Assignment of conserved amino acid residues to the ATP site in the protein kinase domain of the receptor for epidermal growth factor

The ATP substrate site in the epidermal growth factor (EGF) receptor was mapped by using a series of 26 ATP derivatives with modifications at the base, ribose or triphosphate moiety. Ki values for these derivatives were determined by competition with [gamma-32P]ATP. The enzyme seems to interact specifically with the beta-phosphate in an ion-pair bond with the N-6 amino group at the adenine in a hydrogen bond. With ribosyl-2-aminopurine triphosphate and GTP, the enzyme most likely recognizes the 2-amino group in a hydrogen bond. This high specificity for ATP and GTP is unique for the ATP site in the EGF receptor among all investigated protein kinases. The available data on the interaction between ATP derivatives and protein kinases were used to assign conserved amino acid residues found in diverse protein kinases to the ATP site in this type of enzyme.     

Interaction of N1-, N6- and C8-substituted derivatives of adenosine-5'-triphosphate with the catalytic subunit of cAMP-dependent protein kinase from rabbit skeletal muscles

In order to investigate the structure of the active site of the cAMP-dependent protein kinase catalytic subunit a synthesis of several previously unknown adenosine-5'-triphosphate (ATP) derivatives containing substituents of various nature at N(1), N(C6) and C(8) positions of the purine base was carried out. The interaction of these derivatives with a homogeneous preparation of the catalytic subunit of rabbit skeletal muscle cAMP-dependent protein kinase was investigated. All the nucleotide analogs were found to inhibit the enzyme activity; the inhibition was competitive with respect to ATP. It was assumed that the adenine moiety of the ATP molecule is bound to the active site of protein kinase by the hydrophobic interaction with the aromatic amino acid residues and by formation of the hydrogen bond between the exo-NH2-group of the substrate and a corresponding group of the enzyme. The "correct" binding of ATP to the enzyme active center is defined by the anti-conformation of the nucleotide.     

ATP analog specificity of cAMP-dependent protein kinase, cGMP-dependent protein kinase, and phosphorylase kinase

The ATP analog specificities of the homogeneous cGMP-dependent protein kinase and the catalytic subunit of cAMP-dependent protein kinase have been compared by the ability of 27 analogs to compete with ATP in the protein kinase reaction. Although the data suggest general similarities between the ATP sites of the two homologous cyclic-nucleotide-dependent protein kinases, specific differences especially in the adenine binding pocket are indicated. These differences in affinity suggest potentially useful ATP analog inhibitors of each kinase. For example, apparent autophosphorylation of the purified regulatory subunit of the cAMP-dependent protein kinase is blocked by nebularin triphosphate, suggesting that the phosphorylation is catalyzed by trace contamination of cGMP-dependent protein kinase. Some of the ATP analogs have also been tested using phosphorylase b kinase in order to compare this enzyme with the cyclic-nucleotide-dependent enzymes. All three protein kinases have high specificity for the purine moiety of ATP, and lower specificity for the ribose or triphosphate. The similarity between the ATP site of phosphorylase b kinase to that of the cyclic-nucleotide-dependent protein kinases suggests that it is related to them. The ATP analog specificities of enzymes examined in this study are different from those reported for several unrelated ATP-utilizing enzymes.     

Mapping the ATP-binding site in the catalytic subunit of adenosine-3':5'-monophosphate-dependent protein kinase. Spatial relationship with the ATP site of the undissociated enzyme.

A set of 24 ATP analogs modified at various positions of the ATP molecule was used for mapping the ATP-binding site in the free catalytic subunit (C) of cAMP-dependent protein kinase (type I). Ki values for these analogs (of which 23 were shown to be competitive with ATP) were measured and compared with Ki values previously obtained for the same set of analogs upon binding to the undissociated form of the enzyme (R2C2). It was found that modifications at the adenine part of ATP bring about a considerable reduction in affinity between C and the resulting analog. The other parts of the ATP molecule play a less important, though definite, role in the binding of this nucleotide to C. By measuring the effect of each given modification in ATP on its binding to C, and comparing the effect of this modification on the binding of the same analog to R2C2, it was possible to obtain 'specificity profiles' for both forms of the kinase. Using such profiles it is shown that the adenine-binding subsite in C may well coincide with the adenine-binding subsite in R2C2. Two plausible models describing the spatial relationship between the ATP sites in C and R2C2 are proposed.     

Deconstructing nucleotide binding activity of the Mdm2 RING domain

Mdm2, a central negative regulator of the p53 tumor suppressor, possesses a Really Interesting New Gene (RING) domain within its C-terminus. In addition to E3 ubiquitin ligase activity, the Mdm2 RING preferentially binds adenine base nucleotides, and such binding leads to a conformational change in the Mdm2 C-terminus. Here, we present further biochemical analysis of the nucleotide–Mdm2 interaction. We have found that MdmX, an Mdm2 family member with high sequence homology, binds adenine nucleotides with similar affinity and specificity as Mdm2, suggesting that residues involved in nucleotide binding may be conserved between the two proteins and adenosine triphosphate (ATP) binding may have similar functional consequences for both Mdm family members. By generating and testing a series of proteins with deletions and substitution mutations within the Mdm2 RING, we mapped the specific adenine nucleotide binding region of Mdm2 to residues 429–484, encompassing the minimal RING domain. Using a series of ATP derivatives, we demonstrate that phosphate coordination by the Mdm2 P-loop contributes to, but is not primarily responsible for, ATP binding. Additionally, we have identified the 2′ and 3′ hydroxyls of the ribose and the C6 amino group of the adenine base moiety as being essential for binding.     

Interaction of various nucleotide-dependent enzymes with bifunctional analogs of ATP, derivatives of polymethylene diamines

The interaction of bifunctional ATP derivatives, Appp5'[NH-(CH2) n-NH]ppp5'A (n = 0 or 2-8) with tyrosyl-, valyl-, lysyl-, tryptophanyl-tRNA synthetases and creatine kinase was investigated. ATP derivatives don't inhibit the tRNA aminoacylation catalyzed by tyrosyl-tRNA synthetase. These derivatives behave as mixed-type inhibitors with respect to ATP in the case of valyl- and lysyl-tRNA-synthetases. In the case of the other enzymes all analogs of ATP manifest competitive inhibition towards ATP. The affinity of all ATP derivatives to tryptophanyl-tRNA synthetase does not differ significantly (Ki = 0.2 divided by 0.6 mM). The Ki values for these derivatives in the case of creatine kinase are also very similar with the exception of A5'ppp-NH-(CH2)3-NH-ppp5'A. The Ki value for this derivative is one order of magnitude lower than for other ones. The affinity reagents received by periodate oxidation of bifunctional ATP analogs derivatives of di-, tetra- and heptamethylenediamine modify non-identical subunits of creatine kinase with different velocities, but modification of M- and M'-subunits proceeds independently. An analogues derivative of trimethylenediamine interacts simultaneously with two centers of the dimeric form of kinase forming non-equivalent complexes. The covalent attachment of the reagent to one subunit of creatine kinase does not except the complex formation and covalent binding of bifunctional ATP analogs with the other subunit of the dimer, but results in a one order of magnitude decrease in affinity of the ATP derivative to the nonmodified centre of the enzyme. These data permit to evaluate the distance between ATP binding sites of creatine kinase in its dimeric form as 5-6 A approximately. Such a distance between active sites may be the reason for the higher activity of the M- and M'-creatine kinase subunits taken separately as compared to the enzyme dimeric form.     

Derivatives of 17-(2-methylallyl)-substituted noroxymorphone: variation of the delta address and its effects on affinity and selectivity for the delta opioid receptor

In an effort to establish the importance of the N-(2-methylallyl) substituent in the noroxymorphone series, several derivatives have been synthesized, retaining that N-substituent and modifying the delta address moiety. A few compounds showed moderate binding affinity and selectivity for the delta receptor; none displayed a pharmacological profile as exceptional as N-(2-methylallyl)noroxymorphindole. A second study showed that 3-O-methylation of all derivatives decreases binding affinity. The present results indicate that only a combination of the N-(2-methylallyl) group and an indole delta address provided high selectivity for the delta receptor.     

Mapping of the adenosine 5'-triphosphate binding site of type II calmodulin-dependent protein kinase

The specificity of the ATP-binding site of the type II calmodulin-dependent protein kinase was probed with 25 analogues of ATP modified at various positions of the molecule. The analogues were compared by their ability to compete with ATP in the protein kinase reaction. The result of this comparison indicates that the enzyme is most sensitive to modifications at, or replacement of, the purine moiety. Changes at the triphosphate chain are much better tolerated, although the enzyme exhibited a selective sensitivity to changes in the conformation of this group. The smallest contribution to the specificity of ATP binding appears to be made by the ribose ring. The Ki values obtained for a subset of these analogues were compared to those previously reported for phosphorylase b kinase and the cyclic nucleotide dependent protein kinases [Flockhart, D. A., Freist, W., Hoppe, J., Lincoln, T. M., & Corbin, J. D. (1984) Eur. J. Biochem. 140, 289-295]. A striking similarity in the responses of these protein kinases to modifications of the ATP molecule suggests that the type II calmodulin-dependent protein kinase is related to these enzymes. Support for this conclusion was provided, recently, through comparisons of the deduced primary structures of the alpha and beta subunits of the type II calmodulin-dependent protein kinase with the protein sequences of the catalytic subunits of phosphorylase b kinase and cAMP-dependent protein kinase [Hanley, R. M., Means, A. R., Ono, T., Kemp, B. E., Burgin, K. E., Waxham, N., & Kelly, P. T. (1987) Science (Washington, D.C.) 237, 293-297; Bennett, M. K., & Kennedy, M. B. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 1794-1798], which indicated areas of extensive homology.     

Mechanism of activation of protein kinase I from rabbit skeletal muscle. Mapping of the cAMP site by spin-labeled cyclic nucleotides.

Binding of adenosine 3':5'-monophosphate (cAMP) to protein kinase (type I) from rabbit skeletal muscle has been investigated using spin-labeled cAMP derivatives. Different compounds were synthesized with the spin label attached by spacer chains of different length at different positions on the adenine base. Immobilization of the spin label, determined by comparing the electron-spin resonance spectra recorded in the presence of the kinase with those of the free ligand in solutions of different viscosities, gave information about the geometry of the cAMP site. Strong immobilization of the N-6 substituents up to a spacer length of seven atoms indicates a rather deep cleft of the cAMP site. The depth of this cleft differs, however, when the spin label is attached to the different positions at the adenine (N-6, C-2 and C-8). Whereas the N-6 derivatives indicate a rather deep site, the C-2 derivatives reveal a significantly smaller depth and C-8 substituents (syn conformation) obviously occupy a very shallow surface with almost no immobilation. In addition the binding affinities of the spin-labeled cAMP derivatives have been determined, together with those of a series of (diamagnetic) C-2 derivatives bearing hydrophobic alkyl chains of different length. The latter results helped to clarify the differences between the regions near to C-2 and N-6, respectively, of the cAMP site. N-6 spin-labeled derivatives have also been investigated in the presence of ATP and protein kinase. These results are interpreted as indicative of a conformational change at the cAMP site upon formation of the holoenzyme, due to binding of ATP, leaving cAMP less strongly immobilized.     

Inhibitory action of certain cyclophosphate derivatives of cAMP on cAMP-dependent protein kinases

A series cAMP derivatives with modifications in the adenine, ribose and cyclophosphate moiety were screened for their binding affinity for the two types of cAMP-binding sites in mammalian protein kinase type 1. In addition, the activation of the kinase by these analogs was monitored. The binding data indicate that cAMP is bound to both sites in a comparable manner: the adenine appears to have no hydrogen-bond interactions with the binding sites, whereas the ribose may be bound by three hydrogen bonds involving the 2', 3' and 5' positions of cAMP. The binding data are not conclusive about the nature of the interaction with the exocyclic oxygen atoms on phosphorus, though a charge interaction seems to be absent. The cAMP molecule seems to be bound in the syn conformation. The results of activation experiments show that modifications in the adenine and ribose moiety do not affect the maximal activation level, while alteration of the two exocyclic oxygen atoms may result in a reduced maximal activation level and in one case, (Rp)-adenosine 3', 5'-monophosphorothioate [Rp-cAMPS], in total absence of activation even at concentrations at which the analog saturates both binding sites. Since occupancy of the cAMP-binding sites by this derivative apparently did not lead to activation of the enzyme, we examined whether this compound could antagonize the activation by cAMP. Indeed (Rp)-cAMPS was found to inhibit cAMP stimulated kinase activity at concentrations compatible to its binding affinity. Also with mammalian protein kinase type II (Rp)-cAMPS showed antagonistic activity, while with a cAMP-dependent protein kinase from Dictyostelium discoideum partial agonistic activity was observed. Previously a mechanism for activation of protein kinase type I was proposed involving a charge interaction between the equatorial exocyclic oxygen atom and the binding site [De Wit et. al. (1982) Eur. J. Biochem 122, 95-99]. This was based on measurements with impure preparations of (Rp)-cAMPS and the Rp and Sp isomers adenosine 3', 5'-monophosphodimethylamidate. cAMPN(CH3)2. The present work using highly purified compounds suggests the absence of a charge interaction, since the uncharged analog (Sp)-cAMPN(CH3)2 activates the kinase effectively. The data seem compatible with an activation model involving the formation of a covalent bond with phosphorus in both cAMP binding sites.     

Mg X ATP2-dependent interaction of the inhibitor protein of the cAMP-dependent protein kinase with the catalytic subunit

The interaction between the inhibitor protein and the catalytic subunit of the cAMP-dependent protein kinase has been investigated by steady state kinetics and by an assessment of the requirement of this interaction for ATP. By analysis for tightly bound inhibitors, inhibition by the inhibitor protein was shown to be competitive versus peptide substrate and uncompetitive versus Mg X ATP2-. This, together with the observations of Gronot et al. (Gronot, J., Mildvan, A.S., Bramson, H. N., Thomas, N., and Kaiser, E.T. (1981) Biochemistry 20, 602-610) and those given in the accompanying paper (Whitehouse, S., Feramisco, J.R., Casnellie, J.E., Krebs, E.G., and Walsh, D.A. (1983) J. Biol. Chem. 258, 3693-3701), would indicate that the probable reaction mechanism of the protein kinase is ordered with the nucleotide binding first and that the inhibitor protein blocks catalysis by interaction with the catalytic subunit-Mg X ATP complex. The Ki for this interaction at saturating Mg X ATP and zero peptide substrate is 0.49 nM. Multiple inhibition analysis in the presence of 5'-adenylimidodiphosphate (AMP X PNP) indicates that the inhibitor protein does not interact with a catalytic subunit-AMP X PNP complex. The requirement for ATP for the inhibitor protein-catalytic subunit interaction has also been demonstrated by direct binding measurements and by the observation that the efficiency of the inhibitor protein is increased by preincubation of the inhibitor protein, catalytic subunit, and ATP in the absence of peptide substrate. By either measurement, the catalytic subunit in the presence of the inhibitor protein, was shown to exhibit an apparent Kd of 20 approximately 60 nM for ATP; this value is two orders of magnitude higher than the affinity for ATP by the catalytic subunit alone. This high apparent affinity of the catalytic subunit for ATP (in the presence of the inhibitor) does not require that there be a specific binding site on the inhibitor protein for some moiety of the ATP but may simply be a reflection of the formation of a catalytic subunit-Mg X ATP X inhibitor protein complex with resultant displacement of the equilibrium of ATP binding to the protein kinase.     

K-252 compounds, novel and potent inhibitors of protein kinase C and cyclic nucleotide-dependent protein kinases

K-252 compounds (K-252a and b isolated from Nocardiopsis sp. (1) and their synthetic derivatives) were found to inhibit cyclic nucleotide-dependent protein kinases and protein kinase C to various extents. The inhibitions were of the competitive type with respect to ATP. K-252a was a non-selective inhibitor for these three protein kinases with Ki values 18-25 nM. K-252b showed a comparable potency for protein kinase C (Ki, 20nM), whereas inhibitory potencies for cyclic nucleotide-dependent protein kinases were reduced. KT5720 and KT5822 selectively inhibited cAMP-dependent (Ki, 60nM) and cGMP-dependent (Ki, 2.4nM) protein kinases, respectively.     

Some thoughts on the evolutionary basis for the prominent role of ATP and ADP in cellular energy metabolism

The predominance of the adenosine triphosphate/adenosine diphosphate (ATP/ADP) couple in cellular phosphorylation reactions, including those that form the basis for cellular energy metabolism, cannot be explained on thermodynamic grounds since a variety of "high energy phosphate" compounds (including ADP itself) found in the cell would, based on thermodynamic considerations, be at least as effective as ATP in serving as a phosphoryl donor. How then did present-day organisms come to rely on the ATP/ADP couple as the principal mediator of phosphorylation reactions? The early appearance of adenine compounds in the prebiotic environment is suggested by experiments indicating that, relative to other purine or pyridimine compounds, adenine derivatives are preferentially synthesized under simulated prebiotic conditions (Ponnamperuma et al., 1963). In addition to the roles of adenine nucleotides in phosphorylation reactions, other adenine derivatives (e.g. Coenzyme A, flavin adenine dinucleotide, puridine nucleotides) are employed in a variety of metabolic roles. The principal function of the adenine moiety in these latter cases is in the binding of these derivatives to the relevant enzyme. The capability for binding of the adenine moiety appears to have arisen early in evolution and been exploited in a multitude of contexts, a suggestion consistent with observed similarities between the binding sites of several enzymes employing adenine derivatives as substrate. The early availability of suitable adenine compounds in the biosphere and development of complementary binding sites on cellular proteins, coupled with the expected advantages in having a limited number of metabolites as central mediators of endergonic and exergonic metabolism could readily have led to the observed pre-eminence of adenine nucleotides in cellular energy metabolism.(ABSTRACT TRUNCATED AT 250 WORDS)     

N-(2-Aminoethyl)-5-isoquinolinesulfonamide, a newly synthesized protein kinase inhibitor, functions as a ligand in affinity chromatography. Purification of Ca2+-activated, phospholipid-dependent and other protein kinases

We designed a simple procedure for the purification of Ca2+-activated, phospholipid-dependent protein kinase (protein kinase C) from rabbit brain, using affinity chromatography with a new affinity adsorbent. The adsorbent was synthesized by attaching the amino residue of N-(2-aminoethyl)-5-isoquinolinesulfonamide (H-9) to cyanogen bromide-activated Sepharose. H-9 is a potent competitive inhibitor of protein kinase C, cGMP-, and cAMP-dependent protein kinase with respect to ATP and exhibits inhibition constants of 18, 0.87, and 1.9 microM, respectively (Hidaka, H., Inagaki, M., Kawamoto, S., and Sasaki, Y. (1984) Biochemistry, 23, 5036). A 960-fold purification was achieved in the two-step procedure, which entailed DEAE-cellulose and the affinity chromatography. The resultant preparation was essentially homogeneous, as indicated by polyacrylamide gel electrophoresis under conditions of denaturation with sodium dodecyl sulfate. The affinity of protein kinase C for the H-9-Sepharose was high, and the enzyme could not be eluted either by a high concentration of sodium chloride or by 40% glycerol. The protein kinase C could be eluted from H-9-Sepharose by the buffer containing both 0.2 M NaCl and 20% glycerol, thereby suggesting that the binding between protein kinase C and H-9-Sepharose was due to both hydrophobic and electrostatic interactions. H-9 coupled to Sepharose retained both cyclic nucleotide-dependent protein kinases and protein kinase C, and these enzymes could be eluted separately by the buffer containing L-arginine, a potent inhibitor of these three kinases. The novel aspects of these three multifunctional protein kinases can thus be investigated using isoquinolinesulfonamide derivatives.     

Inhibition of tryptophanyl-tRNA synthetase by modifying ATP analogs

The interaction between tryptophanyl-tRNA synthetase (EC 6.1.1.2) from beef pancreas and the ATP analogs containing alkylating or phosphorylating groups in the polyphosphate moiety of ATP was studied as an approach to investigate the structure of the enzyme active center. Some of the compounds under study were shown to irreversibly inhibit the enzyme activity; the presence of ATP in the most cases protects the enzyme against inactivation. The kinetic constants Ki and k2 of interaction of the irreversible inhibitors with the enzyme were determined. It was found that the Ki values for a number of irreversible competitive inhibitors are by 1-2 orders of magnitude less than the Km value for ATP; the k2 values were found equal to 0.02-0.04 min-1. this suggests that the compounds may be used as affinity reagents, the most efficient ones being adenosine 5'-(beta-chloroethyl phosphate) and mixed AMP-mesithylene carbonic acid anhydride. The absence of a protective effect of ATP in the case of adenosine 5'-(beta-bromoethane phosphonate) and non-competitive type of reversible inhibition inhibition of the enzyme by adenosine 5'-chloromethane phosphonate indicate that the molecule of tryptophanyl-tRNA synthetase contains sites interacting with adenine nucleotides, other than the ATP binding sites of the active center.     

Multisubstrate analogs for deoxynucleoside kinases. Triphosphate end products and synthetic bisubstrate analogs exhibit identical modes of binding and are useful probes for distinguishing kinetic mechanisms

Comparative inhibition kinetics with natural dNTP end products (dNp3) and new synthetic bisubstrate-type analogs, dNp4A (deoxynucleoside 5'-adenosine 5'-P1,P4-tetraphosphate), have been studied with their target deoxynucleoside kinases from Lactobacillus acidophilus. Analysis of inhibition specificity, inhibition patterns, and Ki(app) under various conditions has revealed the following conclusions. Both dNTP and dNp4A bind to the active site of the corresponding kinase through multiple binding determinants. The deoxynucleoside moiety of dNTP fits optimally at the deoxynucleoside binding site and provides the basis for its inhibition specificity, whereas the triphosphate group interacts with the ATP binding site, reinforcing the affinity of the molecule as a potent end product inhibitor (Ki = 0.4-3 microM). The adenosine moiety of dNp4A does not contribute to the binding of this compound, whereas the tetraphosphate portion is the second binding determinant, just as in the model developed for dNTP. dNTP and dNp4A proved to be useful tools for distinguishing the kinetic mechanisms of kinases which follow sequential pathways, i.e. the rapid equilibrium Random Bi Bi for dCyd and dGuo kinases and the steady state Ordered Bi Bi mechanism for two dAdo kinases associated either with dCyd kinase or with dGuo kinase on different multifunctional proteins.     

Interaction of ATP with acetyl-CoA carboxylase from rat liver. The role of the polyphosphate chain. Affinity labelling with alkylating amides of ATP and ADP

The interaction of a number of ATP analogs with a modified triphosphate moiety as well as 2-chloro-ethyl-amino derivatives of nucleotides, 4(N-2-chloroethyl-N-methylamino)-benzyl-gamma-amide of ATP and the corresponding ADP beta-amide with acetyl-CoA carboxylase (EC 6.4.1.2.) from rat liver has been studied. Halophosphonate derivatives of ATP have been synthesized from bromomethylene diphosphonic acid and found to be reversible inhibitors of the enzyme. ATP and ADP alkylating amides have proved to form a reversible complex with the ATP-binding site and to modify a group in the acetyl-CoA-binding site. The bicarbonate ion accelerates the process of inactivation. The estimate of the distance between the ATP-binding site and the acetyl-CoA-binding site ranges within 0.8-1.2 nm.     

Effect of some new cAMP analogs on cAMP-dependent protein kinase isoenzymes.

1. Ten new cAMP analogs were synthesized by replacing the purine ring with with indazole, benzimidazole or benztriazole and/or their nitro and amino derivatives. 2. Each analog proved effective in activating cAMP-dependent protein kinase I (PK-I) purified from rabbit skeletal muscle and cAMP-dependent protein kinase II (PK-II) from bovine heart and chasing 8-[3H]cAMP bound to regulatory subunits in the half-maximal effective concentrations of 2 x 10(-8)-8 x 10(-6) M. 3. The N-1-beta-D-ribofuranosyl-indazole-3'5'-cyclophosphate(I) proved a very poor chaser and activator of both isoenzymes, but when indazole was attached at its N-2 to ribose (IV) or when its H at C-4 (equivalent to the position of amino-group in adenine) was substituted by an amino-(III) or especially nitro-group (II) its efficiency was dramatically increased. 4. Analogs containing benztriazole ring proved as powerful as cAMP irrespective of the presence of substituents (VII-X). 5. Benzimidazole derivatives with amino-(VI) or nitro-group (V) activated PK-II 3 and 20 times better than PK-I. 6. Attaching of ribose to N-2 of indazole or benztriazole increased the affinity to PK-II 10 and 4 times, respectively. 7. Chasing efficiency of cAMP analogs at half-saturating [3H]cAMP tended to correlate with activating potency only for PK-I but at saturating [3H]cAMP concentration for both isoenzymes. 8. On the basis of synergistic activation with 8-Br-cAMP a site 2-selective binding of nitro-benzimidazole (V) and unsubstituted benztriazole (VII) derivatives to PK-II is suggested.     

Ribofuranosyl-benzimidazole derivatives as inhibitors of casein kinase-2 and casein kinase-1

5,6-Dichloro-1-(beta-D-ribofuranosyl)benzimidazole (DiCl-RB) is a powerful inhibitor of casein kinase-2 (CK-2) [Zandomeni, R. et al. (1986) J. Biol. Chem. 261, 3414-3420]. Here a series of 17 analogues of DiCl-RB has been employed for studying the specificity and the mode of action of this family of CK-2 inhibitors. The two halogen substituents on the benzene ring are shown to play a prominent role in inhibition, the 5,6-dibromo derivative (DiBr-RB) being fivefold more effective than DiCl-RB (Ki = 2 microM, with GTP as substrate), whereas the difluoro derivative (DiF-RB) is nearly as ineffective as unsubstituted 1-(beta-D-ribofuranosyl)benzimidazole. On the other hand, although some modifications of the ribose group significantly decrease the inhibitory efficiency, the sugar moiety is not strictly required, since dichlorobenzimidazole itself (DiCl-Bz) is an inhibitor almost as good as DiCl-RB. Inhibition of CK-2 by DiCl-RB and by its analogues, DiCl-Bz included, is of the competitive type with respect to the nucleotide substrate, the Ki values being lower with GTP than with ATP. The Ki values of the most potent inhibitor, DiBr-RB, with ATP and GTP, are 6 microM and 2 microM, respectively, denoting an affinity for the enzyme higher than that of the physiological substrates, ATP and GTP. DiBr-RB has been assayed for its inhibitory capacity toward several protein kinase other than CK-2. Protein kinase-C, cAMP-dependent protein kinase, the Ser/Thr protein kinase expressed by Pseudorabies virus, and four different tyrosine protein kinases from spleen, proved insensitive to DiBr-RB concentrations capable of almost entirely suppressing the activity of rat liver and maize seedling CK-2. Casein kinase-1 however is nearly as sensitive as CK-2 to DiBr-RB. Inhibition of CK-1 is also of the competitive type with respect to ATP (Ki = 14 microM). Although the inhibitory spectrum of CK-1 by the various analogues is reminiscent of that observed with CK-2, a remarkable difference is revealed by 5'-phosphorylation of ribose which increases the Ki with CK-2 while decreasing that with CK-1.     

Molecular determinants for ATP-binding in proteins: a data mining and quantum chemical analysis

Adenosine 5'-triphosphate (ATP) plays an essential role in all forms of life. Molecular recognition of ATP in proteins is a subject of great importance for understanding enzymatic mechanism and for drug design. We have carried out a large-scale data mining of the Protein Data Bank (PDB) to analyze molecular determinants for recognition of the adenine moiety of ATP by proteins. Non-bonded intermolecular interactions (hydrogen bonding, pi-pi stacking interactions, and cation-pi interactions) between adenine base and surrounding residues in its binding pockets are systematically analyzed for 68 non-redundant, high-resolution crystal structures of adenylate-binding proteins. In addition to confirming the importance of the widely known hydrogen bonding, we found out that cation-pi interactions between adenine base and positively charged residues (Lys and Arg) and pi-pi stacking interactions between adenine base and surrounding aromatic residues (Phe, Tyr, Trp) are also crucial for adenine binding in proteins. On average, there exist 2.7 hydrogen bonding interactions, 1.0 pi-pi stacking interactions, and 0.8 cation-pi interactions in each adenylate-binding protein complex. Furthermore, a high-level quantum chemical analysis was performed to analyze contributions of each of the three forms of intermolecular interactions (i.e. hydrogen bonding, pi-pi stacking interactions, and cation-pi interactions) to the overall binding force of the adenine moiety of ATP in proteins. Intermolecular interaction energies for representative configurations of intermolecular complexes were analyzed using the supermolecular approach at the MP2/6-311 + G* level, which resulted in substantial interaction strengths for all the three forms of intermolecular interactions. This work represents a timely undertaking at a historical moment when a large number of X-ray crystallographic structures of proteins with bound ATP ligands have become available, and when high-level quantum chemical analysis of intermolecular interactions of large biomolecular systems becomes computationally feasible. The establishment of the molecular basis for recognition of the adenine moiety of ATP in proteins will directly impact molecular design of ATP-binding site targeted enzyme inhibitors such as kinase inhibitors.