The functionalization of fluorescent single-walled carbon nanotubes (SWCNTs) has proven to be a versatile platform for the development of biosensors targeting small molecules, proteins, and environmental factors. In this study, we introduce a novel and generic strategy that utilizes random single-stranded DNA (ssDNA) libraries to evolve synthetic molecular recognition on the surface of SWCNTs.Building upon our previous successes, we present the evolution of molecular recognition, specifically targeting the neuromodulator serotonin. This evolution is achieved through a combination of directed evolution and machine learning-based prediction, focusing on the iterative optimization process to enhance the sensitivity and selectivity of ssDNA-SWCNT nanosensors.In addition to neurotransmitter sensing, we extend the application of our evolved random ssDNA-SWCNT complexes to the development of cancer cell-targeting biosensors. Through positive selection, these complexes can be tailored to recognize and bind selectively to cancer cells. Simultaneously, negative selection ensures minimal interaction with normal cells. Those results showcase the broad applicability of our approach, suggesting that the elaborate functionalization of SWCNTs with specific ssDNA sequences can be universally employed for the rapid and convenient fabrication of novel biosensors and artificial antibodies. This research represents a significant step towards the development of highly specific and efficient nanosensors for diverse biomedical applications.
Read full abstract