Cancer-Specific hNQO1-Responsive Biocompatible Naphthalimides Providing a Rapid Fluorescent Turn-On with an Enhanced Enzyme Affinity

Human NAD(P)H:quinone oxidoreductase 1 (hNQO1) is overexpressed in cancer cells and associated with the drug resistance factor of cancer. The objective of this work is the development of fluorescent probes for the efficient detection of hNQO1 activity in cancer cells, which can be employed for the cancer diagnosis and therapeutic agent development. Herein, we report naphthalimide-based fluorescent probes 1 and 2 that can detect hNQO1. For hNQO1 activity, the probes showed a significant fluorescence increase at 540 nm. In addition, probe 1, the naphthalimide containing a triphenylphosphonium salt, showed an enhanced enzyme efficiency and rapid detection under a physiological condition. The detection ability of probe 1 was superior to that of other previously reported probes. Moreover, probe 1 was less cytotoxic during the cancer cell imaging and readily provided a strong fluorescence in hNQO1-overexpressed cancer cells (A549). We proposed that probe 1 can be used to detect hNQO1 expression in live cells and it will be applied to develop the diagnosis and customized treatment of hNQO1-related disease.


UV/Vis absorption and fluorescence spectroscopic methods
Stock solutions of synthetic probes were prepared in DMSO. Stock solutions of metal chloride salts and thiols were prepared in deionized water. Stock solutions of TBA salts anions and [Cu(CH3CN)4]PF6 for Cu + ion were prepared in CH3CN. Stock solutions of reactive oxygen species (ROS) were prepared by using literature procedures. 1 All spectra were recorded in PBS/DMSO solution (v/v = 99:1, 100 mM, pH = 7.4) containing BSA (0.007%) and KCl (0.1 M) at 37 ℃. Excitation was carried out at 410 nm with all excitation slit widths 5 nm.

Enzyme titration
Stock solutions of hNQO1 and NADH were prepared in deionized water. Time dependent of fluorescence response of 1 (5 μM) or 2 (5 μM) to various concentrations of hNQO1 in the presence of NADH (100 μM) were determined at 540 nm, with a time interval of 0.1 s for 5 min. All measurements were recorded in PBS/DMSO solution (v/v = 99:1, 100 mM, pH = 7.4) containing BSA (0.007%) and KCl (0.1 M) in 37 ℃. λex = 410 nm. The fluorescence units were converted to concentration by relating the signal increase to a fluorescence signal derived from a known concentration of compound 3 or 4. From this, velocity (μM s -1 ) was plotted as a function of [hNQO1] (μM/mL).

Enzyme kinetics
Time dependences of fluorescence response of various concentrations of 1 (0-0.06 μM) or 2 (0-0.5 μM) to hNQO1 (0.5 μg/mL) in the presence of NADH (100 μM) were determined at 560 nm, with a time interval of 0.1 s for 5 min. All measurements were recorded in PBS/DMSO solution (v/v = 99:1, 100 mM, pH = 7.4) containing BSA (0.007%) and KCl (0.1 M). λex = 410 nm. The fluorescence units were converted to concentration by relating the signal increase to a fluorescence signal derived from a known concentration of compounds 3 and 4. Velocity (μmol min -1 mghNQO1 -1 ) was plotted as a function of [probes] (μM) to obtain Km and Vmax values from nonlinear least-squared analysis for Michaelis-Menten kinetics (OriginPro 8.0). From this, kcat and kcat/Km values were calculated from these values.

Cell viability assay
A549 cells (2×10 4 /well) and H596 cells (0.9×10 4 /well) were seeded on 96 well microplate (SPL Life Science) and incubated for 1 h, 24 h and 36 h, respectively. After the cells were stabilized, 1% DMSO was used as a control and probes 1 or 2 was treated for indicated time points. To analyze the viability of cells in the presence and absence of probes 1 and 2, the CytoTox96® Non-Radioactive Cytotoxicity Assay Kit (Promega, Madison, WI, USA) was used, following the manufacturer's protocols. Power Wave XS Microplate Reader (Biotek Instruments Inc., Winooski, VT, USA) was used to measure the fluorescence level. The wavelength was set at 492 nm. Cell viability assays were performed in triplicates, and the viability was expressed as a percentage (%) of measured absorbance, relative to the control cells.

Flow cytometry analysis
Flow cytometry was used for quantitative analysis of cellular internalization of probes 1 and 2. A549 cells were seeded (1 × 10 5 /mL, total volume 4 mL) in 60 mm culture dish (SPL Life Science, Seoul, Republic of Korea) and incubated for 24 h. Cells were treated with 5 µM of probes 1 or 2 and incubated for 10 min, 30 min, and 1 h, respectively. Non-treated cells were used as the control for experiments. All cells were detached from the dish using 0.05 % trypsin-EDTA solution (Welgene, Daegu, Republic of Korea) and washed with PBS for two times before flow cytometry analysis. Fluorescence-activated cell sorting flow cytometry system (FACS Calibur BD Flow Cytometer, BD Biosciences, California, USA) was used for cells analysis.

Colocalization imaging
In order to verify the hNQO1 visualization abilities of probes 1 and 2, the cellular imaging and colocalization images were taken on hNQO1-positive A549 cells. The cells were seeded (2.5 × 10 4 /ml, total volume 2 ml) in glass-bottomed confocal dishes (SPL Life Science) and incubated in the same conditions as mentioned above. The cells were treated with probe 1 (5 µM) or 2 (5 µM) for 1 h and stained with suborganelles trackers. MitoView633 (Biotium Inc., Fremont, CA, USA) was used for mitochondria staining, LysoView633 (Biotium Inc., Fremont, CA, USA) for lysosome and ER-tracker Red (Invitrogen Inc., Carlsbad, CA, USA) for endoplasmic reticulum (ER). The cells were incubated with 500 nM of MitoView633 or LysoView633 for 30 min, or with 2 μΜ ER-tracker Red for 1 h. The fluorescence channel was excited at 555 nm and the emission was recorded at 600-700 nm with a band-pass filter. The dishes were washed by PBS three times to remove free dyes and the fluorescence was imaged. The colocalization images were analyzed by the ZEN software (blue sedition version). And the level of CTCF (Corrected total cell fluorescence) was calculated using the following formula, with the data from confocal microscopy images, using ImageJ software: CTCF = Integrated Density -(Area of selected cell × Mean fluorescence of background readings). Magnification of each image is 40×. Figure S1. The normalized absorption spectra of probe 1, compound 3 and NADH.      Figure S8. (a) Enzyme titration of hNQO1 (0-0.5 μg/mL) with probe 2. (b) Enzyme kinetics of hNQO1 (0.5 μg/mL) with probe 2 (mean ± sd, n = 3). The kinetic parameters were estimated to be Km = 0.48 ± 0.13 μM, kcat = 0.20 ± 0.009 s -1 , Vmax = 0.38 ± 0.02 μmol min -1 mg hNQO1 - Figure S9. The pH effect on the enzymatic reactions of 2 (5 μM, respectively). All data were obtained by measuring the initial rates of fluorescence change at 540 nm in the absence (black) and presence (red) of hNQO1 (5 μg/ml) and NADH (100 μM) in different pH buffer solutions. λex = 410 nm.