Cascade-coupled colorimetric-fluorescent dual-signal detection of EGFR-positive extracellular vesicles via bifunctional MOF@Pt nanozyme integrated with a CHA-CRISPR system.

Abstract

Extracellular vesicle (EV)-based liquid biopsy holds great promise for glioma diagnosis, but its clinical translation remains hindered by inefficient isolation of disease-relevant EV subpopulations and insufficiently integrated signal validation. Herein, we develop a platform that integrates a bifunctional nanozyme system (phosphatase-like MOF and peroxidase-like Pt) with a CHA-CRISPR/Cas12a cascade, achieving selective isolation and ultrasensitive detection of EGFR-positive glioma-derived EVs. Defective UiO-66-NH2 loaded with Pt nanoparticles and functionalized with EGFR aptamers (UiO@Pt@Apt) is immobilized on Hook strand-modified glass 96-well plates via Apt-Hook hybridization. Upon introduction of EGFR-positive EVs, they specifically bind to aptamers on UiO@Pt@Apt and induce the release of UiO@Pt@Apt-EV complexes through perturbation and destabilization of the Apt-Hook interface. After EV lysis, EV-derived miRNA-21 activates the CHA-CRISPR/Cas12a cascade to generate a fluorescent signal, while the nucleotide fragments produced by Cas12a trans-cleavage are proposed to be hydrolyzed by phosphatase-like defective UiO-66-NH2 to generate PO43-. The PO43- may contribute to Pt-mediated TMB oxidation, thereby supporting colorimetric signal amplification. The platform achieved a colorimetric EV detection limit of 427 particles/μL after coupling with the CHA-CRISPR system, representing a 6.9-fold improvement over the UiO@Pt@Apt system alone, and enabled a miRNA-21 detection limit of 87.0 fM. In plasma samples from 25 glioma patients and 20 healthy donors, the combined readout achieved an AUC of 0.968, showing numerically better discriminatory performance than either single readout, although the improvement was not statistically significant by DeLong analysis. This work provides a promising strategy for EV-based glioma liquid biopsy and offers a potentially adaptable framework for cascade-coupled dual-signal biosensing.