Engineering DNA Nanodevices with Multi-site Recognition and Multi-signal Output for Accurate Intracellular LncRNA Imaging

Dynamic DNA nanodevices, known for their high programmability and controllability, are pivotal in intracellular biomarker imaging. However, these nanodevices often suffer from inadequate detection sensitivity and specificity due to limited cellular loading capacity and low signal feedback. Herein, we engineered an integrated multi-site recognition and multi-signal output of four-leaf clover dynamic DNA nanodevice (MEMORY) that enables sensitive and accurate intracellular long noncoding RNA (lncRNA) imaging. MEMORY features one fluorophore (FAM)-modified cross-shaped structure as spatial-confinement scaffolds loaded with four identical quenchers (BHQ1)-modified recognition probes (RPs), ensuring a low background signal initially. In the presence of target lncRNA, the multiple recognition sites of MEMORY facilitate hybridization with the target to selectively release the RPs, exposing the toehold region and outputting the green fluorescence (FAM) signal. Furthermore, the exposed toehold region can trigger efficient and rapid hybridization chain reaction (HCR) amplification, outputting the red fluorescence (Cy5) signal. MEMORY's multiple recognition sites increase the likelihood of target collisions, enhancing reaction efficiency, while its multi-signal output provides sequential feedback through FAM and Cy5, boosting overall signal intensity. With the lncRNA metastasis-related lung adenocarcinoma transcript 1 (MALAT1) as a detection model, MEMORY offers a linear detection range from 1 pM to 100 nM, with a limit of detection of 0.29 pM. We demonstrated that MEMORY can differentiate between normal and tumor cells based on intracellular MALAT1 imaging. This integrated DNA nanodevice will offer valuable tools for sensitive and accurate imaging of intracellular biomarkers.