Fig. 1. Validation of a common “backdoor” problem for NIR probes.

a Construction of typical NIR dyes based on the TICT mechanism, including (I) Cyanine dyes, (II) Nile blue dyes, and (III) DCM dyes, with their electron acceptor (“A”) and donor (“D”) groups illustrated. b Albumin can bind to, and restrict the twisting of, the A end of these TICT dyes, which c produces unspecific and interfering fluorescent signals that mask the specific and expected signals from the analyte. Because of the abundance of albumin in the living system, this issue represents a common, but largely overlooked, “backdoor” problem for the in vivo application of such probes. d The chemical structures of representative TICT-based NIR dyes and fold changes (from white to red) of these dyes in response to HSA, BSA, MSA, heparin, PVA, and glycerol in PBS (pH 7.4, 10 mM). The fluorescence enhancement for a dye is determined by the ratio of its fluorescent intensity in a different condition to that in PBS. Con. indicates the concentration of the analytes. e The relative viscosity of HSA (black), BSA (red), MSA (blue), heparin (dark cyan), PVA (magenta), and glycerol (dark yellow) in different concentrations in PBS (pH 7.4, 10 mM). Data are represented as mean values ± SD (n = 3 independent experiments). f ¹H NMR titration of different equivalent albumin (0.1−0.8 equiv.) to Cy5S in D₂O (from 6.0 to 9.5 ppm). Red frames highlight the proton signals on the A group. The concentrations of HSA, BSA, MSA, heparin, and PVA used: 1, 10 µM; 2, 50 µM; 3, 100 µM; 4, 300 µM; 5, 600 µM. The content of glycerol used: 1, 10%; 2, 20%; 3, 30%; 4, 40%; 5, 50%. Source data for Fig. 1d, e are provided as a Source Data file.