Rationally architected MOF-derived Co₃O₄@NiMn-LDH hollow heterostructure-based sensor array empowering sensitive...

Catecholamine neurotransmitters and their metabolites are important biomarkers associated with neurodegenerative diseases, yet their accurate discrimination and quantification remain challenging due to structural similarity and complex biological environments. Herein, we report a tri-modal sensing platform based on metal-organic framework (MOF)-derived Co₃O₄ hollow nanocubes with in situ grown NiMn layered double hydroxide (Co₃O₄@NiMn-LDH) for the discrimination and quantification of catecholamine-related biomarkers. The oxidase-like performance of Co₃O₄@NiMn-LDH was significantly enhanced (1.98 U mg^-1) through rationally regulation of the thickness of the NiMn-LDH shell. Using o-phenylenediamine (oPD) as the signal-responsive substrate and F-doped SiQDs as a blue-emissive fluorescent probe, the platform enabled colorimetric and ratiometric fluorescence sensing. Owing to their different reducing abilities, catecholamines and their metabolites inhibited oPD oxidation to varying extents, suppressing 2,3-diaminophenazine (DAP) formation and generating distinct absorbance signals. The decreased DAP production reduced fluorescence at 565 nm, while the emission of F-doped SiQDs at 469 nm was restored, yielding a reliable ratiometric fluorescence response. In the electrochemical channel, catecholamine-related compounds generated distinct current responses due to their different electrooxidation activities. Therefore, the tri-modal sensing platform was established for the quantification of epinephrine (EP), dopamine (DA), norepinephrine (NE), vanillylmandelic acid (VMA), and homovanillic acid (HVA) with satisfactory linear responses over a wide concentration range (1-100 μM). Moreover, machine learning-assisted linear discriminant analysis (LDA) enabled effective discrimination of these biomarkers. The tri-modal platform exhibited reliable performance in complex samples, indicating its potential for multimodal analysis of structurally similar neuroactive molecules and neurodegenerative disease diagnosis.