A fast-responsive fluorescent turn-on probe for nitroreductase imaging in living cells

Nitroreductase (NTR) may be more active under the environment of hypoxic conditions, which are the distinctive features of the multiphase solid tumor. It is of great significance to effectively detect and monitor NTR in the living cells for the diagnosis of hypoxia in a tumor. Here, we synthesized a novel turn-on fluorescent probe NTR-NO2 based on a fused four-ring quinoxaline skeleton for NTR detection. The highly efficient probe can be easily synthesized. The probe NTR-NO2 showed satisfactory sensitivity and selectivity to NTR. Upon incubation with NTR, NTR-NO2 could successively undergo a nitro reduction reaction and then generate NTR-NH2 along with significant fluorescence enhancement (30 folds). Moreover, the fluorescent dye NTR-NH2 exhibits a large Stokes shift (Δλ = 111 nm) due to the intramolecular charge transfer (ICT) process. As a result, NTR-NO2 displayed a wide linear range (0–4.5 μg mL−1) and low detection limit (LOD = 58 ng mL−1) after responding to NTR. In addition, this probe was adopted for the detection of endogenous NTR within hypoxic HeLa cells.


Introduction
It is found that the average oxygen concentration is in the range of 4% to 0% in the microenvironment of solid tumors. 1 Hypoxia in tumor growth is caused by the innite growth of tumor cells, which makes blood supply insufficient and causes vascular dysplasia. [2][3][4][5] Clinical studies found that the tumor hypoxic status is tightly associated with tumor diagnosis and therapy, which exerts a vital part in the metastasis, invasion, as well as drug resistance of the tumor. 6,7 NTR is a avinase with nicotinamide adenine dinucleotide (NADH) as a coenzyme, which can catalyze the reduction of nitroaromatic compounds to the corresponding amino compounds. [8][9][10] It has been conrmed that hypoxia can result in the overexpression of reductase, such as nitroreductase, DT-diaphorase, and azo reductase. [11][12][13][14][15] The activity of NTR can be used as an important biomarker to reect the hypoxic conditions in tumor cells and tissues. 16,17 Traditional hypoxia detection methods include magnetic resonance imaging, p O 2 electrodes and positron emission tomography/computed tomography (PET/CT), and it generally requires expensive equipment and complex sample pretreatment. [18][19][20][21][22][23][24][25] Moreover, the integrity of samples and achieving higher spatial and temporal resolution cannot be guaranteed. Furthermore, uorescence imaging has attracted widespread attention to monitoring biological molecules owing to its real-time imaging, high sensitivity and selectivity, and noninvasiveness. Numerous uorescent probes have been designed according to the reduction reaction catalyzed by NTR to detect the hypoxia status for over the last 10 years (Table  S1 †). [26][27][28][29][30][31][32][33] However, many uorescence NTR probes exhibit short Stokes shi (<100 nm) and slight signal changes aer responding to NTR, which is easily disturbed by the background uorescent signal. Thus, there is an urgent requirement to develop simple uorescent probes with obvious signal changes aer responding to NTR.
Here, we successfully synthesized a turn-on uorescent probe NTR-NO 2 based on a fused four-ring quinoxaline skeleton reported by our previous work for NTR detection (Scheme 1). 34 By introducing nitro as the response site, the probe NTR-NO 2 displayed weak emission due to the well-known quenching effect of a nitro group. The probe shows obvious uorescence turn-on signal (30-fold) aer the nitro to amine reduction reaction catalyzed by NTR with the aid of NADH as an electron source. Furthermore, the activated version of the Scheme 1 The proposed detection principle of probe NTR-NO 2 towards NTR.
probe NTR-NH 2 possessed strong uorescent emission and large Stokes shi (Dl ¼ 111 nm) due to ICT it showed low toxicity to cells and high photo-stability, indicating that NTR-NH 2 is potentially employable as a uorophore in the probe. 35 The probe NTR-NO 2 exhibited fast response and high selectivity to NTR. Good linearity between the uorescent intensity and the concentration of NTR ranging from 0 to 4.5 mg mL À1 was observed. The limit of detection (LOD) was calculated to be 58 ng mL À1 . Moreover, the probe NTR-NO 2 was successfully applied in the monitoring of NTR activity within the HeLa cell under hypoxic conditions.
The 1 H NMR and 13 C NMR spectra were measured on the Qone-WNMR-I-AS400 instrument with DMSO-d6 as the solvent. The Agilent 1100 LC/DAD/MSD was used for high-resolution mass spectrometry (HRMS). Absorption spectra were recorded using a UV3600 UV-VIS-NIR spectrometer. Fluorescence emission spectra were recorded using a FluoroMax-4 uorescence spectrophotometer. The pH values were measured using the Ray Magnetic pHS-3E pH meter. In addition, FV1000 confocal microscopy was used to perform cell uorescence imaging.

Spectroscopic characteristics and NTR optical response
Unless specially stated, the absorption and uorescence spectra of NTR-NO 2 (10 mM) at different concentrations of NTR in PBS solution (20% DMSO, pH ¼ 7.4) were recorded at 37 C in the presence or absence of NADH (500 mM) for 30 min. For all uorescence spectra measurements, the excitation wavelength was 430 nm, and the slit widths were 2 nm/2 nm.

Synthesis of NTR-NO 2
Compound 1 was synthesized following a previously reported method. 36-38 4-(Triuoromethyl)phenylhydrazine hydrochloride (638 mg, 3 mmol) and 3-methyl-2-butanone (387 mg, 4.5 mmol) were dissolved in 5 mL acetic acid while stirring. Then, the resulting mixture was heated to 90 C and stirred for 12 h. Aer the reaction was complete, the hot solution was cooled to room temperature. Ethyl acetate was used for the extraction of products from the mixture (3 times). The organic layers were extracted in a vacuum to afford the crude product, which was directly used in subsequent reactions.
Compound 2 was synthesized by a previously reported method. 37,38 Compound 1 and iodomethane (850 mg, 6 mmol) were dissolved in THF (10 mL). The solution was reuxed at 70 C for 12 h and a large amount of insoluble solid precipitated from the solution was obtained. Then, the reaction mixture was ltered and the insoluble solid was washed with cold THF. Aer drying under vacuum, light yellow solid was obtained with a 55% yield.
NTR-NO 2 was synthesized following a previously reported method. 34 Compound 2, 2,4-nitrobenzene-1,2-diamine (850 mg, 6 mmol), and iodine (1 g, 4.5 mmol) were dissolved in DMSO (5 mL). Then, the resulting mixture was heated to 100 C and stirred for 1 h. Aer the reaction was complete, the mixture was cooled to room temperature and added to saturated Na 2 S 2 O 3 solution. Then, the mixture was extracted with ethyl acetate (3 times) and evaporated in vacuum. The residue was puried using column chromatography on silica gel with petroleum ether : ethyl acetate (400 : 1, v/v) as the eluent to obtain the products.

Cytotoxicity assays and hypoxic imaging
HeLa cells were purchased from KeyGen Biotech (Nanjing, China). The cells were cultivated within MEM medium supplemented with 10% fetal bovine serum (FBS), 0.08 mg mL À1 streptomycin and 80 U mL À1 penicillin and incubated at 37 C containing 5% CO 2 /95% O 2 gas. The HeLa cell line was then inoculated in the 96-well plates at 5 Â 10 4 per well. Thereaer, the cultures were removed and the NTR-NO 2 probe at different contents (0-50 mM) was added to incubate with the cells for 24 h. Aer adding MTT (20 mL, 5 g L À1 ), cells were subjected to another 4 h of incubation. Later, 100 mL of DMSO was added to replace the original medium and to dissolve formazan crystals. Then, the multifunctional microplate reader was used to measure absorbance (OD) value at 570 nm, and cell viability was presented as follows: 39 The cell viability ð%Þ ¼ HeLa cells (100 mL) were inoculated onto glass coverslips for adherence. Then, cells were further incubated under 20% O 2 (normoxia) and 1% O 2 (hypoxia) for 12 h, followed by further incubation with the probe NTR-NO 2 for another 30 min under respective conditions. Aer washing with PBS (pH ¼ 7.4) three times, cells were imaged using confocal uorescence on a Olympus FV1000 confocal uorescence microscope.

Results and discussion
Synthesis and sensing mechanism of probe NTR-NO2 We synthesized NTR-NO 2 as a novel off-on uorescent probe for NTR detection based on a fused four-ring quinoxaline skeleton reported in our previous work. 34 The -CF 3 and -NO 2 groups were attached to the dihydroquinoline unit and quinoxaline unit in this skeleton, respectively, where the nitro group was selected as the recognition site for NTR and the uorescence quenching group due to its strong electron attraction. NTR-NO 2 was synthesized as shown in Scheme 2. The chemical structure of NTR-NO 2 was identied using NMR and HRMS (Fig. S1-S3 †). In this probe, fused cyclic derivatives of quinoxaline structure were employed as e-withdrawing parts for uorescent push-pull systems due to its highly p-decient feature; 34 the probe NTR-NO 2 exhibited weak emission due to the well-known quenching effect of a nitro group. Subsequently, the nitro group was reduced into the corresponding amino group aer the reaction was catalyzed by NTR in the presence of NADH, and it resulted in the formation of the "push-pull" structure of NTR-NH 2 . NTR-NH 2 was highly emissive upon photoexcitation due to the ICT process reported by our group. 35 As a result, the obvious turn-on signal was measured. For verifying the above mechanism (Scheme 1), HRMS analysis (Fig. S4 †) along with absorption was performed to test the reaction solution of the probe NTR-NO 2 with NTR (Fig. 1a). As suggested by HRMS for the reaction solution, the main peak was observed at m/z ¼ 359.1479 [M + H] + , which was characterized as NTR-NH 2 . Changes in spectral characteristics of NTR-NO 2 solution, when NTR was added in the presence of NADH showed that the reaction solution had a similar maximum absorption wavelength to that of NTR-NH 2 . All these results clearly demonstrate that the reaction of NTR-NO 2 with NTR in the presence of NADH causes the reduction of the nitro group.

Photophysical performances of the NTR-NO2 probe
We investigated optical properties of the NTR-NO 2 probe with or without NTR at 37 C using the PBS buffer (pH ¼ 7.4, 20% DMSO) in the presence of 500 mM of NADH. The probe showed a sensitive spectral response toward NTR. It can be seen from Fig. 1a that the free-probe (50 mM) showed a maximum absorption peak around 400 nm. Aer the addition of NTR in the presence of 500 mM NADH, a new maximum absorption peak at around 425 nm was observed with the chromogenic changes easily detected by the naked eye from the initial colorless to yellow. As shown in Fig. 1b, the free-probe NTR-NO 2 was nearly non-uorescence upon excitation at 420 nm, which was in good agreement with the absorption spectra. Upon excitation at 430 nm, it was obvious that the uorescence intensity at 541 nm showed gradual enhancement upon the NTR titration in the presence of NADH (Fig. 2).
To demonstrate the sensitivity and selectivity of NTR-NO 2 , the uorescence spectra of NTR-NO 2 (10 mM) with NTR in various concentrations in the presence of 500 mM NADH were measured in the PBS buffer (pH ¼ 7.4, 20% DMSO). Importantly, the emission intensity of NTR-NO 2 (10 mM) at 541 nm was enhanced approximately 30-folds aer incubating with NTR (6 mg mL À1 ) in the presence of 500 mM NADH. The plots of I 541 nm (l ex ¼ 420 nm) against the concentrations of NTR ranging from 0 mg mL À1 to 4.5 mg mL À1 exhibited a good linear relationship (0.9896) (Fig. 2). Moreover, the detection limit of NTR-NO 2 toward NTR was determined to be 58 ng mL À1 . Compared with the probe NTR-NO 2 , the uorescence quantum yield (Y s ) of the active version aer responding to NTR increased by nearly 23folds (from 0.019 to 0.43) (Table S2 †). In addition, the probe NTR-NO 2 (10 mM) was incubated with various biologically relevant species (500 mM) in the presence of 500 mM NADH to assess the selectivity of NTR-NO 2 towards NTR. As shown in Fig. 3, the probe NTR-NO 2 showed apparent uorescence changes only in    the presence of NTR. Other biologically relevant substances cannot cause signicant uorescence changes even at higher concentrations. Moreover, we further incubated the probe of NTR-NO 2 with a strong reducing agent NaBH 4 in the presence of NADH. No notable uorescence changes were observed (Fig. S5 †). These results revealed that the probe NTR-NO 2 could be highly-sensitive and highly selective to monitor the level of NTR by the remarkable turn-on signal.

Time-dependence and pH effect of NTR-NO2 to NTR
The kinetic proles of the probe NTR-NO 2 (10 mM) incubated with different levels of NTR (0-8 mg mL À1 , 500 mM NADH) was studied. As shown in Fig. 4a, the solution of the probe NTR-NO 2 gradually displayed a uorescence increase at 541 nm upon the addition of NTR in the presence of NADH, which was saturated over 20 min. Moreover, no uorescence intensity change was observed in the free-probe, indicating that the probe was stable in such a buffer system. According to the above kinetic results, the probe NTR-NO 2 may be used to rapidly detect NTR. Furthermore, the pH effect on the detection of NTR activity was investigated to assess the potential applications of the probe NTR-NO 2 under different conditions. The probe NTR-NO 2 (10 mM) showed weak emission intensity in diverse pH environments, and the uorescence intensity increased signicantly aer responding to NTR (8 mg mL À1 ) in the presence of NADH within a wide pH range (Fig. 4b). It suggested that our probe could be applied to rapidly monitor NTR in the pathological or physiological condition.

Fluorescence imaging of NTR within living cells
Encouraged by the favorable properties of the probe NTR-NO 2 , we conducted the uorescence imaging experiments in living cells. First, we evaluated the cytotoxicity of the probe NTR-NO 2 via MTT assays. As shown in Fig. S6, † it indicated that the probe NTR-NO 2 has no remarkable cytotoxicity to the cells even up to 50 mM. Then we assessed whether it could monitor intracellular hypoxia. HeLa cells were incubated with probe NTR-NO 2 (10 mM) under normoxic (20% p O2 ) and hypoxic (1% p O2 ) conditions. As shown in these confocal uorescence images (Fig. 5), there was only weak uorescence emission under 20% p O2 condition. However, an obvious uorescence intensity could be observed under 1% p O2 condition resulted in more active NTR and in the formation of NTR-NH 2 . The above ndings demonstrated that our probe could be used to monitor endogenous NTR in tumor cells.

Conclusion
In summary, a novel turn-on probe NTR-NO 2 was prepared in this study based on the fused four-ring quinoxaline skeleton detecting NTR. This probe exhibited signicant uorescence emission change catalyzed by NTR in the presence of NADH based on ICT. Our probe showed high sensitivity and selectivity toward NTR. Moreover, we have demonstrated that the probe NTR-NO 2 can be effectively used to monitor hypoxia via the detection of NTR in HeLa cells. We anticipate that the probe NTR-NO 2 could be used as a powerful tool for investigating the biological functions of NTR in vivo.

Conflicts of interest
There are no conicts to declare. Fig. 4 (a) The plot of fluorescence intensity at 541 nm of NTR-NO 2 (10 mM) with respect reaction time for varied concentrations of NTR in the presence of 500 mM NADH at 37 C. (b) Effect of pH on the fluorescence intensity changes of probe NTR-NO 2 (10 mM) at 541 nm in absence (black) and presence (red) of NTR (8 mg mL À1 ) in PBS buffer (20% DMSO) for 20 min.