Selective Detection of Cu+ Ions in Live Cells via Fluorescence Lifetime Imaging Microscopy

Abstract Copper is an essential trace element in living organisms with its levels and localisation being carefully managed by the cellular machinery. However, if misregulated, deficiency or excess of copper ions can lead to several diseases. Therefore, it is important to have reliable methods to detect, monitor and visualise this metal in cells. Herein we report a new optical probe based on BODIPY, which shows a switch‐on in its fluorescence intensity upon binding to copper(I), but not in the presence of high concentration of other physiologically relevant metal ions. More interestingly, binding to copper(I) leads to significant changes in the fluorescence lifetime of the new probe, which can be used to visualize copper(I) pools in lysosomes of live cells via fluorescence lifetime imaging microscopy (FLIM).


Supporting Information
All chemical reagents and solvents for synthesis were purchased from commercial suppliers and were used without further purification. All moisture or oxygen sensitive reactions were carried out under a nitrogen atmosphere. The composition of solvent mixtures is given by volume ratio (v/v). Analytical and preparative thin layer chromatography was performed using Merck 60 F254 silica gel with 0.25 mm and 2 mm thickness, respectively. Flash chromatography (FC) was performed using Merck Kieselgel 60 at rt under a positive pressure. 1 H NMR spectra were collected in CDCl 3 or DMSO-d 6 (Cambridge Isotope Laboratories, Cambridge, MA) at 25 °C on a Bruker AV-300 or AV400 spectrometers. Chemical shifts (δ) are given in ppm and coupling constants in Hz. Notation for the 1 H-NMR spectral splitting patterns include singlet (s), doublet (d), triplet (t), quartet (q), quintet (qui), broad (br) and multiplet/overlapping peaks (m). Matrix-assisted laser desorption/ionization MALDI analyses were carried out with Thermo Scientific CHCA MALDI Matrix. All compounds shown in the reaction schemes below (except for 5 and FLCS1 which are new) were synthesized following previously reported procedures. [1], [2] Synthesis and characterization
After 15 minutes the solvent was evaporated under reduced pressure in a Schlenk line and the residue was redissolved in THF (8 ml).
2,3-Dichloro-5,6-dicyano-p-benzoquinone, DDQ (276 mg, mmol, 2 eq.) was separately dissolved in THF (1 ml) and slowly added with a syringe through a septum into the Schlenk flask. The reaction was stopped after 20 minutes by adding water; the crude product was extracted with CH 2 Cl 2 . The organic layer was washed with brine and water, dried over Na 2 SO 4 and evaporated. The mixture was partially purified via flash column chromatography (eluent: n-Hexane/CHCl 3 = 40/60; solid phase: alumina) to remove the remaining DDQ from the reaction mixture.
The remaining solid containing crude compound 5 was dissolved in 1,2-dichloroethane under nitrogen; N, N-diisopropylethylamine, DIEA (0.83 ml, 0.61 g, 4.7 mmol, 7.8 eq.) was added and the reaction mixture heated-up to 60°C. Then BF 3 ·Et 2 O (0.83 ml, 0.95 g, 6.7 mmol, 11 eq.) was added which led to an instant colour change from green to very dark blue. After 25 min toluene was added to the reaction mixture and a liquid-liquid extraction with water was carried out. The organic phase was again washed with brine and water and the solvent was evaporated. The final purification steps consisted of passing the product through a silica flash chromatography column (eluent: n-Hexane/EA = 100/00 to 80/20) followed by a preparative thin layer chromatography (eluent: n-Hexane/EA = 66/33); for both purification steps the silica used was firs treated with 0.1% triethylamine in hexane. The product obtained from column chromatography, was dissolved in acetone and deposited on the TLC plate with a pipette. The product band was scratched off from the TLC and extracted with MeOH; the solvent was removed under reduced pressure to yield pure FLCS1, which is a dark blue compound (21.7 mg, 3.3 % yield). A sample of the compound was used to prepare a 10 mM stock solution in DMSO and stored at -18°C. For all further spectroscopic measurements, this sample was diluted to 1 mM and 20 μM which served as a stock solution for the photophysical characterisation and for the cell experiments respectively.     Fluorescence quantum yields were determined by using to Alexa 647 in water as a standard (Φ f = 0.33).

Preparation of metal ion solutions for spectroscopic experiments
For the spectroscopic fluorescence and absorbance experiments MeOH was degassed by purging with N 2 for 2 hours before the experiment to avoid oxidation of the copper(I) salt. During the experiment it was kept under N 2 atmosphere using a nitrogen balloon.

Determination of Dissociation Constant
To determine the dissociation constant (K d ) a methanolic solution of FLCS1 (1 μM) containing thiourea (400 μM) as competitive ligand. To calculate the available copper the stability constants for thiourea binding were used taken from the literature: β 12 = 2.0 × 10 12 , β 13 = 2.0 × 10 14 , β 14 = 3.4 × 10 15 . [3] For this titration the copper(I) stock solution was further diluted to 1 μM (see values in caption of Fig. S4b). The fluorescence spectra was collected by exciting at 610 nm and the emission was integrated from 620 to 750 nm. The binding affinity was calculated following literature procedures [4,5] using the Benesi-Hildebrand plot (see Fig. S4c) with the equation:

Cu-GTSM preparation
The For FLIM imaging the coverslide was mounted in a microscope chamber heated by a circulating thermostat (Lauda GmbH, E200) with feedback control of temperature and 0.2 °C precision. Cells were measured by FLIM, as described below, in the 'before' and 'after' treatment states.

Bafilomycin A1 treatment experiment
SH-SY5Y cells were grown in DMEM with 10% FBS with and added penicillin and streptomycin at 37°C. The healthy cells were plated on LabTek II coverslides at a seeding density of 20.000 cells per well and let grow for 48h more hours in an incubator. Then the media was changed to media containing 0.2% lipofectamine and 60 nM FLCS1 and the cells were incubated for another 24h.
After that the media was changed back to grow media and the after a calming period of 1 hour the cells were imaged in their "before For the control comparisons the control cells were treated with DMSO (0.01%) and also incubated for 24 hours. After the incubation time images were taken with the same settings as for the bafilomycin A1 treated cells.

pH/viscosity/H2O-effect experiments
The TCSPC pH measurements were carried out in a methanolic solution of FLCS1 (1 μM). The pH was calibrated using diluted HCl and NaOH solutions. For the viscosity experiment the probe was dissolved in different ratios of MeOH and glycerine and for the water-effect on the lifetime the solvent ratios were altered between water and MeOH. All the measurements were carried out with excess copper(I) (3 μM).       h) and i) with j). Therefore, the reduce the uncertainty of the fitting parameters, the decision was taken to fix τ1 and τ2 values, see Figure S12 for the results. We note that the goodness of fit χ 2 is not affected by fixing the lifetime values. Also, our conclusions that the amplitude of the long component, 2, is increased upon treatment with Cu-GTSM, is valid, irrespective of the fitting method, Figure S9 or Figure S12.      Figure S12. * p > 0.5 shows no statistical significance ** p < 0.0005 shows statistical significance with student t-test. DMSO treatment (right) showing statistical significance for observed changes with * p > 0.0005 and ** p < 0.00005 with student t-test.