Novel Lysosome-Targeting Fluorescence Off-On Photosensitizer for Near-Infrared Hypoxia Imaging and Photodynamic Therapy In Vitro and In Vivo

Photodynamic therapy (PDT) has emerged as a new antitumor modality. Hypoxia, a vital characteristic of solid tumors, can be explored to stimulate the fluorescence response of photosensitizers (PSs). Considering the characteristics of PDT, the targeting of organelles employing PS would enhance antitumor effects. A new multifunctional cyanine-based PS (CLN) comprising morpholine and nitrobenzene groups was prepared and characterized. It generated fluorescence in the near-infrared (NIR) region in the presence of sodium dithionite (Na2S2O4) and nitroreductase (NTR). The response mechanism of CLN was well investigated, thus revealing that its obtained reduction product was CLNH. The obtained fluorescence and singlet oxygen quantum yield of CLNH were 8.65% and 1.60%, respectively. Additionally, the selective experiment for substrates indicated that CLN exhibited a selective response to NTR. Thus, CLN fluorescence could be selectively switched on and its fluorescence intensity increased, following a prolonged stay in hypoxic cells. Furthermore, fluorescence colocalization demonstrated that CLN could effectively target lysosomes. CLN could generate reactive oxygen species and kill tumor cells (IC50 for 4T1 cells was 7.4 μM under a hypoxic condition), following its response to NTR. NIR imaging and targeted PDT were finally applied in vivo.


Introduction
Many strategies, particularly photodynamic therapy (PDT), are emerging as new and noninvasive therapy modalities for cancer treatment [1][2][3]. PDT relies on specific lightgenerated reactive oxygen species (ROS) in the presence of photosensitizers (PSs) to kill cancer cells. Additionally, PSs generally possess certain fluorescence-emitting abilities and can facilitate antitumor effects and image-guided therapy during PDT [4,5]. However, the fluorescence of PS is "always-on", resulting in low signal-to-noise ratios and poor imaging qualities [6][7][8]. Moreover, PS-generated ROS exhibit a short life (0.03-0.18 ms) and limited distance of action (0.01-0.02 µm) [9][10][11]. Therefore, the location of the PS-generated ROS is directly proportional to the anticancer effect. If vital or fragile organelles can be precisely enriched with PSs, the concentrated ROS will focus on these "lethal sites" of cancer cells and be more lethal than randomly dispersed PSs in tumor tissue or cells [11][12][13][14][15]. Organelles Scheme 1. Targeting and NIR-florescence-activation mechanisms of the CLN. The cyanine fluorophore is marked in bright red in the structure of CLNH; the lysosome-targeted morpholine group is labeled in purple; the nitrobenzene groups as an NTR-response unit are marked in blue in the structure of CLN, which are reduced to amino groups after being triggered by NTR, marked in dark red in the structure of CLNH.

Design and Synthesis of the Photosensitizer CLN
Three factors were considered when designing lysosome-targeted NTR-stimulated NIR PS with fluorescence responses. First, the absorption and fluorescence emission wavelengths of the fluorescent unit of PS were preferably in the NIR region to increase the light-penetration depth, reduce the background, and improve the imaging quality and PDT effect [44]. Second, morpholine, which is known to facilitate the localization of PS in the cellular lysosome, as well as enhance the cell-killing effect of PS, was selected as the lysosome-targeting group [11]. Third, nitrobenzene, which has exhibited high sensitivity to NTR in vitro and in vivo, was selected as the recognition site for NTR. Additionally, the fluorescence of PS was quenched via the PET effect and restored after responding to NTR [16]. Here, we prepared our novel cyanine-based lysosomal PS (CLN) by introducing the morpholine and nitrobenzene groups, which could selectively locate lysosomes, demonstrate a specific response to NTR, and facilitate NIR fluorescence imaging and PDT (Scheme 1). CLN was obtained by a four-step synthesis (Scheme 2). Further, the structures of the intermediates and CLN were characterized by nuclear magnetic resonance (NMR) spectroscopy and high-resolution mass spectrometry (HRMS) (Figures S1-S8). Scheme 2. Synthetic route of the CLN. Figure 1A shows that the maximum absorption wavelength of CLN was 611 nm. Expectedly, almost no fluorescence was detected before the CLN response. The absorption and fluorescence spectra of CLN were recorded under different pH conditions (pH [2][3][4][5][6][7][8][9][10][11][12], as shown in Figure S11A,B. Notably, the fluorescence signals of CLN were extremely weak, due to the PET effect of the nitrobenzene moiety. Further, the pKa value of 6.65 for "-NH-" at meso-CLN was obtained via fitting the absorbance at 411 nm vs. values of pH ( Figure S11C). Thus, we first utilized sodium dithionite (Na 2 S 2 O 4 ) to simulate a reducing environment since it is a reliable reducing agent for aromatic nitro groups. After adding Na 2 S 2 O 4 , the maximum absorption wavelength of CLN shifted from 611 to 623 nm, indicating the probable production of a new product. Concurrently, an intense fluorescence signal was generated in the NIR region (the fluorescence intensity at the maximum wavelength (758 nm) was 71 times higher than before). Moreover, the fluorescence quantum yields of CLN were 0.13% and 8.65% before and after the addition of Na 2 S 2 O 4 in a phosphatebuffered saline (PBS) solution, respectively. Finally, the mechanism of the generation of fluorescence was confirmed via NMR titration and MS. Following the addition of Na 2 S 2 O 4 into the CLN solution (DMSO-d 6 ), a new peak, which could be a signal for the hydrogen of aromatic amines, appeared at 5.00 ppm ( Figure S9). Furthermore, the mass peak, reflecting the reduction product (CLNH), was generated at m/z = 852.5080 ( Figure S10). Thus, the above results indicated that the fluorescence was generated by the reduction of the CLNattached nitrobenzene groups into aminobenzene groups in the presence of Na 2 S 2 O 4 , which also generated CLNH. Scheme 1 outlines the NIR-florescence-activation mechanism of CLN. Further, CLN for detecting Na 2 S 2 O 4 was investigated. Figure 1B shows that the concentration of Na 2 S 2 O 4 and the fluorescence intensity at 758 nm exhibited good linear fitting (0-0.4 mM), demonstrating the quantitative feasibility of Na 2 S 2 O 4 . Following the above quantification curve, Na 2 S 2 O 4 exhibited a detection limit of 0.13 µM.

Photophysical Properties
The singlet oxygen probe, Singlet Oxygen Sensor Green (SOSG), was employed to evaluate the singlet oxygen generation of CLNH (the reduction product CLN). Figure 1C shows that the fluorescence signals of SOSG increased with the laser irradiation time in the CLNH solution. Under the same condition, SOSG displayed almost no fluorescence under laser irradiation ( Figure 1D). Employing methylene blue as the standard (0.39 in water), the obtained singlet oxygen quantum yield of CLNH was 1.60% [45]. All the optical data of CLN and CLNH in PBS were shown in Table S1, including the maximum absorption and emission wavelength, Stokes shifts, molar extinction coefficient, and quantum yields of fluorescence and singlet oxygen. Inspired by the above results, we compared the changes in the fluorescence of CLN before and after its response to NTR ( Figure 1D). CLN barely displayed fluorescence without the addition of NTR. However, the fluorescence intensities at 744 nm demonstrated 17-and 29-fold increases at 4 and 5 µg/mL concentrations of NTR, respectively. Moreover, the maximum emission (744 nm) of CLN reduction by NTR exhibited blue shift compared to the CLN reduction by Na 2 S 2 O 4 (758 nm), which was due to the different solvent environments, one in PBS/10% EtOH and the other in PBS buffer ( Figure S11D). To investigate the selectivity of CLN for NTR, the fluorescence spectra and intensities at 744 nm were recorded in the presence of biological species, including glutathione (GSH), vitamin C (Vc), cysteine (Cys), glycine (Gly), and nicotinamide adenine dinucleotide phosphate (NADPH), as well as other species, including inorganic salts (NaCl, MgCl 2 , and ZnSO 4 ) and reactive oxygen/nitrogen species (ClO − , and H 2 O 2 ). Figure 1D shows that only the "NTR" group exhibited a strong fluorescence signal at 744 nm. No significant fluorescence was generated in the other groups, confirming that CLN exhibited a highly selective response to NTR in complicated environments.

Fluorescence Imaging in Hypoxic Cells
Considering the excellent performance of our CLN, we evaluated its imaging performance under cellular hypoxic conditions. We investigated the selectivity of CLN imaging by more precisely controlling the oxygen content and adding NTR inhibitors (dicoumarin) [46]. In this test, the 4T1 and HeLa cells were first incubated with CLN for 3 h. Thereafter, the cells were incubated under normoxic (20% O 2 ) and hypoxic conditions (<1% O 2 ) for 8 h. Additionally, a hypoxic group was incubated with 0.5 mM dicoumarin for 1 h before hypoxia. Figure 2A,B show that almost no fluorescent signal was observed in the normoxic cells. Expectedly, the bright red fluorescence signal appeared in the hypoxia ones. Under the hypoxia conditions, the red fluorescence disappeared when dicoumarin was added to the cells. The statistics of the fluorescence intensities ( Figure 2C,D) corroborated the cell imaging results. These results indicated that the fluorescence intensity of CLN was closely related to the degree of oxygen concentration and could selectively respond to NTR.

Subcellular Localization of CLN
To confirm that CLN could target lysosomes upon entering the cells, we selected Lyso-Tracker Green, a commercial lysosomal probe, for the colocalization imaging experiments. Further, Mito-Tracker Green, a mitochondria-targeting probe, was employed for the colocalization studies. Figure 3A shows that the fluorescence signal of CLN overlapped well with that of Lyso-Tracker Green in the HeLa cells under hypoxic conditions (the Pearson coefficient was 0.95). However, the region of the distribution of the CLN fluorescence was significantly different from that of Mito-Tracker Green (the Pearson correlation coefficient was 0.18). Thus, we confirmed that CLN could target lysosomes in cells.

In Vitro Generation of ROS and the Phototoxicity of CLN
The intracellular generation of ROS in the 4T1 and HeLa cells was further investigated employing DCFH-DA as the intracellular ROS probe. After the laser irradiation without any other treatment, the CLN-treated cells displayed significant green fluorescence signals, indicating that CLN generated ROS in the cells ( Figure 4). After adding N-acetyl-L-cysteine, the ROS scavenger, the green fluorescent signal disappeared.
After treating the 4T1 cells with CLN for 24 h in a normoxic environment, they were irradiated with a light-emitting diode (LED) light (220.7 mW/cm 2 ) for 10 min to investigate its phototoxicity. The cell viability of the tested tumor cells decreased continuously with the increasing CLN concentrations, exhibiting significant cytotoxicity ( Figure S12A). The IC50 value of CLN for 4T1 was 0.7 µM, indicating that CLN exhibited phototoxicity and potential application for PDT. More crucially, CLN still exerted an excellent killing effect on the 4T1 cells under hypoxic conditions (IC50 = 7.4 µM, Figure S12B). Furthermore, CLN did not exhibit significant dark toxicities in the concentration ranges of 0-1.6 and 0-12 µM under normoxic and hypoxia conditions, respectively ( Figure S12).

In Vivo Imaging and Photo-Theranostic Study of CLN
Motivated by the above results, we evaluated whether CLN could respond to NTR in vivo. The BALB/c mice bearing 4T1 tumors were established and randomly divided into two groups comprising three mice each. The groups were named "CLN" and "CLN + Dicoumarin". CLN was directly injected into the tumors of the mice in the "CLN" group, after which they were imaged at different times (2, 20, 40, 60, 80, 100, and 120 min) by an in vivo imaging system. Regarding the "CLN + Dicoumarin" group, the same procedure as that of the "CLN" group was performed after injecting dicoumarin into their tumors for 10 min. Figure 5A,B show that the fluorescence signals of the tumors of the "CLN" group increased gradually and reached their maximum at 80 min. Expectedly, the fluorescence signals of the tumor target in the "CLN + Dicoumarin" group that was treated with the NTR inhibitor were extremely weaker. The "CLN" group exhibited strong fluorescence signals after 6 h of the intratumor injection of CLN ( Figure 5C). The CLN solutions did not exhibit any fluorescence signal under the same conditions, thus confirming that CLN could respond to NTR, which selectively turned on its fluorescence in vivo. Subsequently, the in vivo biodistribution of CLN, following tail-vein injection, was performed on the BALB/c mice bearing 4T1 tumors by an in vivo imaging system. CLN could be effectively accumulated in the tumor site to turn on the fluorescence, exhibiting the strongest fluorescent signal at 2-6 h ( Figure 6A). The statistical results of the fluorescence intensities indicated that the optimal temporal window was 2 h post-injection for PDT ( Figure 6B). At 24 h post-administration, the ex vivo tissue imaging was performed on a small animal imaging system ( Figure S14). The fluorescence signals were mainly observed at liver, spleen, and kidney, indicating that CLN was mainly metabolized by the kidney, liver, and spleen. In addition, fluorescent signals were still present in the tumor area at 24 h, indicating that CLN could accumulate at the tumor site for a longer period. Thus, CLN could be a promising NIR imaging-guided PDT agent for tumor treatment applications.
Motivated by the accumulation of CLN, as well as the turned-on fluorescent signal in the tumor regions, we studied its PDT efficiency in vivo. The 4T1 tumor-bearing mice were tail-vein injected with CLN (2.5 µmol/kg) and saline as controls. In the in vivo study, the 4T1 tumor-bearing BALB/c mice were randomly divided into four groups comprising five mice each, and the groups were denoted as "CLN-Laser", "CLN-Dark", "Control-Laser", and "Control-Dark". Based on NIR imaging results, the optimal phototherapeutic windows were determined after 2 h of injection. Subsequently, all the mice in the "CLN-Laser" and "Control-Laser" groups were exposed to a 660 nm laser (0.2 W/cm 2 ) every day after injection for 15 min at day 1, 3, and 5. The in vivo antitumor effects of CLN were investigated by monitoring the changes in the tumor volume for seven days. Figure 7A shows the negligible inhibitions of the tumor growths in the "CLN-Dark", "Control-Laser", and "Control-Dark" groups, indicating that irradiation alone did not exert any antitumor effect. Surprisingly, the "CLN-Laser" group revealed the most effective antitumor activity after the treatment, displaying slow tumor growth on day 7. The photographs of the tumor further confirmed the optimal antitumor activity of the "CLN-Laser" group, thereby availing visual evidence ( Figure 7B). Additionally, CLN exhibited low in vivo toxicity in the treatments, as further supported by the serum biochemical indices ( Figure S13), as well as the hematoxylin and eosin (H&E) staining of the major organs, including the heart, liver, spleen, lung, and kidney of each mouse from all the groups, indicating the absence of significant lesions in the normal organs ( Figure 7C). On the seventh day, all four groups of mice were sacrificed to collect their tumor tissues; they were sliced and stained with H&E. The images of the histological sections demonstrated that the "CLN-Laser" group presented the most effective antitumor effect among the different therapeutic agents, effectively inducing tumor necrosis and apoptosis, as well as inhibiting the proliferation of the cancer cells ( Figure 7C).

General Information
All reagents were obtained from Energy Chemical Co. Ltd. (Weinan, China) and used as received. Solvents were of analytical grade. Flash column chromatography was performed with silica gel (200-300 mesh) and dichloromethane/methanol was used as eluent. All NMR experiments were measured by using an Agilent 400 DD2 spectrometer (Agilent, Santa Clara, CA, USA) at room temperature. The HRMS were obtained with Electrospray-quadrupole time-of-flight mass spectrometer (MicrOTOF-Q II 10410, Bruker, NASDAQ, Billerica, MA, USA).

Synthesis of Compound 1
Compound 1 was prepared following a modified procedure of the previous literature [47]. The phosphorous oxychloride (5.0 mL, 53.6 mmol) was added dropwise to the mixed solution of DMF (6.0 mL, 77.6 mmol) and DCM (7 mL) in an ice bath, and stirring was continued in the ice bath for 0.5 h. Then, 4-(4-hydroxyphenyl)cyclohexan-1-one (1.0 g, 5.2 mmol) was added, and the resulting mixture was vigorously stirred and heated to 50 • C for 5 h, cooled to room temperature, and then poured into ice-cooled water 200 mL. The mixture was kept at 4 • C refrigerator overnight and then filtered and freeze dried to obtain compound 1 (0.63 g, 45%).

Synthesis of Compound 2
The 2,3,3-trimethyl-3H-indole (3.0 g, 19.5 mmol) and p-nitrobenzyl bromide (5.4 g, 25.0 mmol) were added to the toluene (30 mL) and refluxed in an oil bath at 90 • C for 12 h; a large amount of purple-red solid was formed, which was cooled to room temperature, filtered under reduced pressure, washed with toluene (5 mL × 3), and dried to obtain 3.9 g of purple-red powder with yield of 70%. The NMR spectra are shown in Supplementary Materials ( Figures S1 and S2). 1

Synthesis of Compound 3
The compound 1 (0.30 g, 1.13 mmol), compound 2 (0.77 g, 2.60 mmol), and anhydrous sodium acetate (0.09 g, 1.13 mmol) were added to the anhydrous ethanol 20 mL, and the mixture was stirred at 80 • C for 4 h, the solvent was evaporated under reduced pressure, the product was purified by using silica gel column chromatography (DCM: MeOH = 50:1) to give green solid 0.38 g, yield 41%. The NMR spectra and HRMS are shown in Supplementary Materials (Figures S3-S5). 1

Synthesis of CLN
The compound 3 (0.1 g, 0.12 mmol) and N-(2-aminoethylmorpholine) (0.03 g, 0.23 mmol, 30 µL) were added to acetonitrile (10 mL), and the mixture were refluxed in an oil bath at 60 • C for 4 h; the solvent was evaporated under reduced pressure, the product was purified by using silica gel column chromatography (MeOH:DCM = 1:20) to give blue solid 60 mg, yield 54%. The NMR spectra and high-resolution mass spectrometry are shown in Supplementary Materials (Figures S6-S8). 1 13

Absorption and Fluorescence Spectroscopy
CLN was dissolved in methanol at a concentration of 1 mM as the stock solution and stored at -20 • C for further use. Na 2 S 2 O 4 was freshly prepared in 5 mM solution with deionized water and ready to use. The absorption and fluorescence spectroscopy of CLN were performed in PBS (10 mM, pH = 7.4) at a concentration of 10 µM in the absence or presence of Na 2 S 2 O 4 (0-0.4 mM) at 37 • C. PBS was used for all experiments at 10 mM, pH = 7.4.

Response Mechanism Elucidation
Na 2 S 2 O 4 was dissolved in D 2 O to a concentration of 1 mM, and 5 mg of CLN was dissolved in DMSO-d 6 and its 1 H NMR spectrum was tested, then an appropriate amount of sodium hydrosulfite solution was added and then its 1 H NMR spectrum was tested at 1, 30, and 60 min. Subsequently, this solution was then subjected to HRMS. The spectral data are presented in Supplementary Materials ( Figures S9 and S10).

Evaluation of Singlet Oxygen Quantum Yield (Φ ∆ )
In order to evaluate the photophysical properties of CLN after response, its reduction products (CLNH) were firstly prepared. Briefly, CLNH was separated by using a flash column after reacting with a 1:1 (V:V) mixture of methanolic solution of CLN (10 mg) and aqueous solution of Na 2 S 2 O 4 (43.5 mg) for 5 min.
The singlet oxygen quantum yield of CLNH was carried out by surveying the fluorescence intensity of SOSG at 525 nm under laser irradiation for various time. Firstly, the lyophilized powder of SOSG was prepared in methanol to 5 mM stock solution for use according to the instruction and diluted to 100 µM with PBS, further prepared in a 3 mL cuvette to 2 µM working solution, mixed with 10 µM of CLNH in PBS, which was exposed to 660 nm (200 mW/cm 2 ) laser lamp irradiation for different time, and the fluorescence spectra were collected by fluorescence spectrophotometer with excitation wavelength of 504 nm. Methylene Blue (MB) was used as the reference. At last, the singlet oxygen quantum yield was calculated with the following Equation (1): Φ ∆ represents the singlet oxygen quantum yield of the tested photosensitizer; Φ MB represents the singlet oxygen quantum yield of MB and the value is 0.39 in water [45]; x represents the tested photosensitizer CLNH; k represents the slope of the increase in the fluorescence intensity at 525 nm of SOSG with the addition of irradiation time; F is the correction factor which is calculated by the following Equation (2): OD represents the absorbance of the mixture at 660 nm.

Determination of Fluorescence Quantum Yield (Φ f )
The fluorescence quantum yields of CLN and its reduction product CLNH in PBS were determined using cresyl violet (CV) as the reference. Using 570 nm as the excitation wavelength, the absorbance of CLN and its solution after reaction with Na 2 S 2 O 4 was adjusted to be less than 0.05 at 570 nm. The fluorescence spectra of CLN, CLNH, and CV were tested under the same conditions and the integrated areas were calculated. The fluorescence quantum yields of CLN and CLNH were calculated by the following Equation (3): Φ f represents the fluorescence quantum yield of the tested photosensitizer; Φ CV represents the fluorescence quantum yield of CV and the value is 0.43 in water [48]; I x and I CV represent integrated areas of the tested PS and CV, respectively; A x and A CV represent absorbance of the tested PS and CV at 570 nm, respectively.

NTR Recognition and Selectivity
An amount of 1 mg of NTR was dissolved in 10 mL of sterile PBS, divided into 20 lyophilization tubes, and stored in a −20 • C refrigerator until use. NADPH is prepared in deionized water as a 15 mM solution, ready to use. In a 5 mL tube, PBS buffer containing 10% EtOH and 1 mM CLN (30 µL) were mixed, and then 15 mM NADPH (100 µL) was added to obtain a final concentration of 500 µM. An appropriate volume of NTR was added to the sample solution. The final solution volume was adjusted to 3 mL with PBS buffer containing 10% EtOH. After rapid mixing of the solution, it was incubated at 37 • C for 30 min, and transferred to a 10 mm × 10 mm quartz cell. Fluorescence spectra were recorded with λ ex = 648 nm.

Cell Culture
The 4T1 cells and HeLa cells were cultured in DMEM media with 10% FBS and 1% penicillin/streptomycin solution. Cells were grown in a saturated humidity and 5% CO 2 atmosphere at 37 • C in a cell incubator.

Fluorescence Imaging in Hypoxia Cells
A cellular hypoxic environment was created by passing the gas mixture (5% CO 2 + 95% N 2 ) through the hypoxia chamber. The 4T1 and HeLa cells were first incubated normally (20% O 2 ) at a density of 5 × 10 4 for 24 h, respectively. Cells were treated with CLN (5 µM) for 3 h and then washed 2-3 times with PBS and fresh medium was added. Cells in the hypoxic group were treated with 0.5 mM NTR inhibitor (dicoumarin) for 1 h and then incubated in the hypoxia chamber with the gas mixture for 8 h at 37 • C in a cell incubator. The normoxic group was incubated normally (20% O 2 ) for 8 h and imaged by inverted fluorescence microscopy (Olympus IX73, Tokyo, Japan). Fluorescence intensity analysis was performed by Image J software.
Fluorescence intensity was quantified as follows: 6 cell areas and 3 backgrounds were randomly circled in the image using Image J software, the integrated density of the circled cells and the mean gray value of the background area by Image J software were calculated, and the corrected integrated density of each circled cell was found by the following Equation (4): IntDen (corrected) represents the integrated density of each circled cell deducting the mean gray value of the background; IntDen (cell) indicates the total fluorescence intensity of the region; Area (cell) indicates the area of the cell; and Mean (BG) indicates the mean gray value of the background region (the average of mean of the three background regions).

Subcellular Colocalization Fluorescence Imaging of Cells
The HeLa cells were seeded in a glass bottom dish and incubated for 24 h under normoxia condition (20% O 2 ) at a density of 5 × 10 4 . Then, the cells were incubated for another 8 h under hypoxia condition with coverslips. Next, cells were treated with CLN (5 µM) for 3 h and then washed 2-3 times with PBS and fresh medium with Hoechest 33,342 (1×) and Lyso-Tracker Green (or Mito-Tracker Green, respectively) (300 nM) and incubated for 30 min. Last, these cells were imaged by laser confocal microscope (Carl Zeiss LSM 900). The blue channel was excited at 405 nm, the green channel was excited at 488 nm, and the red channel was excited at 640 nm. Colocalization analysis was performed by Image J software.

Detection of Cellular ROS within DCFH-DA
The 4T1 and HeLa cells were seeded on a culture dish at a density of 5 × 10 4 for 24 h. CLN (5 µM) was added and incubated for 3 h. Then, the cells were washed with PBS 2-3 times and then treated with DCFH-DA (5 µM) for 30 min and imaged in the presence or absence of a laser (660 nm, 200 mW/cm 2 ) irradiated for 10 min under an inverted fluorescence microscopy (Olympus IX73). In addition, NAC (N-Acetyl-L-cysteine) was used as a reactive oxygen scavenger, and after incubation with CLN, the cells were treated with NAC (5 mM) for 1 h, then DCFH-DA was added, and imaging was performed after light exposure. They were washed 2-3 times with PBS after each change in incubation medium, and the rest of the operation was unchanged.

Cytotoxicity Assay
Cytotoxicity assays in the presence or absence of light were performed by reducing MTT. Firstly, 4T1 cells were grown in a 96-well plate at a density of 5 × 10 4 , and incubated under normoxia for 24 h, and then incubated in a hypoxia chamber for another 8 h, treated with different concentrations of CLN (0-12 µM, containing 1% DMSO) for 4 h, and the light group was irradiated by a 96-well LED light (220.7 mW/cm 2 ) for 10 min and continued to be incubated until 24 h under hypoxia. As for the dark group, the cells were kept out of light with the other treatments unchanged. Lastly, these cells were incubated with 100 µL MTT (5 mg/mL in DMEM) solution for 4 h at 37 • C in cell incubator, and then the medium was removed carefully, 100 µL of DMSO was added to each well, and the absorbance was tested at 490 nm on a microplate reader. The cell viability was calculated according to the following Equation (5):

General Information
All procedures were carried out in accordance with the interrelated legislation and ethical regulation guidelines were endorsed by the Model Animal Research Center of Zunyi Medical University. The experimental animals were 18~22 g BALB/c female mice, purchased from Spelford Beijing Biotechnology Co., Ltd. (SPF grade, license number: SCXK (Beijing) 2016-0002). The BALB/c mice were first adaptively housed for 1 week, the dorsal hair was shaved, and then they were inoculated with 100 µL of 4T1 cells at a density of 1 × 10 7 at the lower right position of their back, normal rearing was continued for 3~5 days, tumor growth observed, and the experiment was performed as the tumor grew to 80 mm 3 . All CLN injections were 500 µM of CLN solution (Tris buffer, 10 mM, pH = 7.4, containing 10% DMSO) administered at a dose of 2.5 µmol/kg. All in vivo imaging experiments were performed on a NightOWL II LB983small animal imaging system (Berthold Technologies Gmbh & Co. KG).

In Vivo Imaging of Mice with Endogenous NTR
Six tumor-bearing BALB/c mice were randomly divided into two groups. One group of tumor-bearing mice was injected intratumorally with CLN at a dose of 2.5 µmol/kg, the other mice were injected intratumorally with 100 µL dicoumarin (500 µM) for 10 min before the injection of CLN (2.5 µmol/kg). The time-dependent fluorescence changes were collected at every 20 min within 120 min after injection with a small animal imaging system. The excitation wavelength was 640 nm, and the emission wavelength was 820 nm.

In Vivo Fluorescence Imaging and Tumor Targeting
Three tumor-bearing BALB/c mice were randomly selected and injected with CLN at a dose of 2.5 µmol/kg in the caudal vein, and imaged with a small animal imaging system at various times. The excitation wavelength was 640 nm, and the emission wavelength was 820 nm.

The Photocytotoxic Efficiency of Tumor-Bearing Mice
Twenty tumor-bearing BALB/c mice were randomly divided into four groups: "Controllaser group", "Control-dark group", "CLN-laser group", and "CLN-dark group". The control group was injected with 100 µL saline via tail-vein injection, the CLN group was injected with CLN at a dose of 2.5 µmol/kg, and the "CLN-laser group" was irradiated with a 660 nm (0.2 W/cm 2 ) laser for 15 min after treated with CLN via tail-vein injection for 2 h. The study period was 7 days, and the CLN was injected once a day at an interval, three times within 7 days, and the laser groups were irradiated for 15 min each day. The mice were weighed, and the tumor volume was measured and recorded daily. The tumor volume was measured by vernier calipers and calculated according to the following Formula (6): a denotes the longest diameter of the tumor, and b represents the vertical direction diameter of the tumor according to a.
At the end of the study on day 7, the tumor tissues were taken for size measurement and H&E staining. Blood was collected from each mouse, and serum was collected to measure serum biochemical indices: aspartate aminotransferase (AST), glutamate transferase (ALT), creatinine (CREA), and urea (UREA), and compared with normal mice for biochemical parameters. The reference ranges of biochemical parameters in normal healthy BALB/c female mice were: ALT (38.31 ± 15.07 U/L), AST (151.2 ± 66.6 U/L), CREA (11.27 ± 4.83 µmol/L), UREA (7.49 ± 1.65 mmol/L) [49]. The serum biochemical data are presented in Supplementary Materials (Figure S12).

H&E Staining
The heart, liver, spleen, lung, kidney, and tumor tissues of the four groups of mice were taken and fixed with 4% formaldehyde solution, and then dehydrated and embedded with paraffin to slice the 4 µm sections, which were conducted for hematoxylin and eosin (H&E) under standard methods and sent to a microscope (Olympus CX41).

Conclusions
In summary, we designed and synthesized a new cyanine-based PS, CLN, which was modified with nitrobenzene and morpholine groups. The maximum absorption wavelength of CLN was located in the red-light region. As expected, almost no fluorescence was detected before the CLN response. However, in the presence of Na 2 S 2 O 4 , a fluorescence signal appeared in the NIR region, and the fluorescence intensity at 758 nm was 71 times higher than before. The mechanism of the generation of fluorescence was confirmed by NMR titration and MS. The MS results demonstrated that the nitro group was reduced into amino to produce a new compound, CLNH. Further, the calculated singlet oxygen quantum yield of CLNH was 1.60%. Similarly, the fluorescence of CLN also exhibited a specific response to NTR in a complicated environment. The fluorescence of CLN could be selectively turned on in hypoxic cells. Cellular fluorescence colocalization experiments revealed that CLN targeted the lysosome effectively. Thus, selective imaging and targeted PDT were concurrently achieved in vitro and in vivo.
Taking an upregulated level of NTR as the tumor biomarker [25], CLN achieves a selective turn-on of fluorescence, thus facilitating precise PDT in potential clinical applications. In terms of available results, perhaps a local application of CLN may be more effective, as CLN is limited by tumor-targeting ability and photodynamic effects under hypoxic conditions. Of course, as a potential PS, the molecular structure and properties of CLN need to be improved if used in the clinic. The following improvement strategies may further increase the potential for clinical application of CLN: (1) enhancement of singlet oxygen quantum yield; (2) development of Type I PDT to reduce the dependence on oxygen concentration; (3) construction of oxygen self-sufficient photosensitizers (e.g., modified perfluorocarbon structures); and (4) modification of tumor-targeting groups to enhance PS concentration in the tumor region. The above improvements may be beneficial for clinical use, and in the future, we expect to develop a PS that can be used in the clinic.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/molecules27113457/s1, Figures S1, S3 and S6: The 1 H NMR spectra of the compounds 2, 3, and CLN; Figures S2, S4 and S7: The 13 C NMR spectra of the compounds 2, 3, and CLN; Figures S5, S8 and S10: Mass spectra of the compounds 3, CLN, and CLNH; Figure S9: NMR titration spectra of CLN and sodium dithionite; Figure S11: Absorption and fluorescence spectra of CLN at varied pH; the pKa fitting curve and the fluorescence spectra of CLNH in PBS buffer and PBS buffer/10% EtOH; Figure S12: MTT tests of 4T1 cells; Figure S13: Serum biochemical values; Figure S14: Fluorescence imaging and intensity of major ex vivo organs and tumor; Table S1: The optical data of CLN and CLNH in PBS; Table S2: The summary of photophysical properties of reported similar probes.
Author Contributions: Investigation and data curation, S.D. and M.Y.; software and formal analysis, J.L. and H.L.; investigation, writing-original draft preparation, funding acquisition, supervision, and visualization, J.G.; Conceptualization, writing-review and editing, supervision, funding acquisition, G.W. and Z.Y. All authors have read and agreed to the published version of the manuscript.