Digital TNA-CHA-TR-FRET platform for ultrasensitive miRNA detection in serum: A diagnostic tool for early esophageal...

Catalytic hairpin assembly (CHA) offers significant potential for point-of-care nucleic acid detection due to its enzyme-free, isothermal amplification. However, its clinical translation is hindered by critical limitations: background leakage, matrix interference in complex samples like serum, and insufficient quantitative accuracy. To address these challenges, we developed a novel Digital TNA-CHA-TR-FRET platform for ultrasensitive miRNA detection. Firstly, we introduced Threose Nucleic Acid (TNA) modifications at the stem ends of the hairpin DNA probes (TNA-CHA). This modification significantly increased probe thermal stability, effectively suppressing background leakage caused by transient DNA breathing while maintaining catalytic efficiency, with optimal performance achieved using two TNA base pairs. Secondly, to overcome serum autofluorescence interference, we integrated TNA-CHA with Time-Resolved Fluorescence Resonance Energy Transfer (TR-FRET). The time-delayed measurement window of TR-FRET effectively filtered out short-lived background fluorescence, achieving an impressive limit of detection of 1 aM in serum. Thirdly, to resolve the poor linearity and repeatability inherent to CHA kinetics, we incorporated a microfluidic digital partitioning strategy (digital-TNA-CHA-TR-FRET). Partitioning the reaction into nanoliter droplets significantly improved linear correlation (R² = 0.986) and repeatability (CV <5 % across concentrations). Clinically validated for early esophageal cancer diagnosis using serum miRNAs (miRNA21, miRNA25, miRNA93) on 100 samples (50 patients, 50 controls), our platform demonstrated high diagnostic performance using a "two-positive-out-of-three" miRNA logic gate: 96 % sensitivity, 90 % specificity, and 93 % accuracy. This integrated platform provides a powerful tool for non-invasive, early cancer diagnosis.