Sulphur‐doped carbon dots as a highly efficient nano‐photodynamic agent against oral squamous cell carcinoma

Abstract Objectives Photodynamic therapy (PDT) is a novel non‐invasive therapeutic method, which has been widely applied for the treatment of human oral cancers. However, the problems of undesirable singlet oxygen (1O2) quantum yields and long‐term phototoxicity were inevitable during the application of traditional photosensitizers. Therefore, it is necessary to explore novel photosensitizers for the improvement of therapeutic effects. In our study, the sulphur‐doped carbon dots (S‐CDs) of high yield of singlet oxygen (1O2) were synthesized as a nano‐photosensitizer for OSCC to improve the PDT efficacy in clinical practice. Materials and methods After synthesis of the novel S‐CDs, the size, morphologic characteristics, surface potential and yield of singlet oxygen (1O2) were determined. In vitro study was performed to compare the therapeutic effect as well as the biocompatibility of the novel S‐CDs to those of 5‐ALA. Besides, possible mechanism of action was illustrated. Results After synthesis of the novel S‐CDs, the size, morphologic characteristics, surface potential and yield of singlet oxygen (1O2) were determined. In vitro study was performed to compare the therapeutic effect as well as the biocompatibility of the novel S‐CDs to those of 5‐ALA. Besides, possible mechanism of action was illustrated. Conclusions These data from the in vitro study demonstrated the promising safety profile of the low dose (nmol/L) S‐CDs, which indicated the novel S‐CDs could be used as a promising photodynamic agent for oral cancer therapy.


| INTRODUC TI ON
Squamous cancer is a common malignant oral carcinoma located in the epidermal or adnexal cells. To date, the treatment methods for squamous cell carcinoma mainly includes surgery, freezing and radiotherapy, which can either directly remove or destroy tumour tissues. However, due to the non-negligible side effects, large wound and potential psychological disorders, the application of the above conventional methods could be limited. Recently, a simple non-invasive and simple therapeutic strategy, namely photodynamic therapy (PDT), has been widely applied in clinical cancer treatments. 1,2 Generally, the toxic effect of the reactive oxygen species (ROS) generated by transferring energy from light to oxygen molecules can directly kill tumour cells through photosensitizers. 3 Besides, ROS can damnify the vasculature of tumour parenchyma, resulting in blood supply deficiency, 4,5 and further provoke the anti-tumour immune response further. 3,6 What is more, even repeated use of PDT can achieve satisfactory therapeutic outcome without causing obvious side effects or any scar formation. 7 Due to the superficial nature of oral and skin tumours, tumour tissue is located in an area that is easy to receive light, which facilitates the efficient implementation of PDT. 7 Therefore, PDT has been regarded as an alternative or supplementary therapeutic method of the traditional regimens for the treatment of superficial tumours.
Typically, PDT is a three-way process consisting of photosensitizer (PS), light and molecular oxygen. 8 The concentration of ROS directly affects the efficiency of PDT, which largely relies on the photo-oxidation ability of the PSs. 9 Over the last few decades, photosensitizers in clinical use are the first and second generation of organic PSs have been approved for clinical use, such as haematoporphyrin (Hp), photofrin and chlorin e6 (Ce6), among which 5-aminolevulinic acid (ALA)-mediated PDT has been widely applied in clinical practice with satisfactory outcomes. 8,10,11 5-ALA is the biological precursor of protoporphyrin IX (PpIX). Several previous studies 12- 14 have demonstrated its superiority in the treatment of both oral pre-cancer and oral cancer treatment since it induces the accumulate high concentration of PpIX in cancer cells. However, certain deficiencies, such as 5-ALA suffered from long-term phototoxicity and poor cell penetration, would lower the PDT therapeutic effect, thus limiting the clinical application of 5-ALA. Therefore, the third generation of photosensitizer came into being to solve the problems of the first two generations.
Wang and his co-workers 15 recently developed a nanosystem by combining chemotherapy with PDT for the targeted treatment of OTSCC, and they found that CAL-27 cells assimilated PEG linked haematoporphyrin (HP) and depicted high PDT efficiency through the target-peptide-mediated cell internalization. Although the PDT efficiency has been assembly promoted by using biomolecular modification and nano-materials technologies, certain problems, such as poor biocompatibility, weak selectivity and fluorescence quenching under light, 16 are still inevitable.
As the fourth generation of PSs, nano-photosensitive agents possess excellent characteristics of adjustable excitation and emission wavelength, strong anti-photobleaching ability, and good biocompatibility. 17 Carbon-based nano-materials, such as carbon nanotubes, 18,19 graphene 20 and carbon dots, [21][22][23] have been drawing more and more attention in the fields of cancer diagnosis and therapy. In this study, the sulphur-doped carbon dots (S-CDs) 24 with a high single oxygen quantum yield were prepared for the PDT treatment for oral squamous cell carcinoma. The S-CDs could automatically enter cancer cells and instigate potent trigger cancer cell death by light.
Moreover, by detecting the expression of apoptotic proteins, we also observed that the UM1 cells treated with S-CDs had higher expression levels of apoptotic proteins compared with those treated with classic photosensitizer 5-ALA at the same concentration (Figure 1), indicating the promising application prospect of the developed nano-photosensitizer in the PDT treatment for oral cancer treatment.

| Preparation and characterization of S-CDs
Sulfur-doped carbon dots were prepared via a hydrothermal process using polythiophene (PT2) as the precursor that developed by Wang et al 24 The detailed synthesis steps are as follows: PT2 (30 mg) and NaOH solution (40 mL, 0.5 mmol/L) were firstly mixed for ultrasound for 30 minutes, and then transferred the mixture into an autoclave and maintained the temperature at 170°C. After reaction for 24 hours, the product were purified with 0.22-µm membranes to remove residue and finally dissolved in water.

| Cell culture
UM1 cell line originated from a HNSCC-diagnosed patient. 25,26 These cell lines were gifts from Dr Chen (Sichuan University). UM1 cells were incubated in DMEM, supplemented with 10% FBS, 100 U/ mL penicillin and 100 µg/mL streptomycin (GIBCO). The culture media was renewed twice or three times a week. Cells were put in incubator with temperature at 37°C and 5% CO 2 .

| S-CD uptake and in vitro imaging
Cell suspension was removed, density of 1 × 10 5 /mL, from culture flask and then transferred to 12-well plates, and 1 µmol/L S-CDs were added for 0 hour, 3 hours, 6 hours, 12 hours, 24 hours and 48 hours, respectively. The trypsin-treated UM1 cells were clustered by centrifugation, and then washed with PBS and finally resuspended into 400 µL PBS for flow cytometry test (Attune NxT, Life).
In addition, to observe cellular assimilation state and obtain images, confocal laser microscope was employed. Cells, rinsed by PBS, were fixed with 4% cold paraformaldehyde for 15 minutes. After three times' PBS rinsing, phalloidin and DAPI were used to tint the cytoskeleton and nucleus, respectively.

UM1 cells after incubation with S-CDs were loaded with
MitoTracker green (a mitochondria probe) and LysoTracker green (a lysosome probe), respectively. Samples were washed with PBS after 30 minutes incubation. Analysed photographs were taken from confocal laser scanning microscopy (AIR-MP, Nikon).

| Determine reactive oxygen species (ROS)
The detected green fluorescence, produced by dichlorofluorescein Additionally, we performed flow cytometry to record ROS fluorescence signal of each group by cell suspension that had hatched with DCFH-DA detector. Every group has three accessory holes.

| AO/PI test
Cell viability could also be detected through acridine orange-propidium iodide (AO/PI) kit. 28 The UM1 cells were cultured in 12-well plates with the protocol described before (Western blot test). The AO/PI stain procedure briefly is keeping the working solution, containing AO: 670 μmol/L, PI: 750 μmol/L, with cell in the dark at 4°C for 20 minutes and observed each well via fluorescent microscope.
Live cells turned into green (AO), whereas the dead appeared red (PI).

| Calcium concentration (Ca 2+ ) detection
Experimental group and cell treatment were the same as ROS assay.
What was slightly different was that the cells were cultured with Fluo-4/AM (1 µmoL) (Beyotime) 29 in the incubator for 30 minutes.
When entering into cell and meet Ca 2+ , Fluo-4/AM would transfer to a strongly fluorescent compound. Subsequently, images were captured on fluorescent inverted microscope.

| Western blotting
Growth-arrested UM1 cells were divided into three groups: the control group, ALA (10 nmol/L) with irradiation group and S-CDs (10 nmol/L) with irradiation group. Then, all groups were harvested 12 hours after the treatment. The whole-cell lysis assay kit (KeyGen) was used to extract proteins. Protein was denatured in sodium dodecyl sulphate (SDS) buffer at 100°C for 5 minutes, dissociated by 12% SDS-PAGE afterwards and transferred to PVDF membranes. Then, target proteins were incubated with the correspondent primary antibody Bax, Bcl-2 and caspase-3 on the first day and secondary F I G U R E 1 Illustration for the S-CDs and 5-ALA-mediated PDT in UM1 cells antibody (Beyotime) on the coming day. Subsequently, developer (Bio-Rad) was added to show the target bands. Band quantification was carried out with the ImageJ software.

| Immunofluorescence
To make an entry for antibody to go through the sample cells: fixed by room-temperatured 4% (w/v) paraformaldehyde for 10 minutes at the punctual time post-PDT, were punched by room-temperatured 0.5% Triton X-100 for additional 10 minutes. Primary antibodies against Bax, Bcl-2 and caspase-3 and secondary antibody were applied to form a camera-detectable fluorescent conjunction afterwards. Then after PBS rinsing, slides were dyed by cytoskeleton and nuclei stainer, and images were captured under confocal laser scanning microscope (AIR-MP). SPSS 18.0 provided the function of t test to tell the differences between each two groups. It is when the P value is <.05 that the variant would be considered significant.

| Characterization of the S-CDs
The S-CDs were synthesized as a novel nano-photosensitizer for oral squamous cell carcinoma based on a polythiophene precursor Meanwhile, an emission band centred at ∼600 nm was obtained, which can be ascribed to the sulphur doping. Furthermore, when the S-CDs was excited at 420 nm, the characteristic phosphorescence of singlet oxygen at 1275 nm can be observed from S-CDs ( Figure 2F).

| Cellular localization of the S-CDs
Previous reports 30,31 have verified that the curative effect of PDT primarily relied on the cellular uptake and localization of the photosensitizers. Considering the red fluorescence emission of the S-CDs (∼600 nm). Therefore, the uptake of S-CDs by UM1 cells was firstly characterized by using flow cytometry and confocal laser scanning microscopy. As shown in Figure A1A, during the first 24 hours of incubation, the red fluorescent signal of S-CDs was significantly increased and then slightly increased for another 48 hours, which was consistent with the results from flow cytometer ( Figure A1B).
Besides, the positive charge on the surface of the S-CDs yielded promising cell uptake. After 24 hours of incubation, the S-CDs could be detected in about 90% of the UM1 cells. The confocal laser scanning microscopy suggested that the S-CDs were accumulated in both lysosomes ( Figure 3A) and mitochondria ( Figure 3B) after incubating for 2 hours. Ten hours later, the fusion signal of red and green fluorescence in lysosomes was brighter, indicating that the S-CDs mainly accumulated in lysosomes.

| Promising PDT efficacy of the S-CDs
In PDT, ROS could directly or indirectly damage cellular constituents through reacting with biological molecules, which plays a vital role in inducing cell apoptosis. 32 Thus, the production of ROS that triggered by S-CD-mediated PDT was examined with the DCFH-DA fluorescence assay. Bright green fluorescence was observed in the UM1 cells receiving S-CD-mediated PDT, while no obvious green fluorescent signal was discovered in the groups receiving S-CDs or 5-ALA without light irradiation as well as the group that merely received culture medium ( Figure 4A). Among the UM1 cells that were administrated with 5-ALA with light irradiation, only a faint green fluorescence can be detected ( Figure 4B). The results demonstrated that the capability to generate singlet oxygen of the S-CDs was higher than that of the 5-ALA, and the S-CD-mediated PDT had the highest singlet oxygen yield, which was further confirmed by the flow cytometry results. As shown in the Moreover, CCK-8 assay was applied to determine the cytotoxicity and PDT efficiency when the S-CDs was applied to perform PDT against UM1 cells. As illustrated in Figure 4C Under illumination, as soon as cellular uptake of the S-CDs occurred, the cell viability was immediately decreased, and the trend was consistent along with the increasing of S-CD concentration (5 nmol/L, 20%, 100 nmol/L, 5%) ( Figure 4C), while the 5-ALA group presented similar cell viability as the illumination-only group ( Figure 4D).
Besides, AO/PI assay was used to further evaluate the high photosensitive oxidation capacity of the S-CDs. The nucleic-acid-highly-affinitive AO penetrated membrane, causing a green fluorescent signals.
While the PI only stained the DNA and the RNA of dead/dying cells with red fluorescence. As shown in Figure A3, the S-CD group with light exposure experienced cell morphology change from spindle-like (control group) into round, and some of them even scattered into fragments forming red luminous points of varied sizes. In the 5-ALA group with light exposure, only slightly scattered red fluorescence could be detected and no drastic appearance change was observed in cell morphology, which was coincided with the CCK-8 results.

| S-CD mediated cell apoptosis
To further emphasize the perfect PDT performance of S-CDs for oral squamous cell carcinoma, the detailed intracellular responses to S-CD-mediated PDT were discussed. Firstly, as PDT treatment could raise the concentration of calcium (Ca 2+ ) in cytoplasm, resulting in cell death, 33,34 the alteration in Ca 2+ level during light exposure was evaluated by using Fluo-4 AM (a Ca 2+ probe). As illustrated in Figure   A4, the green fluorescence in S-CD-treated group with light irradiation was more remarkably than the control groups, while only a faint  In addition, studies have shown that ROS induced cell apoptosis through mitochondrial apoptotic pathway, causing the change of Bcl-2 family proteins (such as Bcl-2 and Bax), followed by activation of caspase. [35][36][37] To further trace the high therapeutic efficiency of S-CDs at biomolecular level, Western blots and immunofluorescence analysis were employed to investigate the Bcl-2, Bax and caspase-3 (apoptosis-related protein) after S-CD-mediated PDT. As shown in Figure 5A, the Western blot demonstrated a decreased relative fluorescence intensity of Bcl-2 in the group receiving S-CD-mediated PDT. Meanwhile, reversed trends were observed in terms of the Bax and caspase-3 proteins.
The confocal laser microscopy image revealed similar observation. In the group receiving S-CD-mediated PDT, the fluorophore intensity of the Bcl-2 was the faintest among the three study proteins ( Figure 5D), while the fluorescent signals of Bax and caspase-3 were significantly stronger than those of the control group ( Figure 5B,C).
With respect to 5-ALA at the same concentration, the fluorescent intensities of Bax and caspase-3 were weaker, but the signal of Bcl-2 was stronger than that of the S-CD group. The results further confirmed the high PDT efficiency of S-CDs. AM analysis indicated that the green fluorescent signal of the S-CD-mediated PDT group was brighter than that of the 5-ALAmediated PDT group, suggesting the high photosensitive oxidation capacity of nano-photosensitizer. Secondly, among the UM1 cell receiving S-CD PDT treatment, the Bcl-2 protein level was obviously decreased, while the level of Bax was significantly up-regulated. Therefore, we inferred that the minimum dosage S-CDs provoked apoptosis of UM1 cells via up-regulating the caspase-3

| D ISCUSS I ON
and Bax and decreasing the Bcl-2 level during the PDT. However, controlled investigations with 5-ALA at the same concentration induced minimal activation of the mitochondria apoptosis pathway. We expect this material could be used in bioimaging, targeted drug delivery and photodynamic therapy.
In conclusion, the novel S-CDs were synthesized as a nano-photosensitizer for oral squamous cell carcinoma therapy, which was proved to have satisfactory cell internalization ability and self-luminous, and depicted low cytotoxicity without the presence of light irradiation. Under the condition of light irradiation, the S-CDs may act as a more effective nano-weapon for anticancer therapy compared with traditional PS 5-ALA. The high therapeutic efficiency of the nano-structure was speculated to be realized by generating high rate of 1 O 2 , inducing acute stress response and Ca 2+ influx, and thereafter the overexpression of caspase-3 and Bax proteins as well as the down-regulation of Bcl-2 protein were triggered. Hence, the newly developed S-CDs might be a promising alternative photosensitizer for the treatment of oral-maxillofacial carcinoma by using PDT.

CO N FLI C T O F I NTE R E S T
The authors declare no potential conflicts of interest with respect to the authorship and/or publication of this article.

AUTH O R CO NTR I B UTI O N S
Qirong Li contributed to conception, design, data acquisition and interpretation, performed all statistical analyses, drafted the manuscript and critically revised the manuscript. Ronghui Zhou synthesize S-CDs, contributed to conception, design and interpretation, and critically revised the manuscript. Yu Xie draw the scheme picture and contributed to conception and design. Yanjing Li and Yu Chen contributed to conception and design and critically revised the manuscript. Xiaoxiao Cai contributed to conception and design, and critically revised the manuscript. All authors gave their final approval and agreed to be accountable for all aspects of the work.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.