In vitro and in vivo phasor analysis of stoichiometry and pharmacokinetics using near-infrared dyes

Abstract

We introduce a simple new approach for time-resolved multiplexed analysis of complex systems using near-infrared (NIR) dyes, applicable to in vitro and in vivo studies. We first show that fast and precise in vitro quantification of NIR fluorophores lifetime and stoichiometry can be done using phasor analysis, a computationally efficient and user-friendly representation of complex fluorescence intensity decays obtained with pulsed laser excitation. We apply this approach to the study of binding equilibria by Förster resonant energy transfer (FRET), using two different model systems: primary/secondary antibody binding in vitro and ligand/receptor binding in cell cultures. We then extend our demonstration to dynamic imaging of the pharmacokinetics of transferrin binding to the transferrin receptor in live mice, elucidating the kinetic of differential transferrin accumulation in specific organs, straightforwardly differentiating specific from non-specific binding. Our method, implemented in a freely-available software package, has all the advantages of time-resolved NIR imaging, including better tissue penetration and background-free imaging, but simplifies and considerably speeds up data processing and interpretation, while remaining quantitative. These advances make this method attractive and of broad applicability for in vitro and in vivo molecular imaging, and could be extended to applications as diverse as image guided-surgery or optical tomography.

A 512 × 512 SPAD Image Sensor With Integrated Gating for Widefield FLIM

“…17 shows a representative decay from each sample, corresponding to a square region of interest (50×50 pixels) in the center of the array. Decays were fitted to a single-exponential integrated with a gate profile provided by the measured periodic IRF, using a standard Levenberg-Marquard algorithm implemented in a custom LabVIEW software (AlliGator [34]). Single-exponential fits of the R6G solution and QD655 decays are shown in Fig.…”

“…Fluorescence lifetime extraction of (a) Rhodamine 6G (R6G) solution, and (b) quantum dots (QD655) by fitting the decay by a single exponential integrated over a gate profile provided by the IRF sample, and pile-up correction [34]. The signal shown corresponds to the total signal within a square region of 50×50 pixels.…”

Computational macroscopic lifetime imaging and concentration unmixing of autofluorescence

“…A 3 × 3 spatial binning was applied to the resulting reconstructed dataset. All decays included herein were fitted using the Levenberg–Marquardt nonlinear least square fit (NLSF) algorithm implemented in Alligator [36]. An example of TPSF fitting is shown in Appendix A Figure A1D.…”

Ultra-fast fit-free analysis of complex fluorescence lifetime imaging via deep learning

“…Hence, quantification of lifetime-based quantities can be very challenging. To demonstrate the potential of FLI-Net for MFLI based on gated ICCD (and hence its potential for widespread FLI applications), we evaluated its performance in two settings: well-plate imaging with concentration-controlled mixtures of two NIR dye mixtures 16 The values obtained from the DNN (b) show a similar trend to that of the least-squares fit (c) but possess a histogram that is centered much more closely to the ground-truth values of both and .…”

“…The approach needs to be modified for techniques such as time-gated fluorescence 15 and typically requires some calibration samples to be quantitative. 16 In parallel, interests in data driven and model-free processing of imaging methodologies has boomed over the last decade. Of particular note, Machine Learning (ML) and Deep Learning (DL) methods have recently profoundly impacted the image processing field.…”