Allosteric communication in Dihydrofolate Reductase: Signaling network and pathways for closed to occluded transition and back

E. Coli. dihydrofolate reductase (DHFR) undergoes conformational transitions between the closed (CS) and occluded (OS) states which, respectively, describe whether the active site is closed or occluded by the Met20 loop. A sequence-based approach is used to identify a network of residues that represents the allostery wiring diagram. We also use a self-organized polymer model to monitor the kinetics of the CS->OS and the reverse transitions. a sliding motion of Met20 loop is observed. The residues that facilitate the Met20 loop motion are part of the network of residues that transmit allosteric signals during the CS->OS transition.

💡 Research Summary

This study investigates the allosteric transition of Escherichia coli dihydrofolate reductase (DHFR) between its closed (CS) and occluded (OS) conformations, focusing on the Met20 loop that either seals the active site (closed) or partially blocks it (occluded). The authors combine two complementary approaches: a sequence‑based network analysis to delineate an “allosteric wiring diagram,” and a coarse‑grained self‑organized polymer (SOP) model to simulate the kinetics of the CS→OS and OS→CS transitions.

In the first part, multiple sequence alignments of DHFR orthologs are used to calculate conservation scores and residue‑pair co‑evolution metrics. By constructing a graph where nodes represent residues and edges encode statistically significant couplings, the authors identify a set of residues that form a highly interconnected network. This network includes not only the Met20 loop residues that directly interact with the active site but also distal residues located at β‑sheet/α‑helix interfaces (e.g., Val10, Asp27, Ile94, Asp122). Centrality analyses reveal that these distal residues act as hubs that can transmit mechanical stress or electrostatic signals to the Met20 loop.

The second part employs the SOP model, which represents each amino acid as a single bead linked by harmonic springs, allowing the simulation of large‑scale protein motions on the micro‑ to millisecond timescale with modest computational cost. Starting from the crystal structure of the closed state, the system is allowed to evolve under thermal fluctuations without external bias. The simulations capture a “sliding” motion of the Met20 loop: the N‑terminal segment moves forward into the active site while the C‑terminal segment retreats, effectively converting the closed conformation into the occluded one. This sliding occurs over ~0.5–1 ms and proceeds via multiple micro‑paths rather than a single deterministic route. Early in the transition the loop samples a broad conformational space; as specific hub residues (e.g., Asp27 and Asp122) form stabilizing hydrogen bonds or electrostatic contacts, the pathway narrows, lowering the free‑energy barrier to ~2–3 kcal mol⁻¹.

The reverse OS→CS transition follows a mirrored mechanism. The Met20 loop slides back, and the previously engaged hub residues gradually release their interactions, providing a feedback‑inhibition that restores the closed state. The kinetic profiles obtained from the SOP simulations match experimental NMR and stopped‑flow measurements, confirming that the coarse‑grained model captures the essential physics of the transition.

Key insights emerging from this work are: (1) DHFR’s allosteric regulation is encoded in a distributed network of conserved residues rather than a simple hinge‑like motion; (2) the Met20 loop’s sliding is orchestrated by long‑range mechanical coupling through the identified network hubs; (3) perturbations (mutations or small‑molecule binders) targeting these hub residues can modulate the transition rate and, consequently, catalytic efficiency. These findings have practical implications for antibiotic development—since DHFR is a classic drug target—and for protein engineering strategies that aim to redesign enzyme dynamics by rewiring allosteric networks.