Structure of PAP in complex with ATP and RNA. (A) The structure of the closed, ternary complex of PAP is shown in purple, and ATP and RNA are shown in white and yellow, respectively. The middle domains of the earlier PAP structures have all been superimposed with that of the ternary complex, and the earlier structures are shown as transparent ghosts to highlight the domain movements. (B). Two views of the surface representations of the PAP/ATP/RNA complex are shown. The color reflects the surface electrostatics where the color gradient from red to white to blue corresponds to −8KT to 0 KT to +8KT. The top view is rotated ~45 degrees relative to that shown in 1A. The -5 and -4, RNA nucleotides (yellow) are seen exiting a cleft formed by the three domains. The bottom view has been rotated ~180 about the vertical axis relative to the top, so that the ATP-binding side of the active site cleft is visible. Part of the ATP base (green) and the -2 nucleotide (yellow) are seen. The nucleotide at the 3′ end is hidden within the active site cavity.

Stereo views of the substrate binding sites of PAP. (A–C) Detailed substrate interactions formed in closed, ternary complex. ATP (yellow carbons) and the 3′ end (blue carbons) are shown along with PAP with the surrounding amino acids (green carbons) and water molecules (red spheres).

Kinetics and mutangenesis of PAP. (A) Total apparent free energy contributions to catalysis (Δ G_cat) from mutated residues. Values are derived from nucleotidyltransfer (utilizing ATP, CTP or GTP) or pyrophosphorolysis (PPi) reactions. (B) Relative difference free energy changes ΔΔG^‡ and ΔΔG_ES, against the total difference free energy changes for catalysis of nucleotidyltransfer, ΔΔG_cat. The ΔΔG_cat is the total difference free energy change of a particular enzyme in either reaction utilizing ATP (right) or CTP (left) relative to the cytidylyltransfer reaction, A_n + MgCTP ⇌ A_n-C + MgPP_i, catalyzed by K215A. The data points for ΔΔG^‡ (red) and ΔΔG_ES (blue) are relative to the appropriate term for the cytidylyltransfer reaction of the K215A and Y224F mutant (the lowest value in each case), respectively. Large values on the y-axis indicate either a faster rate, V₁ (ΔΔG^‡), or greater free energy release upon formation of the Michaelis complex (ΔΔG_ES) relative to the reference reaction. (C and D) Thermodynamics of PAP substrate specificity and catalysis. The free energy change upon E+A+B ⇌ EAB is a function of the product of K_iaKi_b (or K_aK_ib). Here, the relative difference free energies were calculated using K_b as the substrate binding term, rather than K_iaK_b., thus ΔG_ES here refers to EA+B ⇌ EAB. This comparison is valid since the K_ia term for the mutants were determined to be essentially invariant. In each case, the magnitude of the energy barrier from EAB to the transition state was calculated using V₁; the barrier from the transition state to EAP for wild type and Y224F was calculated using V₂ (Table II). (C) Results for the adenylyltransfer reactions catalyzed by wild type PAP (black), Y224F (blue), K215A (green), and for the cytidylyltransfer reaction of w.t. (red). (D) The wild type adenylyltransfer reaction with labels, illustrating ΔG terms discussed in the text. The magnitude of ΔG_IB was estimated from the CTP reaction according to: ΔG_IB = ΔG_cat^ATP −ΔG_cat^CTP, and represents the total free energy difference realized upon molecular recognition of the adenine base of ATP (relative to CTP).

Determinants of ATP and RNA binding. Interacting residues from the N-terminal, middle, and C-terminal domains are colored yellow, green, and cyan, respectively. Residues within cloud-shaped bubbles on either side of the bases indicate hydrophobic/van der Waals interactions. Water molecules (round circles) are color-coded based on the degree to which they are buried within the interior of the protein; those colored dark blue are completely buried (see Methods). Those colored grey-blue are at the protein surface. Others are colored intermediate shades of blue depending on how many shells of water needed to be removed in order for the water atoms to become exposed. Asp154 and the second Mg²⁺ ion are shown in grey. Eight additional water molecules interact with the triphosphate moiety of the ATP. These are not shown for clarity, and because the water structure in this region may be altered due to the D154 A mutation. The base at position -5 interacts with a neighboring PAP molecule and is not shown.

Comparison of the active site with other structures. (A) The alpha carbons of the N-terminal domain of the RNA-bound PAP molecule were superimposed with those of the two crystallographically-independent molecules from PDB code 1FA0. In the earlier structure each of the two molecules contained two molecules of 3′-dATP at the active site. The structure presented in this paper is shown in blue. The earlier structures are shown in green (molecule A) and yellow (molecule B). The middle and C-terminal domains are not shown because they occlude the nucleotides. (B) Superposition of the PAP active site with that of DNA polymerase beta (PDB code 2FMS) demonstrates very similar structure and geometry with respect to the ligand environment about metal B. PAP is shown in blue with Mg²⁺ ion in black. Polymerase beta is shown in green with its two metal ions in teal. As in polymerase beta, a water molecule (also present in the earlier yeast PAP structures) completes the octahedral geometry of metal ion B. Distances between the Mg²⁺ in the ternary PAP complex and its six ligands range from 2.02 – 2.16 Angstroms.

Free energy consideration of PAP. Conceptual free energy diagrams for the reaction segment, (EA + B⇌EAB_(open) ⇌EAB_(closed) ⇌E′AP), including a rate-limiting conversion of enzyme-bound substrates to products, are depicted. The arrows in each panel refer to free energy differences that correspond to important kinetic parameters: blue corresponds to K_m (K_b); red, V₁ (maximal rate in the forward direction); violet, V₂ (maximal rate in reverse direction). Panel (A) describes the reaction catalyzed by wild type PAP in which the closed conformation of the enzyme substrate ternary complex is strongly favored over the open conformation. Panel (B) depicts the energetic effects of the mutation of residues involved in tertiary interactions that stabilize EAB_(closed). The effect of such a mutation is to increase the energy of the EAB_(closed) (ground state) and EAB^‡ (transition state) by an equal amount, i.e. a uniform binding mutation. Panel (C) describes an alternative situation where the lowest-energy ground state is EAB_(open), as is evidently the case for the reaction utilizing the alternative nucleotide substrate, MgCTP. The red arrow in (C) is related to the apparent maximal velocity, V_obs, where V_obs = F_c×V and F_c is the fraction of enzyme substrate complexes that are in the active, closed state.