The ternary extrinsic complex of the coagulation pathway (Tissue Factor (TF): Factor VIIa (FVIIa): Factor X (FX)) is critically important to the regulation of blood clotting. Activation of FX in this membrane-bound complex is vital to both the major pathways which produce the explosive burst of thrombin required for clot formation to proceed. Despite high interest in this complex as a potential drug target for hemophilia and thrombosis, little has been determined experimentally concerning its structure due to the lipid-dependent nature of its formation. As the complex will only form on the surface of a membrane with a specific composition of anionic phospholipids, experimental structure determination through NMR or X-ray crystallography is prohibitively difficult.

We will present the first atomic-level resolution structure of the ternary complex to include membrane interactions, developed using a novel methodology which combines protein-protein docking and extensive non-equilibrium molecular dynamics simulations. Over 560,000 protein-only structures predicting interactions between key domains of the extrinsic complex were first developed using the DOT2 protein docking program. Top docked structures were selected based on agreement with experimental information, and used to identify residues of FX and TF:FVIIa likely to interact. Protein-protein contact information from docking was then used to developed biased non-equilibrium simulations in which the complex was induced to form along a particular pathway. Using five non-equilibrium simulations of 200 ns, binding between membrane-bound structures of FX and TF:FVIIa was slowly induced. The highly mobile membrane mimetic (HMMM) model, an advanced membrane model which replaces slow-moving lipid tails with detergent molecules, was used to allow increased sampling of protein-lipid interacts during this deliberate complex formation. Each of the five resulting complexes was simulated for an additional 200 ns without biases, and putative key protein-protein and protein-lipid interactions involved in complex formation were identified. Residues on the membrane-binding domain of FX where found in the simulations to bind residues of TF previously identified experimentally as critical for complex formation. This structure provides not only the best available information on protein binding for the extrinsic complex but the only atomic-level information on the protein-lipid interactions which are crucial for its formation.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.