Single-molecule fluorescence multiplexing by multi-parameter spectroscopic detection of nanostructured FRET labels

TL;DR

This paper introduces a new set of fluorescent labels called FRETfluors, enabling multiplexed single-molecule detection with high specificity and minimal overlap, advancing real-time molecular analysis.

Contribution

The authors engineered a diverse palette of FRETfluors with unique spectroscopic signatures for enhanced multiplexing at the single-molecule level, surpassing traditional fluorescence detection limits.

Findings

Successfully identified 27 FRETfluors in a mixture

Achieved discrimination between bound and unbound FRETfluors

Minimized classification errors using optimized construct subsets

Abstract

Multiplexed, real-time fluorescence detection at the single-molecule level is highly desirable to reveal the stoichiometry, dynamics, and interactions of individual molecular species within complex systems. However, traditionally fluorescence sensing is limited to 3-4 concurrently detected labels, due to low signal-to-noise, high spectral overlap between labels, and the need to avoid dissimilar dye chemistries. We have engineered a palette of several dozen fluorescent labels, called FRETfluors, for spectroscopic multiplexing at the single-molecule level. Each FRETfluor is a compact nanostructure formed from the same three chemical building blocks (DNA, Cy3, and Cy5). The composition and dye-dye geometries create a characteristic F\"orster Resonance Energy Transfer (FRET) efficiency for each construct. In addition, we varied the local DNA sequence and attachment chemistry to alter the Cy3 and Cy5 emission properties and thereby shift the emission signatures of an entire series of FRET constructs to new sectors of the multi-parameter detection space. Unique spectroscopic emission of each FRETfluor is therefore conferred by a combination of FRET and this site-specific tuning of individual fluorophore photophysics. We show single-molecule identification of a set of 27 FRETfluors in a sample mixture using a subset of constructs statistically selected to minimize classification errors, measured using an Anti-Brownian ELectrokinetic (ABEL) trap which provides precise multi-parameter spectroscopic measurements. The ABEL trap also enables discrimination between FRETfluors attached to a target (here: mRNA) and unbound FRETfluors, eliminating the need for washes or removal of excess label by purification. We show single-molecule identification of a set of 27 FRETfluors in a sample mixture using a subset of constructs selected to minimize classification errors.