The facilitative transport of hexose through the GLUcose Transporter type 4 (GLUT4) is essential for cellular metabolism and is regulated by allosteric nucleotide interactions. In this study, we conducted a comparative biophysical analysis of glucose translocation using both the native Cryo-EM structure (7WSN) and the AI-predicted AlphaFold model using Steered Molecular Dynamics (SMD) at physiological (310.15 K) and attenuated (303.15 K) temperatures with two different allosteric modulators, ATP and ADP. While the AlphaFold model managed to capture baseline static topologies, dynamic profiling revealed that it intrinsically over-optimizes for static stability, immediately collapsing into a sterically occluded, hyper-packed artifact. Translocation through this constricted geometry forces severe steric solvent exclusion, abruptly stripping the substrate's hydration shell and resulting in immediate and immense thermodynamic friction (>300 kJ/mol). Under ATP-bound physiological strain, this hyper-bonded structural clamp prevents functional relaxation, ultimately inducing a catastrophic kinematic failure. Furthermore, ensemble Dynamic Cross-Correlation Matrix (DCCM) analysis demonstrates that the AI-predicted model suffers from persistent rigid-body locking, forcing the transmembrane domain to behave as a kinetically trapped conformation. In contrast, the functionally hydrated Cryo-EM architecture actively maintains a dual-gating pore and successfully isolates the exact mechanical lever driving transport: a native Proline hinge (PRO379) that seamlessly routes allosteric signals from the nucleotide anchor to the pore gate. Within the AlphaFold model, this critical allosteric wiring is disrupted, forcing mechanical coupling through an unphysiological, hyper-correlated aromatic pathway. These findings yield novel atomistic insights into GLUT4 gating mechanics, while definitively establishing that rigid AI-generated architectures lack the essential internal free volume and conformational plasticity required to sustain accurate thermodynamic profiling of dynamic membrane transporters.
S, G., Das, S., Lobo, A., Rahul, C. N., Gideon, D. A.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 3
- Comments 0