Nanophotosensitizers Composed of Phenyl Boronic Acid Pinacol Ester-Conjugated Chitosan Oligosaccharide via Thioketal Linker for Reactive Oxygen Species-Sensitive Delivery of Chlorin e6 against Oral Cancer Cells

Chlorin E6 (Ce6)-incorporated nanophotosensitizers were fabricated for application in photodynamic therapy (PDT) of oral cancer cells. For this purpose, chitosan oligosaccharide (COS) was conjugated with hydrophobic and reactive oxygen species (ROS)-sensitive moieties, such as phenyl boronic acid pinacol ester (PBAP) via a thioketal linker (COSthPBAP). ThdCOOH was conjugated with PBAP to produce ThdCOOH-PBAP conjugates and then attached to amine groups of COS to produce a COSthPBAP copolymer. Ce6-incorporated nanophotosensitizers using the COSthPBAP copolymer were fabricated through the nanoprecipitation and dialysis methods. The Ce6-incorporated COSthPBAP nanophotosensitizers had a small diameter of less than 200 nm with a mono-modal distribution pattern. However, it became a multimodal and/or irregular distribution pattern when H2O2 was added. In a morphological observation using TEM, the nanophotosensitizers were disintegrated by the addition of H2O2, indicating that the COSthPBAP nanophotosensitizers had ROS sensitivity. In addition, the Ce6 release rate from the COSthPBAP nanophotosensitizers accelerated in the presence of H2O2. The SO generation was also higher in the nanophotosensitizers than in the free Ce6. Furthermore, the COSthPBAP nanophotosensitizers showed a higher intracellular Ce6 uptake ratio and ROS generation in all types of oral cancer cells. They efficiently inhibited the viability of oral cancer cells under light irradiation, but they did not significantly affect the viability of either normal cells or cancer cells in the absence of light irradiation. The COSthPBAP nanophotosensitizers showed a tumor-specific delivery capacity and fluorescence imaging of KB tumors in an in vivo animal tumor imaging study. We suggest that COSthPBAP nanophotosensitizers are promising candidates for the imaging and treatment of oral cancers.


Introduction
Most types of oral cancers are squamous cell carcinoma in the oral cavity, and they are easy to find at an early stage compared to systemic cancers [1]. Even though oral cancers are able to be cured at an early stage, they are frequently diagnosed at an advanced stage, with metastasis to other regions. Thus, at this stage, it makes treatment difficult in spite of easy accessibility [2,3]. Furthermore, oral cancers at an advanced stage lead to

Synthesis of COSthPBAP Conjugates
To synthesize COSthPBAP conjugates, PBAP was primarily conjugated with one end of the carboxylic acid of ThdCOOH as follows: 224 mg of ThdCOOH was dissolved in 5 mL of DMSO with 192 mg of EDAC and 115 mg of NHS to activate one carboxylic acid of ThdCOOH. Then, this was magnetically stirred for 3 h, and 270 mg of PBAP was added. This reaction was continued for 12 h to synthesize ThdCOOH-PBAP conjugates. After that, 192 mg of EDAC and 115 mg of NHS were added to this reaction and then stirred for 6 h to activate another end of the carboxylic acid of the ThdCOOH-PBAP conjugates. Following this, 600 mg of COS dissolved in 10 mL of an H 2 O/DMSO mixture (1/4, v/v) was added to this reaction. Then, 24 h later, the reactants were introduced into the dialysis membranes and dialyzed against deionized water for 2 d to remove the organic solvent, unreacted chemicals, and byproducts. To avoid saturation, the deionized water was exchanged every 3 h intervals, and after that, the resulting solution was lyophilized to obtain solid products. The yield of the COSthPBAP conjugates was calculated as follows: Yield (%, w/w) = [(weight of ThdCOOH + weight of PBAP)/weight of COSthPBAP conjugates] × 100.

Characterization of COSthPBAP Conjugates
1 H NMR spectra (500 mHz NB Fourier transform (FT)-NMR spectrometer, Varian Unity Inova; Varian Inc., Santa Clara, CA, USA) was employed to confirm the chemical composition and synthesis procedures of the conjugates. Each component and conjugates were dissolved in DMSO or a mixture of D 2 O/DMSO for analysis.

Fabrication of Ce6-Incorporated Nanophotosensitizers
COSthPBAP (40 mg) was reconstituted in 3 mL of deionized water, and then DMSO (5 mL) was added. To this solution, Ce6 dissolved in DMSO (2 mL) was added, and it was magnetically stirred for 10 min. This solution was poured into 10 mL of deionized water and introduced into a dialysis tube. The solution was dialyzed with deionized water over 1 d and the water was exchanged at 3 h intervals. The resulting solution was used for The Ce6 contents in the COSthPBAP nanoparticles were as follows: the lyophilized solids of the nanoparticles were distributed in deionized water (2 mL), and then DMSO (8 mL) was added. This solution was diluted with DMSO by more than 10 times, and then the Ce6 concentration was measured with a fluorescence spectrofluorophotometer (RF-5301PC, Kyoto, Japan). The excitation and emission wavelengths were 407 nm and 664 nm, respectively. The empty COSthPBAP nanoparticles were dissolved in H 2 O/DMSO (2/8 (v/v), 10 mL) and diluted with DMSO by ten times for a blank test.

Characterization of Nanophotosensitizers
Transmission electron microscopy (TEM) (H7600, Hitachi Instruments Ltd., Tokyo, Japan) was used to observe the morphology of the nanophotosensitizers. The nanophotosensitizer solution 20 µL was dropped onto a carbon film-coated grid followed by drying at room temperature for 6 h. Phosphotungstic acid (0.1%, w/w in H 2 O) was used to negatively stain the nanophotosensitizers. The observation of the nanophotosensitizers was performed at 80 kV.
The ultraviolet-visible (UV) absorption spectrum of the nanophotosensitizers was analyzed with a Genesys 10s UV-VIS spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA).
A Zetasizer (Nano-ZS, Malvern, Worcestershire, UK) was used to measure the particle size of the nanophotosensitizers. For the effect of H 2 O 2 on the particle size, the nanophotosensitizers fabricated as described above were diluted with phosphate-buffered saline (PBS, pH 7.4, 0.01 M), and H 2 O 2 was added to this solution, followed by incubation at 37 • C for 3 h. This solution was used to measure the particle size.
The fluorescence properties of the nanophotosensitizers were measured with the fluorescence spectrofluorophotometer (RF-5301PC, Kyoto, Japan) between 500 nm and 800 nm (excitation wavelength: 400 nm). Images of the fluorescence properties of nanophotosensitizers were observed with a Maestro 2 small animal imaging instrument (Cambridge Research and Instrumentation Inc., Woburn, MA, USA). The nanophotosensitizers were reconstituted in phosphate-buffered saline (PBS, pH 7.4, 0.01 M), and H 2 O 2 was added to this solution, followed by incubation at 37 • C for 3 h.

Drug Release Study
For the release study of Ce6, the nanophotosensitizer solution fabricated as described above was adjusted to 40 mL (1 mg/mg as a polymer) with deionized water, and then 5 mL of this solution was put into a dialysis membrane (MWCO: 2000 Da). After that, the dialysis tube was put into a conical tube with 45 mL of PBS (pH 7.4, 0.01 M). H 2 O 2 was added to this solution to study the oxidative stress on the drug release rate from the nanophotosensitizers. The solution was incubated under shaking (100 rpm) at 37 • C. At predetermined time intervals, whole PBS was collected to measure the Ce6 concentration and then replaced with fresh PBS. The Ce6 concentration was measured with a fluorescence spectrofluorophotometer (RF-5301PC spectrofluorophotometer, Kyoto, Japan) at an excitation wavelength of 407 nm and emission wavelength of 664 nm. All results are the mean ± standard deviation (S.D.) from three separate experiments.

Devices for Light Irradiation
For the singlet oxygen (SO) generation and PDT study, an expanded homogenous beam (SH Systems, Gwangju, Korea) was used as reported previously [42]. The cells or nanophotosensitizers were irradiated at 664 nm and the light density was 2.0 J/cm 2 [42]. The distance between the cells or nanophotosensitizer solutions from the LED panel was 40 cm. A 96-well plate or nanophotosensitizer solution was located in the center of the bottom panel. The light dose in the center of the bottom was determined with a photoradiometer (DeltaOhm, Padova, Italy). The light density was measured at more than 20 points, and then the dose of light was calculated.

Singlet Oxygen (SO) Generation of Nanophotosensitizers
The generation of singlet oxygen (SO) generation by the Ce6 or nanophotosensitizers was evaluated as follows [43,44]: An aqueous solution (1 mL) of free Ce6 or nanophotosensitizers (5 µg/mL of Ce6 equivalent in distilled water, 1% DMSO) was made. SOSG reagent (final concentration: 5 µM) was added to this solution and then irradiated with an expanded homogenous beam (664 nm, SH Systems, Gwangju, Korea) at different time points (0.5, 1, 2, and 5 min). The fluorescence intensity was measured with a fluorescence spectrophotometer (RF-5301PC, Shimadzu Co., Kyoto, Japan) at the excitation wavelength of 488 nm and the emission wavelength of 525 nm. This measurement was performed under dark conditions. (Grand Island, NY, USA). This solution was composed of streptomycin (10,000 µg/mL), amphotericin B (25 µg/mL), and penicillin (10,000 units/mL).

PDT of Oral Cancer Cells In Vitro
YD-38 or KB cells (2 × 10 4 cells) seeded in 96-well plates were treated with Ce6 or nanophotosensitizers. The Ce6 was dissolved in DMSO for the Ce6 treatment and then diluted by more than 100 times with serum-free media (DMSO final concentration: 0.5% (v/v)). The nanophotosensitizers in deionized water were sterilized with a 1.2 µm syringe filter and then diluted with serum-free media. The cells were incubated for 2 h in a 5% CO 2 incubator at 37 • C and, after that, washed with PBS twice. Serum-free fresh media (100 µL) were added to this, and then the cells were irradiated at 664 nm using an expanded homogenous beam (SH Systems, Gwangju, Korea). The light dose was 2.0 J/cm 2 . The light intensity was measured with a photo radiometer (Delta Ohm, Padua, Italy). The cells were further incubated for 24 h in a CO 2  An intrinsic dark toxicity test was performed without the irradiation of light at 664 nm.

Intracellular Ce6 Uptake and ROS Generation of Oral Cancer Cells In Vitro
Cells (2 × 10 4 cells) seeded into 96-well plates were treated with Ce6 or nanophotosensitizers for 2 h as described above and then washed with PBS twice. These cells were lysed with 50 µL of lysis buffer (GenDEPOT, Barker, TX, USA). The intracellular Ce6 uptake ratio was measured with the relative fluorescence intensity with an Infinite M200pro microplate reader (Tecan) (excitation wavelength: 407 nm, emission wavelength: 664 nm).
The intracellular ROS generation by the treatment of the Ce6 or nanophotosensitizers was evaluated with DCFH-DA. The cells (2 × 10 4 cells) were treated with Ce6 or nanophotosensitizers in serum-free RPMI media with DCFH-DA (final concentration: 20 µm) for 2 h and then washed with PBS twice. Fresh phenol red-free RPMI media (100 µL) were added, and then the cells were irradiated at 664 nm (2.0 J/cm 2 ). The intracellular ROS generation was measured at an excitation wavelength of 485 nm and an emission wavelength of 535 nm using a microplate reader (Infinite M200 PRO (Tecan)).
Fluorescence observation of the cells was carried out as follows: The cells (2 × 10 5 ) were seeded on the cover glass in six-well plates and then treated with Ce6 or nanophotosensitizers for 1 h. After that, the cells were washed with PBS twice, fixed with 4% paraformaldehyde (PFA) solution in PBS for 15 m, immobilized with immobilization solution (Immunomount, Thermo Electron Co. Pittsburgh, PA, USA), and then observed with a fluorescence microscope (Eclipse 80i; Nikon, Tokyo, Japan).

Animal Tumor Imaging and In Vivo PDT Efficacy against AT84 Tumor Xenograft Model
For animal tumor imaging, KB cells (1 × 10 6 cells) were subcutaneously injected into the backs of nude BALb/C mice (male, 20 g, five weeks old, OrientBio Co. Ltd. Seongnamsi, Gyeonggido, Korea). The KB cell-bearing mice were used for fluorescence imaging when the diameter of the tumor xenograft became larger than 6 mm. Nanophotosensitizer solution (10 mg Ce6/kg) was intravenously (i.v.) injected via the tail veins of the mice (injection volume: 200 µL). The whole bodies of the mice were observed with a Maestro TM 2 small animal imaging instrument (Cambridge Research and Instruments, Inc. Woburn, MA, USA). For the fluorescence imaging of the tumors, the mice were anesthetized with avertin for whole-body imaging, and the organs were extracted to observe the organ distribution of the COSthPBAP nanophotosensitizers.
According to the Ethical Program of the Pusan National University Institutional Animal Care and Use Committee (PNUIACUC), the mice were treated with CO 2 (20%) and then the fluorescence images of the whole body and each organ were observed. The organs were extracted from the mice bodies and then fluorescence images were observed using a Maestro TM 2 small animal imaging instrument (Cambridge Research and Instruments, Inc. Woburn, MA, USA). For the anticancer PDT treatment, the mice were anesthetized with the avertin solution and then irradiated with PDT equipment. Water and feed were freely provided to the mice. The mice were kept in the cage (3~4 mice/cage, Cage size: 200 mm × 260 mm × 130 mm (W × D × H)).

Statistical Analysis
The statistical significance of the results was evaluated by Student's t-test using SigmaPlot ® (SigmaPlot ® v.11.0, Systat Software, Inc., San Jose, CA, USA), and p < 0.05 was evaluated as the minimal level of significance.

Synthesis and Characterization of COSthPBAP Copolymer
To make ROS-sensitive nanophotosensitizers, COSthPBAP conjugates were synthesized as shown in Figure 1. As shown in in Figure 1, one end of the carboxylic acid of ThdCOOH was activated with an EDAC/NHS system and then conjugated with PBAP to produce ThdCOOH-PBAP conjugates. Each specific peak of the methyl groups of the Thd-COOH ("a" in Figure 1) and PBAP ("b" in Figure 1) was confirmed at around 1.5~1.6 ppm and 1.3 ppm, respectively ( Figure S1a,b). Specific peaks of glucosamine of COS were between 2.0 ppm and 5.0 ppm ( Figure S1c), i.e., protons of carbon 1~6 ("1~6" of Figure 1) was confirmed. Another carboxylic acid end of the ThdCOOH-PBAP conjugates was activated with EDAC/NHS again and then conjugated with COS to produce COSthPBAP conjugates, as shown in Figure 1. The chemical structure and 1 H NMR spectra of the COSthPBAP are shown in Figure S2a,b. As shown in Figure S2b, each specific proton peak of the COS, ThdCOOH, and PBAP was confirmed around 1.0 ppm~5.0 ppm, i.e., the specific peaks of the methyl protons of ThdCOOH and PBAP appeared at 1.8~1.9 ppm and 1.4~1.5 ppm, respectively. The peaks of COS also appeared at around 2.5~5.0 ppm. These results indicate that the COSthPBAP conjugates were successfully synthesized. The yield of the COSthPBAP conjugates was higher than 92.3% (w/w).

Synthesis and Characterization of COSthPBAP Copolymer
To make ROS-sensitive nanophotosensitizers, COSthPBAP conjugate synthesized as shown in Figure 1. As shown in in Figure 1, one end of the carbox of ThdCOOH was activated with an EDAC/NHS system and then conjugated wit to produce ThdCOOH-PBAP conjugates. Each specific peak of the methyl group ThdCOOH ("a" in Figure 1) and PBAP ("b" in Figure 1) was confirmed at around ppm and 1.3 ppm, respectively ( Figure S1a,b). Specific peaks of glucosamine of CO between 2.0 ppm and 5.0 ppm ( Figure S1c), i.e. protons of carbon 1 ~ 6 ("1 ~ 6" o 1) was confirmed. Another carboxylic acid end of the ThdCOOH-PBAP conjuga activated with EDAC/NHS again and then conjugated with COS to produce COS conjugates, as shown in Figure 1. The chemical structure and 1 H NMR spectr COSthPBAP are shown in Figure S2a,b. As shown in Figure S2b, each specific prot of the COS, ThdCOOH, and PBAP was confirmed around 1.0 ppm ~ 5.0 ppm, specific peaks of the methyl protons of ThdCOOH and PBAP appeared at 1.8 ~ and 1.4 ~ 1.5 ppm, respectively. The peaks of COS also appeared at around 2.5 ~ 5 These results indicate that the COSthPBAP conjugates were successfully synthesiz yield of the COSthPBAP conjugates was higher than 92.3% (w/w).

Characterization of Ce6-Incorporated COSthPBAP Nanophotosensitizers
Ce6-incorporated COSthPBAP nanophotosensitizers were prepared by the nanoprecipitation and dialysis method. The characteristics of the Ce6-incorporated COSthPBAP nanophotosensitizers were abbreviated as shown in Table 1. Higher Ce6 contents induced increases in the particle sizes of the nanophotosensitizers. Furthermore, the particle sizes were significantly increased compared to the empty nanoparticles. The Ce6 contents increased with increases in the feeding weight, but the loading efficiency was slightly decreased, as shown in Table 1.  1 Drug content (w/w, %) = (Ce6 weight/total weight of nanophotosensitizers) × 100.; Loading efficiency (w/w, %) = (Ce6 weight in the nanophotosensitizers/feeding weight of Ce6) × 100. 2 Particle sizes were average ± standard deviation from three different measurements. 3 Zeta potential was average ± standard deviation from three different measurements.
As shown in Figure 2a, the COSthPBAP nanoparticles had nano-spherical shapes and small diameters of less than 200 nm. Their size distributions showed a monomodal distribution pattern (Figure 2b). Their average particle sizes were about 146 nm (Table 1).  Figure 2a, the COSthPBAP nanoparticles had nano-spherical shapes a small diameters of less than 200 nm. Their size distributions showed a monomo distribution pattern (Figure 2b). Their average particle sizes were about 146 nm (Table  Table 1). Figure 3 shows the UV spectra of the Ce6, nanophotosensitizers, and em nanoparticles in water (deionized water, DW), DMSO, and/or a DMSO-DW mixture. shown in Figure 3a, no specific peaks of free Ce6 were observed in the DW because it h very low solubility, while it shows specific peaks between 300 nm and 700 nm in DMSO (Figure 3b). The nanophotosensitizers did not have specific peaks higher than nm, as shown in Figure 3c. When the nanophotosensitizers were dissolved in the DMS DW mixture, the UV spectra of the nanophotosensitizers show almost the same p specificity as the free Ce6, as shown in Figure 3d. The empty nanoparticles had spec peaks lower than 400 nm in only the DW and the DMSO-DW mixture (Figure 3e,f). Th  Table 1). Figure 3 shows the UV spectra of the Ce6, nanophotosensitizers, and empty nanoparticles in water (deionized water, DW), DMSO, and/or a DMSO-DW mixture. As shown in Figure 3a, no specific peaks of free Ce6 were observed in the DW because it had very low solubility, while it shows specific peaks between 300 nm and 700 nm in the DMSO (Figure 3b). The nanophotosensitizers did not have specific peaks higher than 400 nm, as shown in Figure 3c. When the nanophotosensitizers were dissolved in the DMSO-DW mixture, the UV spectra of the nanophotosensitizers show almost the same peak specificity as the free Ce6, as shown in Figure 3d. The empty nanoparticles had specific peaks lower than 400 nm in only the DW and the DMSO-DW mixture (Figure 3e,f). These results indicate that the COSthPBAP polymers did not affect the intrinsic properties of the free Ce6 during the nanophotosensitizer fabrication process.
To study whether or not the nanophotosensitizers had ROS sensitivity, the nanophotosensitizers were incubated in the presence of H 2 O 2 , as shown in   To study whether or not the nanophotosensitizers had ROS sensitivity, the nanophotosensitizers were incubated in the presence of H2O2, as shown in Figure 4. The particle size distribution became multimodal and/or irregular (Figure 4b,c) in the presence of H2O2, i.e., the particle size distribution had a multimodal pattern with 0.5 mM of H2O2 and had an irregular distribution pattern with 2.0 mM of H2O2, while they maintained a monomodal distribution pattern with 0 mM of H2O2 (Figure 4a). The morphological observation with TEM also showed that the nanophotosensitizers were disintegrated in the presence of H2O2 (Figure 4e,f) while they maintained spherical shapes in the absence of H2O2 as shown in Figure 4d. These results indicate that the COSthPBAP nanophotosensitizers had ROS sensitivity and they were disintegrated by ROS.    Figure 5 shows the generation of SO from free Ce6 and nanophotosensitizers. As shown in Figure 5, the fluorescence intensity time-dependently increased both the free Ce6 and nanophotosensitizers under light irradiation, while changes in the fluorescence  Figure 5 shows the generation of SO from free Ce6 and nanophotosensitizers. As shown in Figure 5, the fluorescence intensity time-dependently increased both the free Ce6 and nanophotosensitizers under light irradiation, while changes in the fluorescence intensity were negligible in the absence of light irradiation. In addition, the fluorescence intensity of the nanophotosensitizers with light irradiation was more than two times higher than that of Ce6, indicating that the nanophotosensitizers efficiently produced ROS in the aqueous environment.  Figure 5 shows the generation of SO from free Ce6 and nanophotosensitizers. As shown in Figure 5, the fluorescence intensity time-dependently increased both the free Ce6 and nanophotosensitizers under light irradiation, while changes in the fluorescence intensity were negligible in the absence of light irradiation. In addition, the fluorescence intensity of the nanophotosensitizers with light irradiation was more than two times higher than that of Ce6, indicating that the nanophotosensitizers efficiently produced ROS in the aqueous environment.  Figure 6 shows the changes in the fluorescence spectra and fluorescence images in the presence of H2O2. As shown in Figure 6a, the fluorescence intensity of the aqueous solution of the nanophotosensitizers gradually increased with H2O2, indicating that the nanophotosensitizers had ROS sensitivity and were able to respond to oxidative stress. Furthermore, fluorescence intensity in the images was also increased according to the H2O2 concentration (Figure 6b).  Figure 6 shows the changes in the fluorescence spectra and fluorescence images in the presence of H 2 O 2 . As shown in Figure 6a, the fluorescence intensity of the aqueous solution of the nanophotosensitizers gradually increased with H 2 O 2 , indicating that the nanophotosensitizers had ROS sensitivity and were able to respond to oxidative stress. Furthermore, fluorescence intensity in the images was also increased according to the H 2 O 2 concentration (Figure 6b).  Figure 7 shows the Ce6 released from the nanophotosensitizers. As shown in Figure  7a, the Ce6 release rate from the nanophotosensitizers was lower with a high Ce6 content, while the Ce6 release rate was faster with lower contents, indicating that the hydrophobic properties of Ce6 might have hydrophobically interacted in the core of the nanoparticles  Figure 7 shows the Ce6 released from the nanophotosensitizers. As shown in Figure 7a, the Ce6 release rate from the nanophotosensitizers was lower with a high Ce6 content, while the Ce6 release rate was faster with lower contents, indicating that the hydrophobic properties of Ce6 might have hydrophobically interacted in the core of the nanoparticles and then dissolved slowly. When H 2 O 2 was added to the release media, the Ce6 release rate from the nanophotosensitizers was significantly faster with the H 2 O 2 in a dose-dependent manner, as shown in Figure 7b. These results indicate that the Ce6 was released from the nanophotosensitizers in a ROS-sensitive manner.
(b) of nanophotosensitizers. Nanophotosensitizers were incubated for 3 h at 37 °C in the absence or presence of H2O2. 1. H2O2, 0 mM; 2. H2O2, 0.5 mM; 3. H2O2, 2.0 mM. The Ce6 concentration of the nanophotosensitizers in PBS was 0.05 mg/mL. Figure 7 shows the Ce6 released from the nanophotosensitizers. As shown in Figure  7a, the Ce6 release rate from the nanophotosensitizers was lower with a high Ce6 content, while the Ce6 release rate was faster with lower contents, indicating that the hydrophobic properties of Ce6 might have hydrophobically interacted in the core of the nanoparticles and then dissolved slowly. When H2O2 was added to the release media, the Ce6 release rate from the nanophotosensitizers was significantly faster with the H2O2 in a dosedependent manner, as shown in Figure 7b. These results indicate that the Ce6 was released from the nanophotosensitizers in a ROS-sensitive manner.

Cell Culture Study In Vitro
Prior to testing the PDT efficacy against oral cancer cells, the intrinsic cytotoxicity of free Ce6 and nanophotosensitizers were evaluated as a means of dark toxicity ( Figure 8). As shown in Figure 8a-c, both the Ce6 and nanophotosensitizers had low cytotoxicity until a 2 μg/mL Ce6 concentration against YD-38 cells, KB cells, and SCC-15 cells, i.e., the viabilities of YD-38 cells, KB cells, and SCC-15 cells were higher than 80% until a 2 μg/mL Ce6 concentration of the nanophotosensitizers and free Ce6. Furthermore, the nanophotosensitizers also had low intrinsic dark toxicity against HGF-1 human gingival fibroblast cells until 2 μg/mL of Ce6 and free Ce6 as shown in Figure 8d. These results

Cell Culture Study In Vitro
Prior to testing the PDT efficacy against oral cancer cells, the intrinsic cytotoxicity of free Ce6 and nanophotosensitizers were evaluated as a means of dark toxicity ( Figure 8). As shown in Figure 8a-c, both the Ce6 and nanophotosensitizers had low cytotoxicity until a 2 µg/mL Ce6 concentration against YD-38 cells, KB cells, and SCC-15 cells, i.e., the viabilities of YD-38 cells, KB cells, and SCC-15 cells were higher than 80% until a 2 µg/mL Ce6 concentration of the nanophotosensitizers and free Ce6. Furthermore, the nanophotosensitizers also had low intrinsic dark toxicity against HGF-1 human gingival fibroblast cells until 2 µg/mL of Ce6 and free Ce6 as shown in Figure 8d. These results indicate that the nanophotosensitizers and free Ce6 had low toxicity in the absence of light irradiation conditions. At 5 µg/mL, the free Ce6 and nanophotosensitizers resulted in less than 80% cell viability against AT84 cells and HGF-1 cells, respectively. The free Ce6 and nanophotosensitizers were tested until a 2 µg/mL Ce6 concentration for the next experiment. The results of the dark toxicity test indicate that the absence of light irradiation did not significantly affect the viability of the tumor cells or normal cells in either the free Ce6 or the nanophotosensitizers. irradiation conditions. At 5 μg/mL, the free Ce6 and nanophotosensitizers resulted in less than 80% cell viability against AT84 cells and HGF-1 cells, respectively. The free Ce6 and nanophotosensitizers were tested until a 2 μg/mL Ce6 concentration for the next experiment. The results of the dark toxicity test indicate that the absence of light irradiation did not significantly affect the viability of the tumor cells or normal cells in either the free Ce6 or the nanophotosensitizers.  Figure 9 shows the relative Ce6 uptake ratio of the cancer cells. As shown in Figure  9a-c, the Ce6 uptake ratio dose-dependently increased in all cancer cells, including the YD-38 cells, KB cells, and SCC-15 cells, for both the free Ce6 and the nanophotosensitizers. Specifically, the Ce6 uptake ratio of the nanophotosensitizers was significantly higher than those of the free Ce6. These results indicate that the nanophotosensitizers had the superior potential for intracellular delivery. In the morphological observation of the YD-38 cells, the nanophotosensitizers revealed significantly stronger fluorescence intensity compared to treatment with the free Ce6 (Figure 9d). These results indicate that the nanophotosensitizers were internalized in the cells efficiently.  Figure 9 shows the relative Ce6 uptake ratio of the cancer cells. As shown in Figure 9a-c, the Ce6 uptake ratio dose-dependently increased in all cancer cells, including the YD-38 cells, KB cells, and SCC-15 cells, for both the free Ce6 and the nanophotosensitizers. Specifically, the Ce6 uptake ratio of the nanophotosensitizers was significantly higher than those of the free Ce6. These results indicate that the nanophotosensitizers had the superior potential for intracellular delivery. In the morphological observation of the YD-38 cells, the nanophotosensitizers revealed significantly stronger fluorescence intensity compared to treatment with the free Ce6 (Figure 9d). These results indicate that the nanophotosensitizers were internalized in the cells efficiently. Figure 10 shows the relative ROS generation (Figure 10a-c) and PDT efficacy (Figure 10d-f) of the free Ce6 and nanophotosensitizers. As shown in Figure 10a-c, the relative ROS generation dose-dependently increased according to the Ce6 concentrations of both the free Ce6 and the nanophotosensitizers in all cancer cells, including the YD-38 cells (Figure 10a), KB cells (Figure 10b), and SCC-15 cells (Figure 10c). Furthermore, the nanophotosensitizers resulted in a higher ROS generation compared to the free Ce6, indicating that the nanophotosensitizers had a superior potential for generating intracellular ROS in cancer cells. Figure 10d-f shows the PDT efficacy of the free Ce6 and nanophotosensitizers against the YD-38 cells (Figure 10d), KB cells (Figure 10e), and SCC-15 cells (Figure 10f). As expected, the nanophotosensitizers showed higher phototoxicity against the YD-38 cells, KB cells, and SCC-15 cells compared to the free Ce6. These results indicate that the nanophotosensitizers had a higher potential for cellular uptake, ROS generation, and PDT efficacy compared to the free Ce6.   (Figure 10c). Furthermore, the nanophotosensitizers resulted in a higher ROS generation compared to the free Ce6, indicating that the nanophotosensitizers had a superior potential for generating intracellular ROS in cancer cells. Figure 10d-f shows the PDT efficacy of the free Ce6 and nanophotosensitizers against the YD-38 cells (Figure 10d), KB cells (Figure 10e), and SCC-15 cells (Figure 10f). As expected, the nanophotosensitizers showed higher phototoxicity against the YD-38 cells, KB cells, and SCC-15 cells compared to the free Ce6. These results indicate that the nanophotosensitizers had a higher potential for cellular uptake, ROS generation, and PDT efficacy compared to the free Ce6.

Animal Tumor Imaging of Tumor Xenograft Model
To investigate whether or not the nanophotosensitizers could target tumors, KB cells were implanted into the backs of nude mice, and then nanophotosensitizers were intravenously (i.v.) administered through the tail veins of mice.
As shown in Figure 11a,b, strong fluorescence intensity in the tumor tissues was

Animal Tumor Imaging of Tumor Xenograft Model
To investigate whether or not the nanophotosensitizers could target tumors, KB cells were implanted into the backs of nude mice, and then nanophotosensitizers were intravenously (i.v.) administered through the tail veins of mice.
As shown in Figure 11a,b, strong fluorescence intensity in the tumor tissues was observed, i.e., red fluorescence was observed in the field of the tumor xenograft. As shown in Figure 11c,d, the fluorescence images of each organ show that the tumor revealed the strongest fluorescence intensity compared to the other organs. These results indicate that the nanophotosensitizers had the ability to target the tumors and then efficiently accumulated in the tumor tissues. This result may indicate that oral cancer can be imaged by the administration of COSthPBAP nanophotosensitizers and then efficiently cured by PDT.

Animal Tumor Imaging of Tumor Xenograft Model
To investigate whether or not the nanophotosensitizers could target tumors, KB ce were implanted into the backs of nude mice, and then nanophotosensitizers w intravenously (i.v.) administered through the tail veins of mice.
As shown in Figure 11a,b, strong fluorescence intensity in the tumor tissues w observed, i.e., red fluorescence was observed in the field of the tumor xenograft. As sho in Figure 11c,d, the fluorescence images of each organ show that the tumor revealed strongest fluorescence intensity compared to the other organs. These results indicate t the nanophotosensitizers had the ability to target the tumors and then efficien accumulated in the tumor tissues. This result may indicate that oral cancer can be imag by the administration of COSthPBAP nanophotosensitizers and then efficiently cured PDT.

Discussion
The traditional treatment regimen for oral cancer is currently considered to have limited efficacy due to delays in diagnosis and treatment [3,45]. Lauritzen et al. reviewed that delays in diagnosis have a significant correlation with the progression of the tumor stages, and the time from diagnosis to the treatment of oral cancer is significantly related to the survival of oral cancer patients [3]. Specifically, almost all cancers in the oral cavity and oropharynx are typically squamous cell carcinomas, which are flat and thin cells [1]. These cell types form the lining of the oral cavity, and these properties of oral cancers make it difficult to deliver anticancer drugs or biological therapeutics to oral cancer tissues [46].
Although various therapeutic regimens, such as surgery, chemotherapy, radiotherapy, and immunotherapy have been attempted to treat oral cancers, single and/or combined treatments are still problematic because high recurrence rates after these treatments have been reported, leading to a low survival rate of patients with oral cancers [47][48][49]. For these reasons, PDT can be considered a promising candidate for the treatment of oral cancers because PDT is suitable to be applied to squamous carcinoma types [50]. That is, the depth limit of light irradiation is known to be less than 2 mm, and thus, oral squamous carcinoma is suitable for light irradiation [17,51]. Furthermore, PDT using photosensitizers can be utilized to fluorescently detect and diagnose tumors in the oral cavity [52]. In addition, the unwanted side-effects of PDT, such as sun-shade problems, are always problematic in PDT approaches for cancers [28]. In a clinical application, talaporfin sodium-based PDT showed the safe regression of esophageal cancer against local failure after chemoradiotherapy [27].
Nanoparticle-based PDT, such as nanophotosensitizers, is believed to be a promising candidate for the tumor-specific delivery of photosensitizers [38,39,53]. Since nanodimensional carriers can be accumulated in the tumor tissue via the enhanced permeation and retention (EPR) effect, nano-based medicine has been extensively investigated for cancer therapy [53][54][55]. Specifically, tumor microenvironments are quite different compared to normal tissues and cells [55,56]. The redox potential of tumor microenvironments is known to be elevated and can be applicable for tumor-targeting using nano-based medicine [57,58]. Mirhadi et al. reported that redox-sensitive nanomedicine can be used to target cancer cells and emphasize the anticancer activity of therapeutics [58]. In our results, COSthPBAP nanophotosensitizers released Ce6 in an ROS-sensitive manner through the ROS-sensitive disintegration of the COSthPBAP nanophotosensitizers, as shown in Figures 4, 6 and 7. The COSthPBAP nanophotosensitizers responded to oxidative stress in the presence of H 2 O 2 and then were efficiently delivered to cancer cells as free Ce6. These trends of nanophotosensitizers are related to the generation of ROS and PDT efficacy, i.e., the COSth-PBAP nanophotosensitizers showed superior ROS generation and PDT efficacy against oral cancer cells with low dark toxicity against normal HGF-1 cells (Figures 8-10). Photosensitizers, such as 5-amino levulinic acid (5-ALA), can be used to improve the diagnostic contrast/accuracy of oral cancers through the fluorescence detection of anatomic locations of the oral cavity [59]. PDT treatment against oral leukoplakia lesions and oral lichen planus lesions showed positive results [60]. However, the low tumor specificity of traditional photosensitizers frequently results in dispersion in normal cells or tissues, and these properties are related to the side effects of traditional photosensitizers. COSthPBAP nanophotosensitizers can be specifically delivered to tumor tissue, i.e., the fluorescence intensity was the strongest in the tumor tissues compared to other organs (Figure 11), indicating that COSthPBAP nanophotosensitizers can be delivered to tumor tissues specifically.
As illustrated in Figure 12, the degradation of the thioketal group and PBAP moiety is known to be affected by the presence of ROS [61][62][63]. Lee et al. also reported that phenyl boronic acid can be degraded in the presence of ROS, such as H 2 O 2 , and, after that, acid release by H 2 O 2 catalyzes the hydrolysis of the polymer backbone [61]. These behaviors induce changes in the polymer properties from hydrophobic to hydrophilic. Gao and Xiong also showed that the thioketal group can be degraded by reactive oxygens and then degraded into thiol groups [62]. PBAP moieties and thioketal linkers in the polymer backbone resulted in hydrophobic-hydrophilic changes and degradation by ROS-sensitive behavior [63]. These changes accelerated the release rate of bioactive agents and anticancer drugs and then emphasized their anticancer activities. Our results also indicate that the drug release rate was accelerated by H 2 O 2 and then affected the ROS production/PDT effect, as shown in Figures 6-10. degraded into thiol groups [62]. PBAP moieties and thioketal linkers in the polymer backbone resulted in hydrophobic-hydrophilic changes and degradation by ROSsensitive behavior [63]. These changes accelerated the release rate of bioactive agents and anticancer drugs and then emphasized their anticancer activities. Our results also indicate that the drug release rate was accelerated by H2O2 and then affected the ROS production/PDT effect, as shown in Figures 6-10. Nanomaterials, polymer conjugates, and/or nanoparticles are known to accelerate the generation of singlet oxygen rather than free Ce6 [44,64,65]. For example, Park and Na reported the measurement of Ce6-pluronic F127 conjugates generating singlet oxygen using SOSG reagent was significantly better compared to free Ce6 [44]. They argued that the singlet oxygen generation of Ce6-pluronic F127 conjugates was five times higher than that of free Ce6 due to the improved aqueous solubility against distilled water. Nanomaterials, such as carbon nanotubes, are also reported to enhance singlet oxygen generation [64]. Nanoparticles based on polymers can be considered an ideal vehicle to improve aqueous solubility, photostability, and photo dynamic activity [65]. Our results also indicate that the COSthPBAP nanophotosensitizers resulted in higher singlet oxygen generation than free Ce6 ( Figure 5).

Conclusions
COSthPBAP copolymers were synthesized for Ce6 delivery against oral cancer cells. ThdCOOH was conjugated with PBAP to produce ThdCOOH-PBAP conjugates and then added to the amine groups of COS to produce COSthPBAP copolymers. Ce6-incorporated Nanomaterials, polymer conjugates, and/or nanoparticles are known to accelerate the generation of singlet oxygen rather than free Ce6 [44,64,65]. For example, Park and Na reported the measurement of Ce6-pluronic F127 conjugates generating singlet oxygen using SOSG reagent was significantly better compared to free Ce6 [44]. They argued that the singlet oxygen generation of Ce6-pluronic F127 conjugates was five times higher than that of free Ce6 due to the improved aqueous solubility against distilled water. Nanomaterials, such as carbon nanotubes, are also reported to enhance singlet oxygen generation [64]. Nanoparticles based on polymers can be considered an ideal vehicle to improve aqueous solubility, photostability, and photo dynamic activity [65]. Our results also indicate that the COSthPBAP nanophotosensitizers resulted in higher singlet oxygen generation than free Ce6 ( Figure 5).

Conclusions
COSthPBAP copolymers were synthesized for Ce6 delivery against oral cancer cells. ThdCOOH was conjugated with PBAP to produce ThdCOOH-PBAP conjugates and then added to the amine groups of COS to produce COSthPBAP copolymers. Ce6-incorporated nanophotosensitizers using the COSthPBAP copolymers were fabricated using the nanoprecipitation and dialysis methods. The Ce6-incorporated COSthPBAP nanophotosensitizers had small diameters of less than 200 nm, with a mono-modal distribution pattern. However, they became multimodal and/or irregular distribution patterns when H 2 O 2 was added. In the morphological observation using TEM, the nanophotosensitizers were disintegrated by the addition of H 2 O 2 , indicating that the COSthPBAP nanophotosensitizers had ROS sensitivity. In addition, the Ce6 release rate from the COSthPBAP nanophotosensitizers accelerated in the presence of H 2 O 2 . SO generation was higher in the nanophotosensitizers than in the free Ce6. Furthermore, the COSthPBAP nanophotosensitizers showed a higher intracellular Ce6 uptake ratio and ROS generation in all types of oral cancer cells. They also efficiently inhibited the viability of oral cancer cells under light irradiation, but they did not significantly affect the viability of either normal cells or cancer cells in the absence of light irradiation. The COSthPBAP nanophotosensitizers showed a tumor-specific delivery capacity and fluorescence imaging of KB tumors in an in vivo animal tumor imaging study. We suggest that COSthPBAP nanophotosensitizers are promising candidates for the imaging and treatment of oral cancers.

Institutional Review Board Statement:
The animal study was strictly performed according to the guidelines of the Pusan National University Institutional Animal Care and Use Committee (PNUIACUC). The protocol of the animal experiments was reviewed and monitored by the PNUIACUC on their ethical procedures and scientific care and approved (approval number: PNU-2020-2751). We obtained the approval number from PNUIACUC in October 2021, and the approval period for the animal experiment was from 1 September 2020 to 31 August 2021. The animal study was conducted during this period.