The Thrombin Binding Aptamer or TBA (5′-GGTTGGTGTGGTTGG-3′) is a 15-mer G-rich oligonucleotide able to inhibit the thrombin-catalysed fibrinogen-fibrin conversion after specific binding to its exosite I. TBA entered clinical trials but its evaluation was halted after phase I studies due to suboptimal dosing profiles. Aiming at obtaining TBA analogues better performing in vivo, a large number of chemically modified TBA variants have been proposed.

In this frame, we prepared a series of cyclic TBA analogues by linking its 5′ and 3′-ends with a variety of flexible linkers. The first derivative was realized introducing a 20-atom long linker. Compared to native TBA, it exhibited a G4 structure with exceptionally improved stability and nuclease resistance. However, these favourable properties were associated with reduced biological activity, suggesting that higher flexibility in the linker structure was necessary. Therefore, a mini-library of second generation cyclic TBAs (cycTBA I-IV) was prepared, carrying circularizing linkers overall spanning from 22 to 48 atoms. Among these derivatives, cycTBA II showed improved anticoagulant activity, associated with a dramatically stabilized G4 structure and enhanced enzymatic resistance in serum compared to the native TBA. Current studies are focused on pseudocyclic TBA analogues, where the cyclic structure is obtained not through covalent bonds but via p-p stacking or charge-transfer interactions of different aromatic probes inserted at the termini of the oligonucleotide. Among ten different TBA derivatives, we identified a promising candidate in this pseudocyclic series showing improved anticoagulant activity compared to native TBA, also having higher nuclease resistance and G-quadruplex thermal stability.