LUCS (Light-Up Cell System), a universal high throughput assay for homeostasis evaluation in live cells

Observations of fluorescent cyanine dye behavior under illumination at 500 nm lead to a novel concept in cell biology allowing the development of a new live cell assay called LUCS, for Light-Up Cell System, measuring homeostasis in live cells. Optimization of the LUCS process resulted in a standardized, straightforward and high throughput assay with applications in toxicity assessment. The mechanisms of the LUCS process were investigated. Electron Paramagnetic Resonance experiments showed that the singlet oxygen and hydroxyl radical are involved downstream of the light effect, presumably leading to deleterious oxidative stress that massively opens access of the dye to its intracellular target. Reversible modulation of LUCS by both verapamil and proton availability indicated that plasma membrane proton/cation antiporters, possibly of the MATE drug efflux transport family, are involved. A mechanistic model is presented. Our data show that intracellular oxidation can be controlled by tuning light energy, opening applications in regulatory purposes, anti-oxidant research, chemotherapy efficacy and dynamic phototherapy strategies.

3) Experimental arguments that allowed to discarding a "dequenching" process as being involved in the fluorescence increase observed in LUCS assay. Based on fluorescence theory, thiazole orange (TO) fluorescence could be self quenched by the proximity of the molecules acting as intercalators within DNA and/or RNA. According to this hypothesis, the progressive light-induced bleaching of the fluorophores could restore fluorescence by dequenching the remaining intact fluorophores. This assumption was discarded by fluorescence life time measurements (figure S3, table S2).
4) Experimental arguments that allowed to discarding a modification of dye quantum yield as being involved in the fluorescence increase observed in LUCS assay. In a double labelling experiment where SYTO13 and SYTO62 dyes were sequentially added to cells, LUCS was observed in conditions where SYTO62 had not been in contact with light (i.e. after light induction had ended) ( figure S4) showing that SYTO62 fluorescence increase can not be explained by modifications of its photophysical properties.

5)
A list of the different live models on which LUCS has been observed (table S3) Figure S1. LUCS assay dose-responses of the 53 compounds used for the regression analysis study (figure 2). HepG2 cells cultured in 96-well plates were treated with compounds (53 out of the 67 substances of the ACuteTox European Program databank, according to commercial availability) for 24h. Cells were then treated with SYTO13 at 4µM final concentration for 30 min followed by a fluorescence measurement (F pre ) at 520nm. Plates were then exposed to a LED based light source (0.24 J/cm 2 centered at 470nm) followed by a second fluorescence reading (F post ). Dots represent R=F post /F pre values normalized to untreated (control) cells.
Corresponding EC 50 s are given in table S1. All experiments were performed at least 3 times and in each case in triplicate. Dots represent means of the 3 experiments (error bars= SD values) Table S1. EC 50 of the 56 chemical tested on HepG2 cells with the LUCS assay.

5-fluorouracil 2,08E-05 -4,68
Acétaminophen     Table S2   Table S2. Thiazole orange lifetime values (t) before and after light application (440 nm) in live HepG2 cells. According to the fluorescence theory, a dequenching process (increase of photon emission following the loss of resonance energy transfer or other quenching processes) should increase fluorescence lifetimes values (t). We show here that t values tend to decrease after light application suggesting that TO selfdequenching process is not involved in LUCS assay. Figure S4. Double labelling experiment. A. HepG2 cells were treated for 1h with SYTO13 at 4µM final concentration then exposed (or not) to a LED light source (470nm; 0.24 J/cm 2 ). SYTO62 was (or not) added at a final concentration of 12µM and then, fluorescence of SYTO13 and SYTO62 was read at 520nm or 680nm respectively. B. Fluorescence reading at SYTO13 wavelength (520 nm). As expected, in cells labeled with SYTO13 only, light exposure at 470nm triggered an increase of fluorescence emission at 520nm as compared to unexposed cells. After SYTO62 addition, fluorescence of both exposed and unexposed cells were slightly reduced, possibly due to a site competition between the two SYTO dyes on nucleic acids. C. Fluorescence reading at SYTO16 wavelength (680 nm). Without light exposure the presence of SYTO13 does not affect SYTO62 fluorescence. However, light exposure at SYTO13 wavelength triggered SYTO62 fluorescence increase. As the SYTO62 dye was added after light exposure, its fluorescence increase cannot be explained by a photoinduced process. This demonstrates that the fluorescence increase associated to LUCS process is not due to modification of intrinsic photophysical dye properties but rather to a massive entry of the dye triggered by a photo-induced process. Figure S4 Table S3. The 16 prokaryotic and eukaryotic unicellular organisms or cell lines on which LUCS process was evaluated with success