All reagents and solvents were purchased from commercial sources and used without further purification. Thin-layer chromatography was performed using Merck Silica gel 60 F-254 pre-coated plates and visualized using a thin-layer chromatographic chamber equipped with ultraviolet (UV) (λ = 254/365 nm) and visible light. Silica gel from Merck (particle size 100–200 mesh) and neutral alumina from Rankem were used for column chromatography. ¹H and ¹³C nuclear magnetic resonance spectra were recorded on Bruker 400- and 500-MHz spectrometers. High-resolution mass spectrometry data were recorded on MicrOTOF-Q-II mass spectrometer using acetonitrile as the solvent. All absorption spectra and fluorescence measurements were carried out using SHIMADZU UV-1800 spectrophotometer and HORIBA JobinYvon fluorimeter (fluorolog-3) using 1-cm path length quartz cuvettes.

The viscosity of different weight percentages of water/glycerol mixture was calculated from a previous report by Cheng (2008). The different weight percentage of water/glycerol mixture was prepared 10 ml each from this solution, 2 ml taken, and dye was added and mixed well using vortex (SCILOGEX vortex mixture MX-S) for 5 min, then immediately spectra were recorded at a fixed temperature of 25°C.

Dulbecco’s modified Eagle medium (DMEM), trypsin, antibiotic cocktail, and fetal bovine serum (FBS) were purchased from HiMedia (USA). Lyso-Tracker Green and MitoTracker Green were purchased from Thermo Fisher Scientific (United States). Our laboratory synthesized ER Tracker Green previously (Dutta et al., 2020). The 35-mm glass bottom imaging dishes were obtained from Ibidi (Germany, Cat# S28 81158). All the confocal microscopy imaging was performed with an Olympus FV3000 confocal laser scanning microscope. BHK-21 and U-87 MG cells were obtained from the National Centre for Cell Science, Pune, India, and were grown in a 25-cm² cell culture flask (Corning, United States) using DMEM (phenol red-free) containing 10% (v/v) FBS and 1% (v/v) antibiotic cocktail in 5% CO₂ at 37°C in a CO₂ incubator. For imaging purposes, cells were grown to 75–80% confluency in the 35-mm glass bottom imaging dishes (170 ± 5 µm) in DMEM with 10% FBS. The cells were washed twice with phosphate-buffered saline (PBS; pH 7.4) containing 5-mM MgCl₂. For the colocalization experiment, the cells were co-incubated with 0.2 µM of the JIND-Mor, and 300 nM of LysoTracker Green, 300-nM MitoTracker Green, and 2.5-µM ER Tracker Green for 15 min, and washed with PBS (pH 7.4) containing 5-mM MgCl₂ twice before imaging. For viscosity tracing, firstly, U-87 MG cells were incubated with 0.2-µM JIND-Mor for 15 min and washed twice with PBS (pH 7.4) containing 5-mM MgCl₂, then 50-µM dexamethasone (Dexa) was added and immediately observed on the confocal microscope for 60 min. Quantification of the lysosomal and Caenorhabditis elegans intensity was done using Image J software.

Hermaphrodite worms were grown in a nematode growth medium (NGM) at 20°C. For staining, C. elegans were synchronized and grown to young adult stage in NGM treated with 10-µM JIND-Mor for 60 h in 20°C. To induce osmotic stress, worms were synchronized and grown at 20°C until the first day of adulthood. The animals were transferred to NGM plates containing 200-mM NaCl for 8 h. Here, NGM plates containing 50-mM NaCl were used as control conditions. They were transferred to an agar pad on a glass slide and paralyzed using 5-mM levamisole and imaged under a confocal microscope. Quantification was done using Image J software using three different worms’ images.

The theoretical calculations were performed using the Gaussian 09 suite of the quantum chemical program (Frisch et al., 2009). Ground-state geometry optimization was performed with Becke’s three-parameter hybrid exchange functional with Lee–Yang–Parr correlation (B3LYP functional) using 6-311G as a basis set.

DCAJ was synthesized by condensation of compounds 1 and 2 (Scheme 2; see SI for details). Compound 1 was synthesized from the formylation of Julolidine using Vilsmeier–Haack reaction in an 83% isolated yield. However, compound 2 was obtained from the reaction of 4-methylacetophenone and malononitrile in a 75% yield. The reaction of 3-acetylindole and malononitrile yielded compound 3. Now to get JIND, compounds 1 and 3 were reacted in the presence of piperidine in isopropanol. JIND was further reacted with 1,4-dibromobutane to get the compound 5, and it was subsequently reacted with morpholine in dry dimethylformamide to obtain JIND-Mor in a 40% isolated yield. All compounds were characterized with nuclear magnetic resonance spectroscopy and mass spectrometry (see SI, Supplementary Figures S1–S17).

The solvent-dependent optical properties, such as ultraviolet–visible, and absorption and fluorescence of DCAJ, JIND, and JIND-Mor were investigated in detail (Supplementary Table S1; Figures 1A–F, Supplementary Figures S18–S20). A red-shift in the absorption and emission maxima of these compounds with solvent polarity confirms their ICT property. The ICT character of these molecules is also evident from the fragment molecular orbital calculation, as electron density in highest occupied molecular orbital is mostly distributed on julolidine (donor) and on malononitrile (acceptor) in the case of lowest unoccupied molecular orbital (Supplementary Figure S21). In water, these molecules have an absorption maximum of around 530 nm with an emission maximum of around 680 nm (Figure 1B). As shown in Figures 1G–J, depending on the solvent polarity, the color of the solution of DCAJ and JIND-Mor changes under visible and UV (365 nm) light. The probes are highly photostable, thermostable, and pH tolerant, as assessed from their unaltered fluorescence intensity (see SI, Supplementary Figure S22–S24).

To apprehend the molecular rotor properties of DCAJ, JIND, and JIND-Mor, we have investigated the viscosity-dependent change in fluorescence intensity and lifetime. In highly viscous solvents such as ethylene glycol and glycerol, we observed a noteworthy >150 times intensity enhancement, without any correlation with solvent polarity (Supplementary Figure S25). Such enhancement in the fluorescence intensity prompted us to investigate the viscosity-dependent emission properties of these compounds in detail. We observed a fluorogenic response upon moving from pure water (viscosity = 0.89 cP) to pure glycerol solution (viscosity = 905 cP), as shown in Figures 2A–C. A distinct visible color change from a nearly nonfluorescent state to a highly fluorescent state is also observed (inset of Figures 2A–C), signifying the molecular rotor nature of the compounds. We have used the Forster–Hoffmann equation to quantify the relation between fluorescence quantum yield and the viscosity of the solution (Wu et al., 2013), i.e., ϕ f = C η x , taking logarithm on both sides, we can get: log ( ϕ f ) = log ⁡ C + x ⁡ log ⁡ η and for a lifetime log ( τ f ) = log ⁡ C ' + x ⁡ log ⁡ η , where ϕ f is the fluorescence quantum yield, C and C' are the constants, C ' = C × ( k r − 1 ) , k r = radiative rate constant, and η = the viscosity of the medium. The restriction of molecular motion on moving from low to high viscous environment suppresses the non-radiative pathways and results in fluorescence enhancement. The double logarithmic plot of log (Intensity) measured at emission maxima with log (viscosity) of all these compounds well fitted with the equation discussed earlier, as shown in Figures 2D–F and Supplementary Figure S26. The fluorescence lifetime of these molecular rotors is also enhanced by increasing the medium viscosity. The fluorescence lifetime of these molecules in water is quite short and increases gradually upon increasing the viscosity of the medium (Figures 3A–C). The lifetime of DCAJ in glycerol is the shortest among all these rotors. On moving from DCAJ to JIND, the fluorescence lifetime increases upon substitution with the bulkier group, as evident from their lifetime in glycerol. The lifetime increases linearly with the viscosity of the medium (Figures 3D–F) as envisioned. The increased stiffness of the slope from this linear dependence for DCAJ to JIND-Mor confirms their improved sensitivity upon increases in bulkiness.

Furthermore, JIND-Mor was selected as a potential molecular rotor for monitoring viscosity changes in a living system. To start with, we have investigated the cellular toxicity of JIND-Mor in noncancerous cells (BHK-21) using a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide assay. The IC₅₀ value is more than 5 µM, and more than 70% of cells were viable even after 24 h (Supplementary Figure S27). After that, we have investigated the localization of JIND-Mor in the cellular compartments via live-cell fluorescence imaging using confocal laser scanning microscopy. To our pleasure, JIND-Mor selectively localizes in the lysosomal compartments of cells. The selectivity was assessed using a commercially available lysotracker green dye, known to localize selectively in the lysosome (Figures 4A–E, top panel, Supplementary Figure S28). The high Pearson-correlation coefficient of 0.87 establishes the selectivity. Furthermore, we investigated the localization with other organelle trackers such as mitochondrion and ER (middle and bottom panel of Figure 4). Contrarily, the Pearson coefficients obtained for ER (0.20) and mitochondrion (0.32) were quite low (Figures 4J,O). A comparison of the previously reported molecular rotor for lysosomal viscosity determination is provided in Supplementary Table S2.

After establishing the lysosome-specific localization of JIND-Mor, we monitored the lysosomal viscosity change in glioblastoma (GBM), a known fast-growing and aggressive cancer cell (Taylor et al., 2019). It is established that cellular viscosity has strong influences on their progression, invasion, and morphological stability (Streitberger et al., 2020). Therefore, the determination of the lysosomal viscosity of GBM can provide useful information for its diagnosis and treatment (Perini et al., 2020). To assess the temporal changes of lysosomal viscosity in human GBM cells (U87-MG) using JIND-Mor, we have used Dexa as the stimulation reagent. Dexa acts as a lysosomal membrane stabilizer and an inhibitor of lysosomal enzymatic release, which causes an increase in the lysosomal viscosity (Yang et al., 2020). Therefore, the dynamic fluorescence change of the lysosomal compartment of U87-MG cells was measured after treating with JIND-Mor for 15 min and then stimulated with 50 µM of Dexa (Figure 5A). The fluorescence enhancement clearly indicated an increase in lysosomal viscosity upon treatment with Dexa without interrupting the lysosomal structural integrity (see Figure 5B). Furthermore, we have quantified the time-dependent fluorescence intensity of lysosomes (Figure 5C). The results discussed earlier indicate that JIND-Mor is potent for selective lysosomal localization and determination of viscosity change in live conditions.

To further validate the applicability of JIND-Mor for imaging lysosome-related organelle stress in C. elegans, they were grown to young adult stage in nematode growth medium with 10-µM JIND-Mor for 60 h at 20°C. The brighter green fluorescence under osmotic stress-induced conditions (Figures 6A–F) clearly indicates the increment in gut granules viscosity. These results revealed the potential applicability of JIND-Mor for viscosity sensing in living C. elegans.