The role of the C8 proton of ATP in the regulation of phosphoryl transfer within kinases and synthetases

The International Union of Pure and Applied Chemistry and the International Union of Biochemistry (IUPAC/IUB) commission on the classification and nomenclature of enzymes placed the enzymes that transfer high energy phosphate bonds from nucleotides into two divisions: the transferases (kinases) and ligases (synthetases) [1].

The transferases have been placed in Division 2 and the ligases into Division 6 of the Enzyme Commission (EC) classification. The ligases are enzymes catalysing the joining of two molecules with the concomitant hydrolysis of the pyrophosphate bond of ATP, while a kinase is defined as an enzyme which catalyses the transfer of the phosphate group from ATP (or GTP) to a substrate containing an alcohol, amino, carboxyl, or phosphate group as the phosphoryl acceptor [2, 3]. The kinases are a large number of structurally diverse enzymes that play a critical role in numerous metabolic and signalling pathways and whose substrates may be a small molecule, lipid, or protein. They have been classified into 25 families of homologous proteins, with the families assembled into 12 fold-groups based on the similarity of their structural folds [2]. However, this classification relays little, if any, information on the catalytic and regulatory mechanisms employed in nucleotide binding and phosphoryl transfer. Within a single group, both prokaryotic and eukaryotic organisms are represented with kinase isoenzymes that appear to be kinetically and functionally distinct based on the rate of phosphoryl transfer and the regulation thereof.

This investigation was undertaken to ascertain the extent to which the adenyl group within ATP plays a direct role in the regulation of ATP binding and/or phosphoryl transfer within a range of kinase and synthetase enzymes. To this end the role of the C8-H of ATP on the binding and/or phosphoryl transfer on the enzyme activity of a number of kinase and synthetase enzymes was elucidated in comparative enzyme activity essays using ATP and ATP deuterated at the C8 position. Kinetic isotope effects (KIE) are broadly classified into primary, secondary and stearic effects. The extent of proton/deuterium KIE's is estimated from the rate constants (KIE = v_H/v_D) and a KIE of the order of 2 or more is strong evidence that the bond to the isotopically substituted hydrogen atom is being broken in the rate determining step of the reaction for the primary KIE and are as a result of bond breaking [4, 5]. The calculated maximum for the KIE involving C-H bonds is approximately 7 at room temperature as determined by the difference in the zero-point energy-difference between the bond to the deuterium and the bond to the hydrogen . The secondary deuterium KIE's are defined as the isotope effect when the bond to the isotopically substituted atom is not cleaved but occur as a result of a hybridization change. Secondary KIE's are classified as α or β, depending on whether the isotopic substitution is made on the α or β centre relative to the atom undergoing the chemical change or further away within the molecule. If the hybridization change is from sp³ to sp², the KIE is normal however if the hybridization change is sp² to sp³ the KIE is inverse, with the range being 1.4 to 0.7, respectively. Stearic effects also affect the KIE to the same extent as the secondary KIE's.

In oligomeric enzymes it is proposed that the deuteration of ATP not only affects the binding of ATP to the site where catalysis is occurring but the deuteration also affects the interaction between sites. In oligomeric kinases it is proposed that mechanistically two modes of regulation occur, one which is dependent on the release of ADP from the first active site before ATP binds to the second active site and the second mode of regulation depends on the conversion of ATP to ADP prior to the binding of the ATP to the second active site. To this end a range of kinase and synthetase enzymes were tested to assess the role of the C8-H of ATP in the binding and steady state enzyme activity of these enzymes.

Acetate kinase (EC 2.7.2.1) is a homodimer which catalyses the Mg²⁺-dependent, reversible transfer of phosphate from ATP to acetate in the following reaction:

Acetate kinase forms part of the acetate and sugar kinase/Hsc70/actin (ASKHA) structural superfamily (PFam Clan: Actin_ATPase:CL0108) [6]. The enzyme is a homodimer and monomer interaction plays a role in the regulation of the enzyme activity and ligand binding with the enzyme active sites functioning in a coordinated half-the-sites manner [7–9]. The actin ATPase clan contains both the acetate kinases and sugar kinases and are all known to undergo a catalytically essential domain closure upon ligand binding.

Hexokinase (ATP: D-hexose 6-phosphotransferase, EC 2.7.1.1) catalyses the Mg²⁺-dependent phosphorylation of glucose, from ATP:

The two isoenzymes of yeast hexokinase, designated P-I and P-II, are dimers of subunit molecular mass 52 kDa [10–15]. Hexokinase also forms part of the acetate and sugar structural superfamily (PFam Clan: Actin_ATPase:CL0108). Yeast hexokinase enzymes are structurally well characterised with each subunit of the homodimer comprising two domains and in the open conformation these domains are separated by a cleft containing the sugar binding site [16–22]. Binding of glucose induces a large conformational change in which the two lobes of the subunit rotate relative to each other. The enzymes also exist in a monomer-dimer association-dissociation equilibrium that is influenced by pH, ionic strength and substrates. There are major differences in the glucose binding behaviour of both forms where binding to dimeric P-I shows strong positive cooperativity, whereas in P-II the two sites are equivalent and binding is non-cooperative [23–26].

The shikimate pathway is a seven-step biosynthetic route that links the metabolism of carbohydrates to the synthesis of aromatic amino acids via the conversion of erythrose-4-phosphate to chorismic acid [27]. Shikimate kinase (SK, EC 2.7.1.71), the fifth enzyme in the shikimate biosynthetic catalyzes phosphate transfer from ATP to the carbon-3-hydroxyl group of shikimate, forming shikimate 3-phosphate. SK belongs to the nucleoside monophosphate (NMP) kinase structural family where the characteristic feature of the NMP kinases is that they undergo large conformational changes during catalysis and belongs to the P-loop containing nucleoside tri-phosphate hydrolase superfamily (Pfam Clan: AAA:CL0023) [28]. The NMP kinases are composed of three domains. The CORE contains a highly conserved phosphate-binding loop (P-loop), the LID domain, which undergoes substantial conformational changes upon substrate binding, and the NMP-binding domain, which is responsible for the recognition and binding of a specific substrate [29]. The SK crystal structures show that SK exists as a monomer with a single ATP binding site and MgADP induces concerted hinged movements of the shikimate binding and LID domains causing the two domains to move towards each other in the presence of this ligand [29].

Phosphofructokinase (PFK, fructose-6-phosphate 1-kinase, EC 2.7.1.11) is a classical allosteric enzyme that catalyzes the phosphorylation of D-fructose 6-phosphate (Fru-6-P) by Mg-ATP to form D-fructose 1,6-bisphosphate and MgADP. PFK from B. stearothermophilus is a homo-tetramer with each subunit having a molecular weight of 34 kDa, which undergoes a concerted two-state allosteric transition [30]. PFK belongs to the PFK-like superfamily (Pfam Clan: PFK:CL0240) The enzyme from B. stearothermophilus (Bs-PFK) shows hyperbolic Michaelis-Menten kinetics with respect to both Fru-6-P and Mg-ATP, but cooperative kinetics in the presence of allosteric inhibitor phosphoenolpyruvate(PEP) [31]. Unliganded Bs-PFK is in the active R state, which has high affinity for substrate, switching to the inactive T state with low affinity for substrate only in the presence of PEP.

Glutamine synthetase (GS) (EC 6.3.1.2) catalyzes the reversible conversion of L-glutamic acid, ATP and ammonia to L-glutamine, ADP and inorganic phosphate via a γ-glutamyl phosphate intermediate [32]. As GS is a central enzyme in nitrogen metabolism the enzyme is regulated by at least four different mechanisms: (a) adenylylation and deadenylylation of the tyrosine 397 residue, (b) conversion between a relaxed (inactive) and taut (active) state depending on the divalent metal cation present, (c) cumulative feedback inhibition by multiple end products of glutamine metabolism, and (d) repression and derepression of GS biosynthesis in response to nitrogen availability [32]. Escherichia coli GS is a large, metalloenzyme (~624 kDa) comprising 12 identical subunits arranged in two face-to-face hexagonal rings [33].

E. coli GS belongs to the glutamine synthetase 1-β group of enzymes that are regulated via adenylylation of a single tyrosine residue [34], with each subunit requiring two structurally implicated divalent cations (either Mg²⁺ or Mn²⁺) for its catalytic activity.

The extent of adenylylation of the E. coli GS in response to an excess or deficiency of nitrogen in the growth environment is regulated in response to the intracellular concentrations of 2-ketoglutarate and glutamine, via the reversible adenylylation of a tyrosine residue (Tyr397) in each subunit of GS [32, 35–39]. The presence of adenylylated GS (GS₁₂) predominates in a nitrogen-rich, carbon-limited media, while the deadenylylated form (GS₀) tends to predominate under conditions of nitrogen limitation [32, 35–46].

Comparative steady-state enzyme activity assays were run to determine the effect of ATP and C8D-ATP on the specific activity of a range of kinase and synthetase enzymes. The reason for the investigation was to establish whether the C8H of ATP plays a primary role in the binding of ATP. The enzymes investigated were hexokinase, acetate kinase, shikimate kinase, phosphofructokinase, adenylylated (GS₁₂) and deadenylylated GS (GS₀). The coordination of ADP/ATP in the active sites of the enzymes under investigation, showing the interactions of the C8H with the active site, is outlined in Figure 1. It was demonstrated that the enzymes selectively bind C8D-ATP at low concentrations giving a KIE in excess of 2. It was also demonstrated that if given a combination of ATP and C8D-ATP the enzymes preferentially utilised the C8D-ATP.

The role of the KIE in the kinetics of the enzymes investigated has led to models for the regulation of the binding of ATP to be proposed (Figure 10). In classical steady-state kinetics as represented by the Briggs-Haldane modification of the Michaelis-Menton formulation (Equation 1),

(1)

and k_on = k₁, k_off = k_-1 and k₂ >> k_-1, and the Michaelis constant, K_M is obtained from

(2)

Where E, S and P are the enzyme, substrate (ATP) and product (ADP) concentrations respectively. In monomeric enzymes such a shikimate kinase the effect of the concentration of ATP and C8D-ATP on the relative enzyme activities results in the KIE asymptoting from 2 to 1. If the C8H of ATP is not directly implicated in phosphoryl transfer, the deuteration of ATP may affect the equilibrium of the reaction either by affecting the binding of ATP, k₁, or by the release of ADP, k₃. In monomeric enzymes which follow Michaelis-Menton kinetics, K_M is dependent on k₁, k₂ and k₃. If the C8D impacts directly on the overall reaction mechanism then the KIE is as a result of breaking the bond to the isotopically labelled atom. The regulation may however, be dependent on the equilibrium of binding of ATP, k₁, or the equilibrium of release of ADP, k₃, also being implicated in the breaking and re-making to the C8H bond. The regulation of the reaction rate by both C8D-ATP per se and the concentration of C8D-ATP or ATP probably arise as a result of the same equilibrium constant, either k₁ or k₃. At low ATP concentrations the deuteration of C8 may result in the retardation of the release of ADP from the active site and at high ATP concentrations the concentration effect of ATP results in a reduction in the rate of release of ADP from the active site. Due to the magnitude of the KIE it would appear that bond-breaking of C8H(D) is implicated in the binding of ATP and release of ADP. The impact on k₃ does not appear to be the cause of the high KIE at low concentrations where the KIE is the largest. The effect of the increase in the ATP/C8D-ATP concentration on the KIE therefore only manifests as the classical effect with the KIE being of the order of 2.0 as determined by v_H/v_D, at low concentrations, asymptoting to 1 at high ATP concentrations where both concentration and the role of C8D impact on the KIE. At low concentrations of ATP the enzyme activity is dominated by the impact of the C8H/C8D on the equilibrium of binding. At low concentrations the C8H plays the predominant role in the equilibrium of binding as in the case of shikimate kinase if the enzyme is exposed to a mixture of ATP and C8D-ATP, where the activity follows that of the C8D-ATP (Figure 9). The K_M for ATP is however lower than for C8D-ATP (Table 1). If k₁ for ATP is greater than k₁ for C8D-ATP the response seen in the mixture would manifest similar to a reduction in the concentration of ATP thereby reducing the overall rate of reaction. At low ATP concentrations the mass spectroscopy data in fact indicates that the ATP is preferentially utilised as seen by the relative concentrations of ADP in the assay solutions whereas at high concentrations the relative concentrations of ATP and C8D-ATP are equivalent (Additional files 1, Figure S1B). At low ATP concentrations k₁ (k_on) predominates and as the concentration of ATP increases the concentration effect plays an increasing role in the KIE thereby negating the impact of the C8H on the KIE. The reduction in the overall enzyme activity by C8D-ATP may be as a result of the breaking of the C8D bond having a direct impact on k₂, the phosphoryl transfer reaction. As the classical H/D KIE is of the order of 2, as the concentration of ATP tends towards the concentration at the maximum specific activity, v_max, where the concentration effect is at its maximum the effect of the C8H/C8D on the KIE is at a minimum and the KIE tends towards 1.

In oligomeric enzymes it is proposed that the deuteration of ATP not only affects the binding of ATP to the site where catalysis is occurring but the deuteration also affects the interaction between sites. In oligomeric kinases it is proposed that mechanistically two modes of regulation occur, one which is dependent on the release of ADP from the first active site before ATP binds to the second active site (Figure 10B) and the second mode of regulation depends on the conversion of ATP to ADP prior to the binding of the ATP to the second active site (Figure 10C). In the mechanism outlined in Figure 10C binding to the second site can occur prior to the release of ATP from the first site provided the reaction from ATP to ADP has occurred.

It is proposed in enzymes such as acetate kinase, hexokinase and GS₀, which utilise the coordinated half-sites mechanism of regulation, the enzyme kinetics follows classical allosteric kinetics where an equilibrium is set up between the enzyme concentration [E] and the substrate concentration [S] and binding of the second ATP is dependent on the conversion of the second active site into an ATP binding form by the release of ATP from the first active site. In enzymes using this mechanism of regulation, K_M is dependent primarily on k_-1, k₂ and k₃. The KIE obtained in these enzymes asymptotes to a value of 1 from low levels (KIE_D values asymptote to levels significantly above 2). At low ATP concentrations the effect of the deuteration of C8 is to allow binding to occur for long enough to allow the reaction to occur and negate the effect of k_-1, thereby shifting the equilibrium to k₂. At low ATP concentrations therefore the impact of the deuteration on the binding is to retard the release of the ATP. At high ATP concentrations the impact of the ATP concentration relative to the impact of ATP binding on the rate of reaction is significantly higher and as a result there is a concomitant increase in the KIE. The impact of binding and the reaction rate however equilibrate to a KIE of 1. The maximum rate of binding can only ever be equivalent to the maximum rate at which the second ATP binding site is converted to the ATP binding form by the release of ADP from the first site (Figure 10B). The inverse KIE, KIE_D, at low ATP concentrations is as a result of an increase in the probability of the reaction occurring resulting from the deuteration and not as a result of activation of binding per se, as in the half-sites-activity mechanism there is no activation of the activity of the second site as a result of the interaction with the first site. The classical impact of deuteration on the KIE when the KIE is a primary effect, as determined by v_H/v_D, should yield a KIE of 2 or more [4, 5]. As the regulation of the enzyme activity and ligand binding in these enzymes function in a coordinated half-the-sites manner binding in the second site only occurs on release of the ADP from the first site, it is therefore proposed that deuteration of the ATP improves the binding characteristics. As the equilibrium shifts towards the impact of increasing ATP concentration on the enzyme activity the deuterated ATP binds effectively twice as efficiently as the non-deuterated ATP thereby negating the impact of the deuteration on the apparent enzyme activity at high ATP concentrations, yielding a KIE of 1.

In enzymes where the second active site is made amenable to ATP binding by the conversion of ATP to ADP, in other words binding may occur to the second site prior to the release of the ATP from the first site, the K_M is dependent on k₁ and k₂. This occurs in the case of phosphofructokinase and GS₁₂ where the KIE becomes 2 at v_max (Figure 6 & 7). The impact of this binding is that at any point in time up to two or more reactions might be occurring simultaneously in two active sites. Unlike in the half-sites mechanism in these enzymes activation of the activity in the second site (or more) might occur from the first site. The binding per se therefore might have an effect on the KIE_D. In multimeric enzymes this effect might be greater. At high concentrations the deuterated ATP binds twice as efficiently as the non-deuterated ATP this allows the KIE to asymptote to 2 or more. At low concentrations the deuteration has the same effect as occurs in the previous model, whereby binding occurs for a long enough period to negate the effect of k-₁. At high concentrations the effect of deuteration is superseded by the concentration effect and as two or more active sites are able to function simultaneously, this allows the KIE to asymptote to 2 or more. It is proposed that a result of the adenylylation of GS it allows for the regulation of the enzyme by a similar mechanism as occurs in phosphofructokinase. Bacterial PFK is a homoteramer, with the four subunits assembled as a dimer of dimers [47–49]. It is conceivable that on adenylylation of GS the interaction between two-subunits effectively creates a dimer of dimer interaction.

The data outlined clearly demonstrates the role of C8H of ATP in the kinetics and regulation of a number of kinase and synthetase enzymes. The KIE is clearly a primary KIE. However, the extremely high values of the KIE_D obtained at low at concentrations in the case of the oligomeric enzymes does not appear to be as a result of the impact of the deuterium on the rate the phosphoryl transfer mechanism per se, but rather as a result of the role that the C8H plays in the equilibrium of binding of the ATP to the active site. Clearly the regulation of enzyme activity in kinases and synthetases is complex, which manifests in the apparent K_M of the kinases ranging from less than 0.4 μM to in excess of 1000 μM for ATP (Carna Biosciences, Inc., Kinase Profiling Book:http://www.carnabio.com). The findings of this investigation have demonstrated that the C8H of ATP plays a direct role in binding of ATP to the active site of enzymes. The labile nature of the C8H of ATP is well documented [50–52]. It is therefore conceivable that the role of the C8H of ATP in the kinetics and regulation of enzyme activity has been conserved in all kinase and synthetase enzymes as one of the regulatory mechanisms associated with binding of ATP to the active site of this diverse range of enzymes. The induction of the C8H to be labile by active site residues coordinated to the ATP purine ring may play a significant role in explaining the broad range of K_m associated with kinase enzymes.

The exact role of the C8H in the stabilization of the ATP-substrate transition state is unclear. All kinase and synthetase enzymes have an absolute requirement for the presence of a divalent metal ion, either Mg²⁺ Mn²⁺, for enzyme activity. The principal effect of the metal ion is to facilitate the nucleophilic attack by charge neutralization. It is conceivable that the labile proton also plays a role in the transient redistribution of charge during nucleophilic attack. It has been demonstrated that ATP analogues substituted at the C8 position significantly decrease the affinity of the analogues for cAMP-dependent protein kinase. As large molecular volume substitution at the C8 position leads to compounds existing primarily in the syn-conformation the data lead to the conclusion that ATP preferentially binds in the anti-conformation [53]. Competitive binding and inhibition of Na, K-ATPase was also obtained using ATP analogues substituted at the C8 position [54]. Na, K-ATPase was found to be able to hydrolyze a number of the C8-substituted analogues having unaltered triosphate fragment all be it at reduced rates. The 5 mM concentrations used would be in the domain where the concentration effect on binding would be at its highest. Notwithstanding this data it is conceivable that purine analogues deuterated at the C8 position will have enhanced inhibition properties over their non-deuterated counterparts. The role of C8H in GTP hydrolyzing enzymes and the extent to which the C8H also becomes labile during catalysis in GTP has not been defined; however, it is probable that similar mechanisms occur within these enzymes.

In an investigation on the conformational dynamics of M. tuberculosis shikimate kinase, the phosphate binding domain was found to be rigid and protected from solvent access, while the shikimate binding domain is highly flexible and the nucleotide binding domain is rigid at low temperatures, becoming more flexible at temperatures above 30°C [55]. The phosphate binding domain contains residues associated with the C8H of ATP [29, 55]. It is conceivable that the hydrogen bonding interaction that exists between the Thr17-OH on the phosphate loop, the C8H of ATP and the oxygen on the ATP α-phosphate plays a significant role in the labile nature of the C8H. The fact that the phosphate loop was also found to be rigid could also be significant in the role of the residues in facilitating binding and catalysis associated with the C8H-ATP.