Dynamic Combinatorial Chemistry Delivers a Reversible, Sub-Picomolar Trivalent Thrombin Inhibitor
- A PNA-templated dynamic combinatorial chemistry (DCC) platform assembles a three-fragment library (~125,000 possible combinations) using DNA/PNA hybridization handles for unbiased equilibration.
- Thrombin itself templates the library, selecting and amplifying the best-binding fragment combination; size-exclusion filtration coupled with MALDI mass spectrometry gives a full selection-to-readout cycle in under an hour.
- The selected trivalent assembly engages thrombin's active site plus both exosites simultaneously, driving exceptionally high (sub-picomolar-range) binding affinity and near-stoichiometric inhibition.
- Unlike covalent thrombin inhibitors, binding is fully reversible — pharmacologically relevant for an anticoagulant, where reversibility matters for managing bleeding risk.
- Demonstrates target-templated DCC as a rapid route to multivalent, exosite-spanning inhibitors, even against a well-studied target.
Sub-picomolar affinity out of a 125,000-member DCC library gets my attention immediately — that's an avidity effect, not a single tight pocket, and it's essentially bivalirudin's active-site/exosite-1 logic pushed one contact further to include exosite II as well. Exosite engagement should also buy real selectivity over trypsin-like serine proteases that don't share thrombin's exosite topology, though I'd want that confirmed head-to-head against fXa and trypsin. My hesitation is tractability: a DNA/PNA-templated trivalent assembly is a fantastic proof-of-concept and chemical tool, not yet a lead — that's a long way from oral. Against DOACs' one-pocket, small-molecule simplicity, this wins on affinity and reversibility but loses badly on drug-likeness; the real test is whether target-templated DCC can be steered toward a druggable combination, not just the tightest binder.
Discovery of a Reversible Sub-Picomolar Thrombin Inhibitor Using DCC ↗