Self-Assembled Lanthanide Antenna Glutathione Sensor for the Study of Immune Cells

The small molecule 8-methoxy-2-oxo-1,2,4,5-tetrahydrocyclopenta[de]quinoline-3-carboxylic acid (2b) behaves as a reactive non-fluorescent Michael acceptor, which after reaction with thiols becomes fluorescent, and an efficient Eu3+ antenna, after self-assembling with this cation in water. This behavior makes 2b a highly selective GSH biosensor, which has demonstrated high potential for studies in murine and human cells of the immune system (CD4+ T, CD8+ T, and B cells) using flow cytometry. GSH can be monitored by the fluorescence of the product of addition to 2b (445 nm) or by the luminescence of Eu3+ (592 nm). 2b was able to capture baseline differences in GSH intracellular levels among murine and human CD4+ T, CD8+ T, and B cells. We also successfully used 2b to monitor intracellular changes in GSH associated with the metabolic variations governing the induction of CD4+ naïve T cells into regulatory T cells (TREG).


I. General methods S1
II. Experimental synthetic methods S2

IV. Spectroscopy Methods S7
V. Kinetic study S13 VI. Binding model and fitting parameters S14 VII. Cell Viability S15 VIII. Time-resolved luminescence assays S16

IX. Mice and Human subjects S16
X. Lymphocyte isolation from the spleen and PBMCs S17 XI. Flow cytometry, surface staining and biothiol levels S17

I. General Methods
All reagents were of commercial quality. Solvents were dried and purified by standard methods. Analytical TLC was performed on aluminum sheets coated with a 0.2 mm layer of silica gel 60 F254. Silica gel 60 (230-400 mesh) was used for flash chromatography. HPLC-MS was performed on a Sunfire C18 (4.6×50 mm, 3.5 μm) column at 30°C, with a flow rate of 1 mL/min and gradient of 0.1% of formic acid in CH3CN (solvent A) in 0.1% of formic acid in H2O (solvent B) was used as mobile phase. Electrospray in positive mode was used for ionization. NMR spectra were recorded using Varian Inova or Mercury 400, and Varian Unity 500 spectrometers. The NMR spectra assignments were based on COSY, HSQC, and HMBC spectra. High resolution mass spectra (HRMS) were recorded on an Agilent 6520 Q-TOF instrument with an ESI source.

III. Computational Methods
The geometry of the molecules in the singlet ground state (S0) has been optimized with the B3LYP functional 2,3 and the 6-31+G(d,p) basis set. 4 Frequency calculations at the same computational level have been performed to confirm that the geometries obtained correspond to energetic minima. These geometries have been used as starting point for the optimization of the first excited single (S1) and triplet (T1) states. For the latter, TD-DFT 5 has been used. All the calculations have been performed with the Gaussian-16 program.
The aromaticity of the systems has been evaluated with the harmonic oscillator model of aromaticity (HOMA) parameter 6,7 as expressed in the following equation (S1).
The values used of Ropt and α were 1.388 and 257.7 for C-C bonds and 1.334 and 93.52 for C-N bonds, respectively.

Figure S1
. Tautomeric/rotameric structures considered in the calculations of the minimum energy structures of the S0, S1 and T1 states for the free carboxylic acids 1b and 2b.

a. Photophysical properties
Excitation and emission spectra of compounds were determined for 12 μM solutions in H2O or CH3CN. The spectra were recorded between 300 and 690 nm (0.5 nm increments and 0.1 s integration time) with excitation set at the appropriate excitation wavelength. Slit widths were set to 5 nm for excitation and to 5 nm for emission. All the spectra were corrected for background fluorescence by subtracting a blank scan of the solvent solution and spectrally corrected using certified fluorescence standards.
Fluorescence quantum yield determination. Fluorescence quantum yields (F) of quinolin-2(1H)-one derivatives 1b and 2b were determined in solvents of different polarity and calculated using quinine sulfate dihydrate (in 0.1M H2SO4) as reference. 8 Concentration of the sample and the reference was set to assure that the absorbance was less than 0.1 at identical excitation wavelengths. The following equation (S2) was used to calculate the quantum yield: (S2) where x and r denote the sample and standard, respectively, A is the absorption at the excitation wavelength, I is the integrated fluorescence intensity, and n is the refractive index of the solvent.

V. Kinetic study
In the kinetic study, determined the rate law from the experimental was determined data using the differential method.
Since we worked with constant concentrations of the sensor, we can focus on the reaction order with respect to GSH, using the following logarithmic equation (S3): One way to obtain the data is to plot the initial rate of the reaction (v0) with different initial [GSH]0. This technique is known as the initial-rate method. A plot of log v0 against log [GSH]0 would be a line with a slope equal to the reaction order with respect to GSH ( Figure S9). The observed rate constant, which would include components depending on the concentration of the sensor, can be determined by the intercept. To obtain the data, we used different concentrations of GSH (5, 25, 50 and 500 M) in the presence of EuCl3

VII. Cell Viability
We incubated mouse splenocytes (Figures S10A and S10B) or human PBMCs ( Figure S10C) with either compound 2b (Figures S1A and S1C) or EuCl3 ( Figure S10B) for 1 hour in PBS at 37 °C. We then washed the cells and stained them with fluorescent Fixable viability dye eFluor 780 (eBioscience, USA) as detailed in the manufacturer's protocol, and monitoring on a 3-laser Canto flow cytometer (BD bioscience) and analyzed the cells using FCS express software.

VIII. Time-resolved luminescence assays
Luminescent assays were performed in 96 well plates in 200 µL 50 mM potassium phosphate (50 mM KH2PO4/K2PO4, pH 7.4) 150 mM NaCl, using a SpectraMax M2 spectrofluorimeter. Splenocytes were cultured in PBS with the sensor 2b (25 μM) and EuCl3 (250 µM) added at the same time, and then excited at 320 nm. The emission signal was acquired at 620 nm unless stated otherwise (measurement start 50 s; measurement end 850 s; 20 flashes per read). Data analysis was performed using Prism software.
Experiments were performed in triplicate, and error bars indicate standard deviation from average. Figure S11. Time resolved-fluorescence kinetics of europium luminescence overtime (λex= 320 nm, λem=620 nm, 14 h) in mouse splenocytes.

IX. Mice and human samples
Male and female C57BL/6-Foxp3-YFP and Wild type B6 mice were purchased from The Jackson Laboratory (stocks #016959 and #000651, respectively) and bred at mouse facilities of the Icahn School of For experiments in human cells we used peripheral blood mononuclear cells (PBMCs) from buffy coats obtained from Fully anonymous deidentified donors to the New York Blood Bank of the USA, and thus were determined by the institution not to constitute human subjects research.