Programmable DNA-Based Dimerization Networks for Dynamic and Competitive Control of Transcription.

Here, we introduce a minimal, experimentally validated DNA-encoded analogue of many-to-many protein dimerization networks to achieve programmable control of cell-free transcription. To do so, we rationally designed DNA monomers bearing orthogonal azide or DBCO handles that assemble via SPAAC chemistry into a combinatorial library of covalent dimers where only a single dimer completes an otherwise inactive promoter and activates transcription of a light-up RNA aptamer. By expanding the network from 2 to 22 monomers, we show that the RNA yield can be finely tuned over more than an order of magnitude, in quantitative agreement with a simple combinatorial model in which the activator dimer fraction decreases as a function of network size. We also designed sequence-defined inputs that selectively sequester monomers to reduce network size and upregulate chosen dimer activators, enabling multiplexed, orthogonal activation of different templates in the same solution from a shared monomer pool. Coupling the dimerization layer to a CRISPR-Cas12a collateral cleavage module, we further show that competitive DNA dimerization transcription logic can be interfaced with downstream enzymatic reactions to convert input sets into amplified cleavage activity. Together, these results establish competitive covalent DNA dimerization networks as a modular and quantitatively predictable platform for implementing competitive many-to-many dimerization logic for controlling cell-free transcription outputs.