Structural fluctuations within a membrane protein propagate between domains, but the direction in which these fluctuations propagate has been inaccessible. Here we develop two causal inference frameworks, CASCADE and MERIT, that extract this direction from molecular dynamics trajectories. They reveal a quantity invisible to experiment: conformational directivity. In transmembrane (TM)-juxtamembrane (JM) dimers of EGFR and FGFR3, fluid membranes produce TM-drives-JM directivity, the direction consistent with signal propagation. Tight packing between sphingomyelin and cholesterol reverses this entirely, imposing JM-drives-TM coupling inconsistent with signaling. The reversal is nonlinear, making directivity a binary switch controlled by the molecular interactions that define membrane phase. Disease-associated TM mutations (EGFR L658Q, FGFR3 A391E) destroy the switch by disrupting the lipid-protein interface through which ordered membranes constrain TM orientation. In wild-type receptors, the control is intact: the receptor is ready to signal; the membrane decides whether it may.
Sato, T., Tamagaki-Asahina, H.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 6
- Comments 0