Mutation of a kinase allosteric node uncouples dynamics linked to phosphotransfer

Abstract

The expertise of protein kinases lies in their dynamic structure, wherein they are able to modulate cellular signaling by their phosphotransferase activity. Only a few hundreds of protein kinases regulate key processes in human cells, and protein kinases play a pivotal role in health and disease. The present study dwells on understanding the working of the protein kinase-molecular switch as an allosteric network of "communities" composed of congruently dynamic residues that make up the protein kinase core. Girvan-Newman algorithm-based community maps of the kinase domain of cAMP-dependent protein kinase A allow for a molecular explanation for the role of protein conformational entropy in its catalytic cycle. The community map of a mutant, Y204A, is analyzed vis-à-vis the wild-type protein to study the perturbations in its dynamic profile such that it interferes with transfer of the γ-phosphate to a protein substrate. Conventional biochemical measurements are used to ascertain the effect of these dynamic perturbations on the kinetic profiles of both proteins. These studies pave the way for understanding how mutations far from the kinase active site can alter its dynamic properties and catalytic function even when major structural perturbations are not obvious from static crystal structures.

Keywords: allostery; catalytic cycle; community maps; protein dynamics; protein kinases.

Fig. 1.

The Y204A node: sequence conservation, structural role, and dynamic changes in PKA. (A) Position of the Y204 residue in the C lobe of PKA kinase. Y204 is adjacent to the P+1 loop and makes extensive hydrogen interactions with key residues including E230 from the αF-helix that supports the kinase spines. (B) Sequence conservation score for Y204 across the kinome. The residue is almost as conserved as other key elements of the kinase structure like the conserved K72-E91 salt bridge, the DFG motif, and HRD motif. In AGC kinases, this residue is usually a Tyr, whereas in some tyrosine kinases it is a Trp. (C) Communities detected in the allosteric network of PKA kinase mapped onto its structure. Each community is around 75 residues big but is smaller than a subdomain. Functional roles of communities are provided in detail in Fig. S1. (D) Communities and their residue populations detected for the wtPKA:ATP:1Mg²⁺ complex. (E) Communities and their residue populations detected for Y204A:ATP:1Mg²⁺ complex. (F) Hydrophobic spines that connect the two lobes of the kinase domain. (G) Community distribution of the main chains and side chains of the hydrophobic spine residues in the wtPKA and Y204A.

Fig. S1.

Each state of the kinase catalytic cycle is described by a distinct “community map.” (A) The kinase catalytic cycle comprises distinct dynamic phases of the kinase domain that are allosterically regulated for commencing an efficient phosphotransfer. (B) Each dynamic phase of kinase domain in the catalytic cycle can be described by a distinct resonance that couples each residue of the kinase bilobal structure. Every phase is hence defined by a distinct community map. (C) Community maps of PKA kinase available for study. A direct comparison of the community maps are made for the wtPKA and Y204A molecular simulations at the same phase of catalytic cycle (protein:ATP:1Mg²⁺). (D) Mapping of the different dynamic communities on the secondary structures of PKA. Thirteen distinct communities can be identified for PKA. Each phase in the catalytic cycle has a subset of these 13 communities.

Fig. S2.

Comparison of the community map network for the wtPKA kinase vs. the Y204A mutant. (A) The 13 communities detected for PKA kinase and their functional roles. (B) The community distribution of the wtPKA and Y204A as described by the “bridging residues.” These residues lie at the boundaries of various communities such that their main chains are a part of one community and side chains are a part of another community. These boundary residues describe the landscape distributions of the community map. For the wtPKA vs. Y204A, the distribution of these boundary residues is heavily altered, explaining the extreme change in protein dynamics achieved by a single point mutation.

Fig. 2.

Enzyme kinetic rates for the transfer of phosphate from ATP to substrate peptide or water for the Y204A vs. the wtPKA. (A) Kinetic plots for steady-state phosphotransfer assay for transfer of γ-PO₄ from ATP to Kemptide. (B) Kinetic plots for pre–steady-state phosphotransfer assay for transfer of γ-PO₄ from ATP to Kemptide. (C) Rate constants for the steady-state, pre–steady-state phosphotransfer assay, and steady ATPase assay. (D) Kinetic plots for steady-state ATPase assay for transfer of γ-PO₄ from ATP to water.

Fig. 3.

Dynamics of nucleotide binding at the PKA active site. (A) ATP binding pocket of PKA. (B) Isothermal binding constants K_D for various nucleotides obtained for wtPKA and Y204. Isotherms are provided in Fig. S3. (C) Community structure for the main chains and side chains of residues lining the ATP binding pocket for wild-type PKA and Y204A mutant.

Fig. 4.

Dynamics of pseudosubstrate binding at the PKA active-site cleft. (A) Pseudosubstrate peptide interacting residues at the PKA active-site cleft. (B) Stability of apo, binary (protein:nucleotide), and ternary (protein:nucleotide:pseudosubstrate) states of wtPKA and Y204A mutant as seen by differential scanning fluorimetry. Melt plots are provided in Fig. S4. (C) Community structure for the main chains and side chains of pseudosubstrate interacting residues for the wtPKA and Y204A mutant.

Fig. S3.

Isotherms for nucleotide binding of wtPKA and Y204A mutant using internal tryptophan fluorescence spectroscopy. (A) Binding of ATP to wtPKA and Y204A mutant. (B) Binding of AMP-PNP to wtPKA and Y204A mutant. (C) Binding of ADP to wtPKA and Y204A mutant. (D) Binding constants obtained for various nucleotide for the wtPKA and Y204A mutant proteins. All curves were fitted to single-ligand binding isotherms using the nonlinear regression module of SigmaPlot software.

Fig. S4.

Differential scanning fluorimetry melt curves for the WT PKA and Y204A mutant using Sypro-orange hydrophobic dye as an indicator. (A) Melt curve for the apo wtPKA and Y204A mutant proteins. (B) Melt curve for the binary complex (kinase plus nucleotide) wtPKA and Y204A mutant proteins. (C) Melt curve for the ternary complex (kinase plus nucleotide plus pseudosubstrate inhibitor peptide) wtPKA and Y204A mutant proteins. (D) Temperature of melting for the apo, binary, and ternary complexes of wtPKA and Y204A proteins obtained by fitting the sigmoid melt curves by the nonlinear regression module of SigmaPlot software.

Fig. 5.

Nucleotide and substrate synergy forms a pivotal aspect of the kinase reaction. (A) Kinase catalytic cycle described for PKA. Synergy between stages I and II form the basis of assay used in B and C. Inset shows the stage of the catalytic cycle used for community map analysis for comparing the dynamics of Y204A mutant with wtPKA. (B) Isotherm for binding of nucleotide saturated wild-type PKA vs. Y204A to FAM-IP20 by fluorescence polarization. (C) Isotherm for FAM-IP20 saturated wild-type PKA vs. Y204A to the nucleotide AMP-PNP as seen by fluorescence polarization.

Fig. 6.

Changes in conformational entropy of catalytic residues in Y204A. (A) Dihedral angles for main chains and side chains of catalytic residues that vary in their populations for the wild-type PKA simulations vs. Y204A mutant. Crucial catalytic residues have altered rotamer preferences in the Y204A dynamics as seen by their dihedral plots. (B) Community structure for the main chains and side chains of residues that line the active site cleft of connecting the two lobes of the kinase structure in wtPKA and Y204A. (C) H/D exchange seen for the active-site peptides 162–172 and 184–187 that show altered dynamics in molecular simulations (reproduced and adapted from ref. 19).

Fig. S5.

Y204A shows altered global H/D exchange for the kinase domain. (A) H/D exchange for peptides obtained for the kinase domains of wtPKA and Y204A after 10 s. (B) H/D exchange for peptides obtained for the kinase domains of wtPKA and Y204A after 30 s. (C) H/D exchange for peptides obtained for the kinase domains of wtPKA and Y204A after 100 s. (D) H/D exchange for peptides obtained for the kinase domains of wtPKA and Y204A after 300 s. (E) H/D exchange for peptides obtained for the kinase domains of wtPKA and Y204A after 1,000 s. (F) H/D exchange for peptides obtained for the kinase domains of wtPKA and Y204A after 3,000 s. (G) H/D exchange and the various peptides for various time points mapped onto the structure for wtPKA and Y204A mutant (data reproduced and adapted from ref. 21).

Fig. S6.

The Y204A shows altered internal side-chain dynamics as seen from the residue dihedral angles. Dihedral plots for the various backbone (Phi-Psi) and side chains (Chi 1, Chi 2, Chi 3, Chi 4, Chi 5 dihedral angle lots for various residues in the Y204A simulations that deviated significantly from the wtPKA simulations). A vast number of residues showed this altered behavior and included residues that were much further away from the position of the Y204 itself. These plots are also indicative of a global rather than a localized impact on protein dynamics in the Y204A mutant protein.