Exosomes are important cancer biomarkers, however, the accuracy of exosome detection is greatly reduced due to heterogeneity of each exosome. Detecting exosomes with a larger field-of-view (FOV) might be a good solution. Compound eyes offer unique advantages such as a large field of view, low aberration, and high temporal resolution. Bionic compound eyes aim to replicate such features and have broad applications in fields like machine vision and medical imaging. In this paper, we propose the fabrication and application of a bionic compound eye for quantitative detection of exosomes, which allows fluorescence imaging of exosomes with an enlarged FOV, achieving a detection limit as low as 9.1 × 102 particles/mL. The bionic compound eye is formed by simply replicating a fly eye with polydimethylsiloxane (PDMS). To detect exosomes, a microfluidic array chip compatible with the compound eye is designed. Exosomes are captured on the chip using CD63 aptamers as the capturing probes. Another kind of fluorescent aptamers are utilized to recognize the captured exosomes. Large FOV dual-color fluorescence (LFDF) imaging of these exosomes is realized by inserting the compound eye between the objective and microfluidic chip. The advantages of LFDF imaging include, first, dual-color fluorescence imaging can guarantee that we are indeed imaging exosomes; second, large FOV can reduce the impact of heterogeneity of exosomes. Thus, the reliability of assay results would be greatly improved. As a proof-of-concept, breast cancer exosomes were used as the example. The experimental results showed that, compared to imaging without the compound eye, the standard deviation of LFDF imaging results decreased by approximately 38%. Thus, the detection errors could be greatly reduced. The feasibility of using LFDF imaging for subtype classification of breast cancer exosomes was also preliminarily validated. This technology offers a new, low-cost, and highly accurate solution for exosome based cancer diagnosis.
Publications
Article type
Year

Nano Research 2025, 18(3): 94907203
Published: 11 February 2025
Downloads:40