Exploring the conformational landscape of protein kinases

Abstract

Protein kinases are dynamic enzymes that display complex regulatory mechanisms. Although they possess a structurally conserved catalytic domain, significant conformational dynamics are evident both within a single kinase and across different kinases in the kinome. Here, we highlight methods for exploring this conformational space and its dynamics using kinase domains from ABL1 (Abelson kinase), PKA (protein kinase A), AurA (Aurora A), and PYK2 (proline-rich tyrosine kinase 2) as examples. Such experimental approaches combined with AI-driven methods, such as AlphaFold, will yield discoveries about kinase regulation, the catalytic process, substrate specificity, the effect of disease-associated mutations, as well as new opportunities for structure-based drug design.

Figure 1.. Structural requirements of an active kinase.

The C-spine and R-spine are shown as gray spacefill. The C-spine is completed by ATP (not shown). Extension of the A-loop enables access of the substrate to the substrate docking site. The αC helix-in position is stabilized by formation of a critical salt bridge shown as ball-and-stick. The DFG residues with are also shown as ball-and-stick with the F pointing “in” and defining the DFG-in conformation.

Figure 2.. Conformational states of Abl detected by NMR.

Energy well diagram showing the ground (active) state and two excited (inactive) states of the kinase domain of Abl as determined by Xie et al. 2020 [19]. The proportion of each conformational state to the population is indicated beneath each well. PDB 6XR6 for the active state, 6XR7 for inactive state 1, and 6XRG for inactive state 2.

Figure 3.. Predicted conformational landscape of Abl.

Each dot represents a transient state with each color representing conformations similar to a common macrostate. Macrostates 1–3 represent conformations related to the active conformation defined by macrostate 1, macrostates 4 and 6 represent conformations related to the inactive I₁ (also referred to as E₁) conformation defined by macrostate 6, and macrostates 5–8 represent conformations related to the inactive I₂ (also referred as E₂) conformation defined by macrostate 7. The arrows represent the probability of transitioning between or within macrostates. Reprinted from Krishnan et al. (2022) [47], with the permission of AIP Publishing.