(4ed) Near-Infrared Fluorescent Nanosensors for High Spatiotemporal Neuropeptide Imaging

Neurochemical signaling underlies neuronal communication, and its dysfunction lies at the core of neurological disorders. However, many of these molecules remain invisible due to the absence of real-time biosensors, leaving neurochemistry as “the missing dimension of neurobiology”. I leverage the intrinsic fluorescence of single-walled carbon nanotubes (SWCNTs) to develop biological sensors capable of real-time neurochemical imaging. SWCNT can be noncovalently functionalized with molecular recognition elements, such as single stranded DNA (ssDNA), to bind analytes of interest. Due to exciton-based photophysics, SWCNT fluorescence is modulated by the surrounding dielectric environment and hence by ssDNA-analyte binding, enabling analyte detection. The resultant nongenetically encoded nanosensors are biocompatible with fluorescence that is tissue transparent, highly photostable, and reversible on a millisecond timescale, and can be universally applied across species and developmental stages. Coupled with the one dimensionality of SWCNT nanomaterials, these nanosensors can image neurochemical signaling with high spatiotemporal resolution.

The sensitivity and selectivity of ssDNA-SWCNT nanosensors is dependent on the ssDNA sequence. Previously, the Landry Lab developed high spatiotemporal nanosensors for imaging neurochemicals using a sequence-to-analyte screening approach, in which (GT)₆ was found to respond most strongly to catecholamine neuromodulators. Achieving an analyte-to-sequence approach for increasingly complex neurochemical target analytes, such as neuropeptides, necessitated the design of a technique for rational nanosensor discovery. We thus established an evolution-based platform called SELEC, or Systematic Evolution of Ligands by Exponential Enrichment on Carbon Nanotubes, to identify high-affinity sequences for analyte detection with high-throughput. Following the first application of SELEC to discovering high-response nanosensors for the small-molecule neuromodulator serotonin, I validated sequences predicted by machine learning models to be high- and low-responders to serotonin, identifying five nanosensors with higher fluorescence response to serotonin than was previously found (>1.9). Further, analysis of the data revealed that machine learning model predictions improved when incorporating fluorescence measurements calculated over entire spectra rather than single wavelength peaks. I next confirmed the broad utility of SELEC to target neuropeptides for nanosensor development, identifying and characterizing nIROT-SELEC, a probe that reversibly enables real-time imaging of oxytocin while demonstrating selectivity over a suite of pharmacological agents and other neurochemicals, including the closely-related neuropeptide vasopressin. Most recently, I have utilized sequence analysis of an aptamer selection library to identify nIRPP, a probe that selectively detects pancreatic polypeptide over homologs Neuropeptide Y and Peptide YY.

Research Interests

I am keenly interested in studying the properties of nanosensors, encompassing both their fundamental aspects and practical applications in understanding neurochemistry. With expertise in nanomaterial synthesis and characterization, I particularly seek to develop my skills in the application of synthetic and chemical tools to interrogating neurobiological questions in model systems and toward clinical utility. Ultimately, by studying the chemical underpinnings of neurological diseases and disorders, I aim to improve diagnostic and therapeutic outcomes.