Molecular simulations of enzymatic phosphorylation of disordered proteins and their condensates

Understanding the condensation and aggregation of intrinsically disordered proteins in a non-equilibrium environment is crucial for many biological processes. For instance, enzymatic phosphorylation of disordered protein regions profoundly affects their properties and interactions.

Multi-scale molecular dynamics simulations have contributed to a better understanding of the phase behavior of disordered proteins and the biological functionalities of their condensates. However, modeling enzymatic reactions remains a challenge, particularly verifying thermodynamic consistency in non-equilibrium conditions.

In our work, we use a Metropolis-like step to mimic out-of-equilibrium phosphorylation reactions and we show how to verify via Markov-state modeling that the resulting MD simulations satisfy local-detailed balance, enabling the study of driven biochemical systems.

Building on these results, we simulate (at residue-level resolution) the enzymatic phosphorylation of the disordered protein TDP-43 by the kinase CK1δ. TDP-43 is prone to phase separation and its pathological aggregation is a hallmark of neurodegenerative diseases. Our simulations show that sequence context and charge distribution, rather than residue position, determine phosphorylation rates in TDP-43. Enzymatic phosphorylation drives condensate dissolution, suggesting a cytoprotective mechanism against aggregation in disease. The simulations further reveal that the disordered region of CK1δ facilitates condensate binding, but can also auto-inhibit the enzyme in line with experiments.