substrate yielding the final product. Nucleotides are depicted in yellow, the acceptor substrate in purple, and the transferred phosphoryl group in red. (b) For either mechanism described above, the transition state for phosphoryl transfer may be described as one of two possibilities, either associative (SN2-like) or dissociative (SN1-like). In the associative case, the transition state is dominated by bond formation, and involves a pentagonal bipyrimidal (phosphorane) structure with the axial ligands being the nucleophile of the attacking substrate and bridging oxygen atom between the β- and γ-phosphates, respectively. The dissociative case involves a planar, trigonal, metaphosphate (PO3–) structure, and is dominated by bond breaking. The two transition states are further defined by differences in charge distribution, with a net charge of –3 on the phosphoryl group in the associative case, and –1 on metaphosphate in the dissociative case. Various experimental criteria used to distinguish the two transition states have been reviewed elsewhere [3,41].
Minireview Enzyme-catalyzed phosphoryl transfer Matte, Tari and Delbaere
Figure 4 Recent crystal structures of transition state analogs. (a) Stereo view of the active-site region of nucleoside diphosphate kinase complexed with ADP–Mg2+ and AlF3, showing the transition state geometry . The Al atom (magenta) is 2.5 Å from O7 of ADP and Nδ1 of His122, the site of phosphorylation. The Mg2+ ion (light blue sphere) is octahedrally coordinated by one F atom (green), oxygens of the α- and β-phosphates, and three water molecules (red spheres). A hydrogen bond is present (2.6 Å) between the 3′OH of the ribose and the bridging oxygen O7. Residues Lys16, Tyr56, Arg92 and Gly123 make interactions with the F atoms of AlF3. (b) Stereo view of the activesite region of UMP/CMP kinase with bound ADP–Mg2+, CMP and AlF3 . Atoms are coloured as in (a). The distances between the Al atom and the oxygen atoms of ADP and CMP are 2.0 Å and 2.2 Å, respectively. The coordination sphere of the Mg2+ ion (light blue) consists of one oxygen atom from the
terminal phosphate of ADP, one F atom, and four water molecules. The sidechains of
Lys19, Arg42, Arg93, Arg131 and Arg137 interact with substrates at the active site.
In order to more rigorously distinguish between associative and dissociative transition states in a structural way, it is necessary to have a view of the transition state complex. There are obvious experimental challenges to identifying the transition state for an enzyme-catalyzed phosphoryl transfer process, whose lifetime is of the order 10–13 s, using structural techniques. However, the use of structural analogs of the γ-phosphoryl group transferred in these reactions, such as tetrahedral BeF3– or trigonal planar AlF3 (described in ), in complex with other substrates, together permit snapshots of the ground state-like and transition state-like structures, respectively.
atom is 2.0 Å from the bridging O of ADP and 2.2 Å from CMP, suggesting a significant amount of bond formation in the transition state. This study may represent the best structural evidence so far in favor of an associative mechanism of phosphoryl transfer. In contrast, a similar study performed on nucleoside diphosphate kinase complexes with ADP–Mg2+ and either BeF3– or AlF3 suggested partial associative and dissociative character in the transition state . Finally, the crystal structure of human HRas bound to the GTPase-activating protein p120GAP (GAP-334) with ADP, Mg2+ and AlF3 has been recently solved . In this structure, the observed distances between the β-phosphoryl group of GDP and the catalytic water molecule (2.2 Å), the disposition of the Mg2+ ion bridging the β-phosphoryl group of ADP and one of the F atoms of AlF3, and the orientation of additional positively charged residues about the AlF3 group, which indicate charge neutralization of the transferred phosphoryl moiety, are together consistent with an associative transition state. Interestingly, binding of the GTPase activator GAP-334 introduces a single arginine sidechain into the active-site region, bridging the GDP and AlF3 molecules in the complex. This result supports the importance of charge neutralization during GTP hydrolysis and is also consistent with an associative-type mechanism.
Recently, crystal structures of three enzymes in complex with these analogs have been determined (Figure 4). Crystal structures of UMP/CMP kinase with substrates and either BeF2 or AlF3 indicate that charged arginine residues rearrange during catalysis to stabilize negative charges on the transferred phosphoryl group, in support of an associative mechanism . In this structure, the Al
In the course of the phosphoryl transfer reaction with a dissociative-type transition state, the largest change in charge is expected to be for the bridging oxygen atom which connects the β- and γ-phosphoryl groups . The observation of a hydrogen bond between the protein backbone and bridging O in p21Haras has been interpreted as evidence for a dissociative transition state through stabilization of
acceptor substrate in ground state enzyme–substrate or enzyme–inhibitor complexes. For the transition state to have some degree of associative character, these distances are expected to be less than the sum of a P–O van der Waals contact and a P–O single bond (i.e. in the order of approximately 4.9 Å); in the dissociative case these distances would be expected to be longer than this [3,47]. In cAMP-dependent protein kinase, an in-line, nucleophillic attack with a pentacoordinate transition state has been postulated [19,28]. In all cases, such analyses are beset by the limitation that they represent, that is, the structural arrangement prior to the actual phosphoryl transfer event.
Structure 1998, Vol 6 No 4
the GDP leaving group . Such hydrogen bonds are also observed in the BeF3– and AlF3 complexes of nucleoside diphosphate kinase, although in this case the hydrogen bond is donated by the substrate itself . Such an interaction is not inconsistent with an associative process; it is also possible that such a hydrogen bond serves to stabilize the orientation of the nucleotide during catalysis. Summary
There are a number of recurrent structural features of kinases complexed with substrates or substrate analogs which contribute to phosphoryl transfer. These features include global and local conformational changes, the use of Mg2+ and in some cases other divalent cations which interact with the transferred phosphoryl group, and the use of positively charged residues to stabilize the putative transition state, to name but a few. We are left to wonder whether the transition state for phosphoryl transfer from nucleotides under catalytic versus non-catalytic conditions are genuinely different, or whether some enzymes simply favor an associative over dissociative transition state. Further application of linear free-energy relationships (LFERs) to kinases, as described for alkaline phosphatase , should help to clarify this point. The continued development of novel experimental and theoretical approaches to a variety of kinases will be necessary to help answer these questions. Acknowledgements We thank Wim Hol, Jacqueline Cherfils and Ilme Schlichting for providing coordinates prior to release. LWT is the recipient of a Medical Research Council Post-doctoral Fellowship from the Medical Research Council of Canada. This work was supported by an operating grant from the Medical Research Council of Canada (LTJD).
References 1. Knowles, J.R. (1980). Enzyme-catalyzed phosphoryl transfer reactions. Annu. Rev. Biochem. 49, 877–919. 2. Knowles, J.R. (1982). Phosphotransfer enzymes: lessons from stereochemistry. Fed. Proc. 41, 2424–2431. 3. Mildvan, A.S. & Fry, D.C. (1987). NMR studies of the mechanism of enzyme action. Adv. Enzymol. 59, 241–313. 4. Frey, P.A. (1992). Nucleotidyltransferases and phosphotransferases: stereochemistry and covalent intermediates. In The Enzymes. (Sigman, D.S., ed.) 20, pp. 141–186, Academic Press, New York. 5. Cleland, W.W. & Hengge, A.C. (1995). Mechanisms of phosphoryl and acyl transfer. FASEB J. 9, 1585–1594. 6. Sinev, M.A., Sineva, E.V., Ittah, V. & Haas, E. (1996). Domain closure in adenylate kinase. Biochemistry 35, 6425–6437. 7. Tari, L.W., Matte, A., Pugazhenthi, U., Goldie, H. & Delbaere, L.T.J. (1996). Snapshot of an enzyme reaction intermediate in the structure of the ATP–Mg2+–oxalate ternary complex of phosphoenolpyruvate carboxykinase. Nat. Struct. Biol. 3, 355–363. 8. Bernstein, B.E., Michels, P.A.M. & Hol, W.G.J. (1997). Synergistic effects of substrate-induced conformational changes in phosphoglycerate kinase activation. Nature 385, 275–278. 9. Auerbach, G., et al., & Jacob, U. (1997). Closed structure of phosphoglycerate kinase from Thermotoga maritima reveals the catalytic mechanism and determinants of thermal stability. Structure 5, 1475–1483. 10. Larsen, T.M., Benning, M.M., Wesenberg, G.E., Rayment, I. & Reed, G.H. (1997). Ligand-induced domain movement in pyruvate kinase: structure of the enzyme from rabbit muscle with Mg2+, K+, and L-phospholactate at 2.7 Å resolution. Arch. Biochem. Biophys. 345, 199–206. 11. Gerstein, M., Lesk, A.M. & Chothia, C. (1994). Structural mechanisms for domain movements in proteins. Biochemistry 33, 6739–6749.
12. Cox, S., Radzio-Andzelm, E. & Taylor, S.S. (1994). Domain movements in protein kinases. Curr. Opin. Struct. Biol. 4, 893–901. 13. Vonrhein, C., Schlauderer, G.J. & Schulz, G.E. (1995). Movie of the structural changes during a catalytic cycle of nucleoside monophosphate kinases. Structure 3, 483–490. 14. Herzberg, O., et al., & Dunaway-Mariano, D. (1996). Swiveling-domain mechanism for enzymatic phosphotransfer between remote reaction sites. Proc. Natl. Acad. Sci. USA 93, 2652–2657. 15. Williams, R.L., Oren, D.A., Muñoz-Dorado, J., Inouye, S., Inouye, M. & Arnold, E. (1993). Crystal structure of Myxococcus xanthus nucleoside diphosphate kinase and its interaction with a nucleotide substrate at 2.0 Å resolution. J. Mol. Biol. 234, 1230–1247. 16. Moréra, S., et al., & Janin, J. (1994). Adenosine 5′-diphosphate binding and the active site of nucleoside diphosphate kinase. Biochemistry 33, 459–467. 17. Cherfils, J., Moréra, S., Lascu, I., Véron, M. & Janin, J. (1994). X-ray structure of nucleoside diphosphate kinase complexed with thymidine diphosphate and Mg2+ at 2-Å resolution. Biochemistry 33, 9062–9069. 18. Shirakihara, Y. & Evans, P.R. (1988). Crystal structure of the complex of phosphofructokinase from Escherichia coli with its reaction products. J. Mol. Biol. 204, 973–994. 19. Madhusudan, et al., & Sowadski, J.M. (1994). cAMP-dependent protein kinase: crystallographic insights into substrate recognition and phosphotransfer. Protein Sci. 3, 176–187. 20. Scheffzek, K., Kliche, W., Wiesmüller, L. & Reinstein, J. (1996). Crystal structure of the complex of UMP/CMP kinase from Dictyostelium discoideum and the bisubstrate inhibitor P1-(5′-adenosyl) P5-(5′-uridyl)pentaphosphate (UP5A) and Mg2+ at 2.2 Å: implications for water-mediated specificity. Biochemistry 35, 9716–9727. 21. Tari, L.W., Matte, A., Goldie, H. & Delbaere, L.T.J. (1997). Mg2+–Mn2+ clusters in enzyme-catalyzed phosphoryl-transfer reactions. Nat. Struct. Biol. 4, 990–994. 22. Hurley, J.H., et al., & Remington, S.J. (1993). Structure of the regulatory complex of Escherichia coli IIIGlc with glycerol kinase. Science 259, 673–677. 23. Baek, Y.H. & Nowak, T. (1982). Kinetic evidence for a dual cation role for muscle pyruvate kinase. Arch. Biochem. Biophys. 217, 491–497. 24. Lee, M.H., Hebda, C.A. & Nowak, T. (1981). The role of cations in avian liver phosphoenolpyruvate carboxykinase catalysis. J. Biol. Chem. 256, 12793–12801. 25. Buchbinder, J.L. & Reed, G.H. (1990). Electron paramagnetic resonance studies of the coordination schemes and site selectivities for divalent metal ions in complexes with pyruvate kinase. Biochemistry 29, 1799–1806. 26. Buchbinder, J.L., Baraniak, J., Frey, P.A. & Reed, G.H. (1993). Stereochemistry of metal ion coordination to the terminal thiophosphoryl group of adenosine 5′-O-(3-thiotriphosphate) at the active site of pyruvate kinase. Biochemistry 32, 14111–14116. 27. Bossemeyer, D., Engh, R.A., Kinzel, V., Ponstingl, H. & Huber, R. (1993). Phosphotransferase and substrate binding mechanism of the cAMP-dependent protein kinase catalytic subunit from porcine heart as deduced from the 2.0 Å structure of the complex with adenylyl imidodiphosphate and inhibitor peptide PKI(5–24). EMBO J. 12, 849–859. 28. Zheng, J., et al., & Sowadski, J.M. (1993). Crystal structure of the catalytic subunit of cAMP-dependent protein kinase complexed with MgATP and peptide inhibitor. Biochemistry 32, 2154–2161. 29. Duffy, T.H. & Nowak, T. (1985). 1H and 31P relaxation rate studies of the interaction of phosphoenolpyruvate and its analogues with avian liver phosphoenolpyruvate carboxykinase. Biochemistry 24, 1152–1160. 30. Hebda, C.A. & Nowak, T. (1982). Phosphoenolpyruvate carboxykinase. Mn2+ and Mn2+ substrate complexes. J. Biol. Chem. 257, 5515–5522. 31. Lee, M.H. & Nowak, T. (1984). Phosphorus-31 nuclear relaxation rate studies of the nucleotides on phosphoenolpyruvate carboxykinase. Biochemistry 23, 6506–6513. 32. Schulz, G.E. (1992). Binding of nucleotides by proteins. Curr. Opin. Struct. Biol. 2, 61–67. 33. Saraste, M., Sibbald, P.R. & Wittinghofer, A. (1990). The P-loop — a common motif in ATP- and GTP-binding proteins. Trends Biochem. Sci. 15, 430–434. 34. Schlichting, I. & Reinstein, J. (1997). Structures of active conformations of UMP kinase from Dictyostelium discoideum suggest phosphoryl transfer is associative. Biochemistry 36, 9290–9296. 35. Bossemeyer, D. (1994). The glycine-rich sequence of protein kinases: a multifunctional element. Trends Biochem. Sci. 19, 201–205.
Minireview Enzyme-catalyzed phosphoryl transfer Matte, Tari and Delbaere
36. Fritz-Wolf, K., Schnyder, T., Wallimann, T. & Kabsch, W. (1996). Structure of mitochondrial creatine kinase. Nature 381, 341–345. 37. Kobayashi, N. & Go, N. (1997). ATP binding proteins with different folds share a common ATP-binding structural motif. Nat. Struct. Biol. 4, 6–7. 38. Kobayashi, N. & Go, N. (1997). A method to search for similar protein local structures at ligand binding sites and its application to adenine recognition. Eur. Biophys. J. 26, 135–144. 39. Traut, T.W. (1994). The functions and consensus motifs of nine types of peptide sequences that form different types of nucleotide-binding sites. Eur. J. Biochem. 222, 9–19. 40. Herschlag, D. & Jencks, W.P. (1990). Catalysis of the hydrolysis of phosphorylated pyridines by Mg(OH)+: a possible model for enzymatic phosphoryl transfer. Biochemistry 29, 5172–5179. 41. Admiraal, S.J. & Herschlag, D. (1995). Mapping the transition state for ATP hydrolysis: implications for enzymatic catalysis. Chem. Biol. 2, 729–739. 42. Maegley, K.A., Admiraal, S.J. & Herschlag, D. (1996). Ras-catalysed hydrolysis of GTP: a new perspective from model studies. Proc. Natl. Acad. Sci. USA 93, 8160–8166. 43. Hassett, A., Blättler, W. & Knowles, J.R. (1982). Pyruvate kinase: is the mechanism of phosphotransfer associative or dissociative? Biochemistry 21, 6335–6340. 44. Peliska, J.A. & O’Leary, M.H. (1991). Sulfuryl transfer catalyzed by phosphokinases. Biochemistry 30, 1049–1057. 45. Jones, J.P., Weiss, P.M. & Cleland, W.W. (1991). Secondary 18O isotope effects for hexokinase-catalyzed phosphoryl transfer from ATP. Biochemistry 30, 3634–3639. 46. Moréra, S., Chiadmi, M., LeBras, G., Lascu, I. & Janin, J. (1995). Mechanism of phosphate transfer by nucleoside diphosphate kinase: X-ray structures of the phosphohistidine intermediate of the enzymes from Drosophila and Dictyostelium. Biochemistry 34, 11062–11070. 47. Müller-Dieckmann, H.-J. & Schulz, G.E. (1994). The structure of uridylate kinase with its substrates, showing the transition state geometry. J. Mol. Biol. 236, 361–367. 48. Abele, U. & Schulz, G.E. (1995). High-resolution structures of adenylate kinase from yeast ligated with inhibitor Ap5A, showing the pathway of phosphoryl transfer. Protein Sci. 4, 1262–1271. 49. Antonny, B. & Chabre, M. (1992). Characterization of the aluminum and beryllium fluoride species which activate transducin. Analysis of the binding and dissociation kinetics. J. Biol. Chem. 267, 6710–6718. 50. Xu, Y.-W., Moréra, S., Janin, J. & Cherfils, J. (1997). AlF3 mimics the transition state of protein phosphorylation in the crystal structure of nucleoside diphosphate kinase and MgADP. Proc. Natl. Acad. Sci. USA 94, 3579–3583. 51. Scheffzek, K., et al., & Wittinghofer, A. (1997). The Ras–RasGAP complex: structural basis for GTPase activation and its loss in oncogenic ras mutants. Science 277, 333–338. 52. Hollfelder, F. & Herschlag, D. (1995). The nature of the transition state for enzyme-catalyzed phosphoryl transfer. Hydrolysis of O-aryl phosphothioates by alkaline phosphatase. Biochemistry 34, 12255–12264.