Black Phosphorus Quantum Dots Enhance the Radiosensitivity of Human Renal Cell Carcinoma Cells through Inhibition of DNA-PKcs Kinase

Renal cell carcinoma (RCC) is one of the most aggressive urological malignancies and has a poor prognosis, especially in patients with metastasis. Although RCC is traditionally considered to be radioresistant, radiotherapy (RT) is still a common treatment for palliative management of metastatic RCC. Novel approaches are urgently needed to overcome radioresistance of RCC. Black phosphorus quantum dots (BPQDs) have recently received great attention due to their unique physicochemical properties and good biocompatibility. In the present study, we found that BPQDs enhance ionizing radiation (IR)-induced apoptotic cell death of RCC cells. BPQDs treatment significantly increases IR-induced DNA double-strand breaks (DSBs), as indicated by the neutral comet assay and the DSBs biomarkers γH2AX and 53BP1. Mechanistically, BPQDs can interact with purified DNA–protein kinase catalytic subunit (DNA-PKcs) and promote its kinase activity in vitro. BPQDs impair the autophosphorylation of DNA-PKcs at S2056, and this site phosphorylation is essential for efficient DNA DSBs repair and the release of DNA-PKcs from the damage sites. Consistent with this, BPQDs suppress nonhomologous end-joining (NHEJ) repair and lead to sustained high levels of autophosphorylated DNA-PKcs on the damaged sites. Moreover, animal experiments indicate that the combined approach with both BPQDs and IR displays better efficacy than monotreatment. These findings demonstrate that BPQDs have potential applications in radiosensitizing RCC cells.


Flow Cytometry
The 786-O cells were culture onto 60 mm 2 dishes. Cells were exposed to 20 µg/mL BPQDs along or in combination with 5 Gy IR for 24 h. Cells were harvested, centrifuged, and resuspended. Then, PE-Annexin V and 7-AAD were incubated with cell suspension in the dark condition. Cells were subjected to flow cytometry analysis using a flow cytometer (FACSVerse, BDBiosciences, San Diego, CA, USA).

Comet Assay for DNA Double-Strand Breaks
Cells were treated with 20 µg/mL of BPQDs for 12 h, and washed out with PBS, and were then irradiation with 4 Gy of X-ray. A duration of 4 h post IR, the cells were mixed with low melting point agarose, spread on a comet assay slide. Those slides were left into 4 • C for drying, and incubated with neutral lysis buffer and subjected to electrophoresis. Cells were stained with SYBR Green I, and comet tails were visualized using a confocal microscope (FV1200, Olympus, Tokyo, Japan). Experiments were performed at least three times for each sample.

NHEJ Repair Efficiency
The 786-O cells were cultured in 6-well plates, and 3 µg of CRISPR/Cas9 plasmid was co-transfected with 25 pmol dsODN into 786-O cells. BPQDs (20 µg/mL or 40 µg/mL) were added to the cells 24 h post transfection. Then, the genomic DNA of 786-O cells was extracted by a Genomic DNA Kit (Tiangen, Beijing, China) and real-time PCR (Applied Biosystems, Foster City, CA) was performed.

Animal Experiment
The 1.5-month-old male BALB/c nude mice (Shanghai SLAC Laboratory Animal Co. Ltd., Shanghai, China) were raised in the SPF-level laboratory animal room of Soochow University. A total of 1 × 10 7 786-O cells were injected subcutaneously into the flanks of nude mice. When the tumor grew to 50-100 mm 3 , the mice were treated with PBS or BPQDs (1 mg/kg), 10 Gy X-rays or BPQDs + X-rays. Radiation was administered (2 Gy/min) to the tumor xenografts in mice by a linear accelerator (Varian). BPQDs were administered to the tumors on days 0, 3, 6, and 9. Every other day, the volume of tumors was measured and recorded. Mice were sacrificed and the tumor tissues were harvested and fixed in tissue fixation fluid on day 20.

Statistical Analyses
All results are presented as the mean ± standard deviation (s.d.). Comparisons were evaluated by Student's t-test for differences between two groups and ANOVA for differences among three or more groups.

Synthesis and Characterization and BPQDs
The BPQDs used in this study were prepared according to our previous study [20]. The morphology of the obtained BPQDs is shown in the supporting information. BPQDs have an average diameter of approximately 10 nm and a thickness of 3-6 nm ( Figure S1). In addition, the surface zeta potential and hydrodynamic size of BPQDs were measured and are summarized in Table S1.

BPQDs Increase Radiation-Induced Apoptosis of RCC
Our previous study showed that BPQDs enhance the chemosensitivity of RCC cells by impairing spindle assembly [20]. Spindle-targeting drugs have been proven to be highly active as radiosensitizers to enhance the killing effect of tumor cells and improve clinical outcome for patients with cancers. Here, we determined the potential role of BPQDs in RCC 786-O cell radiosensitivity by treating 786-O cells with IR alone or in combination with BPQDs. The percentage of apoptotic cells was significantly higher in BPQDs-and radiation-treated cells than in either monotreatment group ( Figure 1A,B). An increase in combination-treatment-induced apoptosis was also evidenced by cleavage of PARP-1 through immunoblotting analysis ( Figure 1C). In addition, we found that BPQDs decrease the capacity of DNA DSBs and enhance IR-induced apoptosis in another RCC cell line, A498 cells ( Figures S2 and S3). These results suggest that BPQDs dramatically enhance radiation-induced apoptotic cell death of RCC.

Statistical Analyses
All results are presented as the mean ± standard deviation (s.d.). Comparisons w evaluated by Student's t-test for differences between two groups and ANOVA for diff ences among three or more groups.

Synthesis and Characterization and BPQDs
The BPQDs used in this study were prepared according to our previous study [2 The morphology of the obtained BPQDs is shown in the supporting information. BPQ have an average diameter of approximately 10 nm and a thickness of 3-6 nm ( Figure S In addition, the surface zeta potential and hydrodynamic size of BPQDs were measur and are summarized in Table S1.

BPQDs Increase Radiation-Induced Apoptosis of RCC
Our previous study showed that BPQDs enhance the chemosensitivity of RCC ce by impairing spindle assembly [20]. Spindle-targeting drugs have been proven to highly active as radiosensitizers to enhance the killing effect of tumor cells and impro clinical outcome for patients with cancers. Here, we determined the potential role BPQDs in RCC 786-O cell radiosensitivity by treating 786-O cells with IR alone or in co bination with BPQDs. The percentage of apoptotic cells was significantly higher BPQDs-and radiation-treated cells than in either monotreatment group ( Figure 1A,B). increase in combination-treatment-induced apoptosis was also evidenced by cleavage PARP-1 through immunoblotting analysis ( Figure 1C). In addition, we found that BPQ decrease the capacity of DNA DSBs and enhance IR-induced apoptosis in another R cell line, A498 cells ( Figures S2 and S3). These results suggest that BPQDs dramatica enhance radiation-induced apoptotic cell death of RCC. Cells were exposed 20 μg/mL BPQDs along or in combination with 5 Gy IR for 24 h, and representative flow cytome plots; (B) the percentage of AnnexinV-positive apoptotic cells (* p < 0.05, BRQDs + IR versus BPQ and IR monotreatment groups). (C) Cells were exposed to 20 μg/mL BPQDs along or in combinat with 5 Gy IR for 24 h and subjected to immunoblotting with anti-Cleaved PARP (Asp214), and a actin antibodies. Cells were exposed to 20 µg/mL BPQDs along or in combination with 5 Gy IR for 24 h, and representative flow cytometry plots; (B) the percentage of AnnexinV-positive apoptotic cells (* p < 0.05, BRQDs + IR versus BPQDs and IR monotreatment groups). (C) Cells were exposed to 20 µg/mL BPQDs along or in combination with 5 Gy IR for 24 h and subjected to immunoblotting with anti-Cleaved PARP (Asp214), and anti-actin antibodies.

BPQDs Enhance IR-Induced DNA Damage and Slow Damage Repair
DNA is the principal target of IR in cells [21]. To understand how BPQDs contribute to the IR-induced cell-killing effect on 786-O cells, a comet assay was used to detect DNA DSBs after treatment of 786-O cells with IR along or in combination with BPQDs. The comet tails of cotreated 786-O cells were much longer than those in either mono-treated cells (Figure 2A,B). The phosphorylation of H2AX on its Ser139 (γH2AX) will occur around DNA DSBs and is a sensitive molecular marker of DNA DSBs. Immunofluorescence staining for γH2AX foci was adopted, and showed that BPQDs-treated 786-O and A498 cells had prolonged repair kinetics compared to the mock group of 786-O cells at 0.5-24 h post 2 Gy of IR ( Figures 2C,D and S3). An early step in DNA DSBs repair involves the recruitment of 53BP1 to form foci at the damaged DNA ends. The DNA DSBs repair kinetics were also supported by counting 53BP1 foci numbers. We observed a significantly slower rate of DNA DSBs repair in combined-treated 786-O cells than in either BPQDs or IR monotreated cells ( Figure 2E,F). There are two main pathways to repair damaged DSBs: HR and NHEJ. Thus, the efficiencies of NHEJ and HR were quantitatively monitored in vivo via a CRISPR/Cas9-induced oligodeoxynucleotide (ODN) detection system as described in a previous report [22]. BPQDs treatment markedly decreased NHEJ activity in 786-O cells ( Figure 2G) but did not affect HR repair (data not shown). These results suggest that BPQDs impair IR-induced DNA DSBs repair.

BPQDs Enhance IR-Induced DNA Damage and Slow Damage Repair
DNA is the principal target of IR in cells [21]. To understand how BPQDs contribute to the IR-induced cell-killing effect on 786-O cells, a comet assay was used to detect DNA DSBs after treatment of 786-O cells with IR along or in combination with BPQDs. The comet tails of cotreated 786-O cells were much longer than those in either mono-treated cells (Figure 2A,B). The phosphorylation of H2AX on its Ser139 (γH2AX) will occur around DNA DSBs and is a sensitive molecular marker of DNA DSBs. Immunofluorescence staining for γH2AX foci was adopted, and showed that BPQDs-treated 786-O and A498 cells had prolonged repair kinetics compared to the mock group of 786-O cells at 0.5-24 h post 2 Gy of IR ( Figures 2C,D and S3). An early step in DNA DSBs repair involves the recruitment of 53BP1 to form foci at the damaged DNA ends. The DNA DSBs repair kinetics were also supported by counting 53BP1 foci numbers. We observed a significantly slower rate of DNA DSBs repair in combined-treated 786-O cells than in either BPQDs or IR mono-treated cells ( Figure 2E,F). There are two main pathways to repair damaged DSBs: HR and NHEJ. Thus, the efficiencies of NHEJ and HR were quantitatively monitored in vivo via a CRISPR/Cas9-induced oligodeoxynucleotide (ODN) detection system as described in a previous report [22]. BPQDs treatment markedly decreased NHEJ activity in 786-O cells ( Figure 2G) but did not affect HR repair (data not shown). These results suggest that BPQDs impair IR-induced DNA DSBs repair.

BPQDs Suppress DNA-PKcs Activity and Limit the Dynamics of DNA-PKcs at Damage Sites
DNA-PKcs is the key regulator of the NHEJ repair pathway [23]. We investigated whether BPQDs affect the activity of DNA-PKcs in response to IR. IR-induced autophosphorylation of DNA-PKcs on its Ser2056 site was significantly reduced in BPQDs-pretreated 786-O cells compared with the IR-only group ( Figure 3A). We further investigated whether BPQDs affect DNA-PKcs activity in an in vitro system. BPQDs (20 and 40 µg/mL) were mixed with purified DNA-PK complexes, and DNA-PK activity was determined using p53 peptide as a substrate. The BPQDs significantly inhibited DNA-PKcs kinase activity in a dose-dependent manner ( Figure 3B). The system was verified using treatment with the DNA-PKcs inhibitor Nu7441 (0.5 µM). The direct interaction between BP NPs and biological systems has attracted more attention. Here, we examined the possible association of BPQDs with the DNA-PK complex. As shown, purified DNA-PKcs and Ku80 could be well pulled down by BPQDs ( Figure 3C). When we examined the foci of phosphorylated DNA-PKcs at Ser2056, an unexpected result showed that the frequency of pDNA-PKcs-Ser2056 foci after BPQDs treatment was significantly enhanced (Figure 3D,E). Autophosphorylation of DNA-PKcs at Ser2056 is essential for DNA-PKcs dissociation and the accessibility of its downstream factors at damage sites [8]. Therefore, BPQDs impair DNA-PKcs activity and trap it at the damage sites.
experiments. *** p < 0.001). (G) The 786-O cells were transfected with 3 μg Cas9/sgHPRT plasmid and 25 pmol dsODN. At 24 h post-transfection, real-time PCR analysis was performed to measure the dosage of BPQDs that had suppressive effects on NHEJ repair. Nu7441 was adopted as a positive control.

BPQDs Suppress DNA-PKcs Activity and Limit the Dynamics of DNA-PKcs at Damage Sites
DNA-PKcs is the key regulator of the NHEJ repair pathway [23]. We investigated whether BPQDs affect the activity of DNA-PKcs in response to IR. IR-induced autophosphorylation of DNA-PKcs on its Ser2056 site was significantly reduced in BPQDs-pretreated 786-O cells compared with the IR-only group ( Figure 3A). We further investigated whether BPQDs affect DNA-PKcs activity in an in vitro system. BPQDs (20 and 40 μg/mL) were mixed with purified DNA-PK complexes, and DNA-PK activity was determined using p53 peptide as a substrate. The BPQDs significantly inhibited DNA-PKcs kinase activity in a dose-dependent manner ( Figure 3B). The system was verified using treatment with the DNA-PKcs inhibitor Nu7441 (0.5 μM). The direct interaction between BP NPs and biological systems has attracted more attention. Here, we examined the possible association of BPQDs with the DNA-PK complex. As shown, purified DNA-PKcs and Ku80 could be well pulled down by BPQDs ( Figure 3C). When we examined the foci of phosphorylated DNA-PKcs at Ser2056, an unexpected result showed that the frequency of pDNA-PKcs-Ser2056 foci after BPQDs treatment was significantly enhanced (Figure 3D,E). Autophosphorylation of DNA-PKcs at Ser2056 is essential for DNA-PKcs dissociation and the accessibility of its downstream factors at damage sites [8]. Therefore, BPQDs impair DNA-PKcs activity and trap it at the damage sites.

BPQDs Increase IR-Induced Mitotic Errors and Subsequent Micronuclei Formation
Massive unrepaired DNA entering mitosis will lead to mitotic error and form lagging chromosomes and chromatin bridges. Emerging evidence shows that such errors in chromosome segregation trigger the generation of micronuclei (MNs), and the recognition of MNs by innate immune sensors, such as cGAS, leads to autoinflammation or antitumor immunity [24,25]. Here, we found that BPQDs markedly increased IR-induced lagging chromosomes and chromatin bridges in 786-O cells ( Figure 4A,B). We further assessed whether BPQDs would exacerbate the formation of cGAS-positive micronuclei. As shown, BPQDs-treated RCC cells showed significantly increased formation of micronuclei as well as cGAS-and γH2AX-positive micronuclei in response to 10 Gy X-ray irradiation (22.6% in combined-treated cells versus 5.5% in IR-only cells, and 4.3% in BPQDs-treated cells) ( Figure 4C,D).

BPQDs Increase IR-Induced Mitotic Errors and Subsequent Micronuclei Formation
Massive unrepaired DNA entering mitosis will lead to mitotic error and form lagging chromosomes and chromatin bridges. Emerging evidence shows that such errors in chromosome segregation trigger the generation of micronuclei (MNs), and the recognition of MNs by innate immune sensors, such as cGAS, leads to autoinflammation or antitumor immunity [24,25]. Here, we found that BPQDs markedly increased IR-induced lagging chromosomes and chromatin bridges in 786-O cells ( Figure 4A,B). We further assessed whether BPQDs would exacerbate the formation of cGAS-positive micronuclei. As shown, BPQDs-treated RCC cells showed significantly increased formation of micronuclei as well as cGAS-and γH2AX-positive micronuclei in response to 10 Gy X-ray irradiation (22.6% in combined-treated cells versus 5.5% in IR-only cells, and 4.3% in BPQDs-treated cells) ( Figure 4C,D).

BPQDs Sensitize RCC Cells to IR In Vivo
To evaluate the potential radiosensitizing activity of BPQDs, we subcutaneously injected 786-O cells into athymic nude mice and recorded the volume of tumors. We found that the relative tumor volume was dramatically decreased when comparing mice with combined treatment (BPQDs + IR) versus either the BPQDs or IR-treated mice, suggesting the benefit of BPQDs on radiosensitization ( Figure 5A,B). We also performed the histopathological examination of the dissected tumor tissues and found severe vacuolization and structural damage in the combination-treated versus the monotherapy and control

BPQDs Sensitize RCC Cells to IR In Vivo
To evaluate the potential radiosensitizing activity of BPQDs, we subcutaneously injected 786-O cells into athymic nude mice and recorded the volume of tumors. We found that the relative tumor volume was dramatically decreased when comparing mice with combined treatment (BPQDs + IR) versus either the BPQDs or IR-treated mice, suggesting the benefit of BPQDs on radiosensitization ( Figure 5A,B). We also performed the histopathological examination of the dissected tumor tissues and found severe vacuolization and structural damage in the combination-treated versus the monotherapy and control groups ( Figure 5C). These results suggested that BPQDs have the potential to radiosensitize the 786-O cells in in vivo.

Discussion
Our present study demonstrated that BPQDs can act as radiosensitizers in RCC because BPQDs-treated RCC cells exhibit sustained DNA damage signaling, which reflects defects in DNA DSB repair, particularly through NHEJ repair, and consequently enhance IR-induced apoptotic cell death. Furthermore, the results from the in vitro system showed that BPQDs can pull down the DNA-PK complex and inhibit DNA-PK kinase activity. Although our study showed that BPQDs impair IR-induced autophosphorylation of DNA-PKcs in RCC cells, the direct interaction between BPQDs and the DNA-PK complex in vivo and the translocation of BPQDs to the cell nucleus, especially on the damaged DNA ends, are still unanswered questions. Several studies have suggested that DNA-PK can be regulated by various cytoplasmic signaling pathways, including EGFR-Akt signaling [26,27], NF-κB signaling [28], and cytoskeleton-related signaling [29]. Our previous reports showed that BPQDs treatment leads to the stress fiber of microtubule [20], suggesting that BPQDs might regulate DNA-PKcs function through the cytoplasmic signaling pathway. BPQDs also suppress the deacetylase activity of HDAC1 in RCC cells. Histone deacetylases play multiple roles in regulating the DNA damage response, including NHEJ repair. HDAC1 and HDAC2 have been identified as upstream participants of NHEJ, at least in part by regulating the proper dynamics of NHEJ factors from damaged sites [30]. Here, we found that BPQDs trap phosphorylated DNA-PKcs at damaged sites. Whether BPQDs induce inappropriate disassembly of DNA-PKcs from DSBs sites by eliminating HDAC1 activity remains unclear and needs further investigation.
DNA-PKcs also functions as a key mitotic signaling kinase other than the DNA damage response [31,32]. Mitotic activation of DNA-PKcs is required for phosphorylation of downstream target factors, including Chk2 [33] and PLK1 [34,35]. DNA-PKcs-dependent

Discussion
Our present study demonstrated that BPQDs can act as radiosensitizers in RCC because BPQDs-treated RCC cells exhibit sustained DNA damage signaling, which reflects defects in DNA DSB repair, particularly through NHEJ repair, and consequently enhance IR-induced apoptotic cell death. Furthermore, the results from the in vitro system showed that BPQDs can pull down the DNA-PK complex and inhibit DNA-PK kinase activity. Although our study showed that BPQDs impair IR-induced autophosphorylation of DNA-PKcs in RCC cells, the direct interaction between BPQDs and the DNA-PK complex in vivo and the translocation of BPQDs to the cell nucleus, especially on the damaged DNA ends, are still unanswered questions. Several studies have suggested that DNA-PK can be regulated by various cytoplasmic signaling pathways, including EGFR-Akt signaling [26,27], NF-κB signaling [28], and cytoskeleton-related signaling [29]. Our previous reports showed that BPQDs treatment leads to the stress fiber of microtubule [20], suggesting that BPQDs might regulate DNA-PKcs function through the cytoplasmic signaling pathway. BPQDs also suppress the deacetylase activity of HDAC1 in RCC cells. Histone deacetylases play multiple roles in regulating the DNA damage response, including NHEJ repair. HDAC1 and HDAC2 have been identified as upstream participants of NHEJ, at least in part by regulating the proper dynamics of NHEJ factors from damaged sites [30]. Here, we found that BPQDs trap phosphorylated DNA-PKcs at damaged sites. Whether BPQDs induce inappropriate disassembly of DNA-PKcs from DSBs sites by eliminating HDAC1 activity remains unclear and needs further investigation.
DNA-PKcs also functions as a key mitotic signaling kinase other than the DNA damage response [31,32]. Mitotic activation of DNA-PKcs is required for phosphorylation of downstream target factors, including Chk2 [33] and PLK1 [34,35]. DNA-PKcs-dependent Chk2-phosphorylation on its Thr68 site facilitates activation of the Chk2-BRCA1 pathway [33]. DNA-PKcs also colocalizes with PLK1, which is an essential kinase during mitosis progression [36], at the centrosome, and DNA-PKcs promotes the PLK1 activitymediated G 2 /M transition [34]. Shao and colleagues showed that BP nanomaterials can exist on the centrosome to compromise centrosome integrity by deactivating the PLK1 activity [19]. The BP nanomaterials interact with PLK1, leading to its aggregation and restricting the recruitment of PLK1 to centrosomes. The BPQDs may block the interaction between DNA-PKcs and PLK1, and disrupt DNA-PKcs-mediated activation of PLK1. Our recent study revealed that DNA-PKcs associates with HDAC6 and modulates HDAC6-mediated deacetylation of HSP90, which is important to maintain the protein stability of the mitotic kinase Aurora A [37]. Aurora A plays essential roles in regulating mitotic spindle formation, and inhibition of Aurora A leads to failure of chromosome congression at metaphase and lagging chromosomes and chromatin bridges in anaphase [38]. Consistent with this, our present data showed that BPQDs treatment enhanced IR-induced lagging chromosomes and chromatin bridges in RCC cells, indicating that BPQDs may also influence HDAC6-HSP90 signaling via suppression of DNA-PKcs.
Cytosolic self-DNA, such as micronuclei, can be generated by mitosis error following DNA damage in mammalian cells, triggering cGAS-STING-dependent inflammatory signaling [24,25]. Most recent work identified that DNA-PKcs phosphorylates cGAS and inhibits its enzymatic activity. DNA-PKcs deficiency enhances the cGAS-mediated innate immune response [38]. The BPQDs suppress the kinase activity of DNA-PKcs, impair DNA DSBs repair efficiency, and disrupt the mitotic spindle structure. In line with this notion, we observed that BPQDs-pretreated RCC cells exhibit an increased number of IR-induced micronuclei and an elevated amount of cGAS localization to micronuclei, suggesting that BPQDs may have the potential to enhance IR-induced innate immunity in RCC cells. RCC is traditionally considered to be resistant to conventionally fractionated radiotherapy with the dose 1.8-2.1 Gy per fraction [39]. Modern technological advances in radiation oncology have increased the efficacy of radiotherapy, allowing higher dosage delivery to tumor, and leading to effective management of cancer patients [40]. Recent studies showed that the application of stereotactic body radiotherapy (SBRT) was associated with better local control of metastatic RCC [41][42][43]. These studies show that RCC can no longer simply be recognized as radioresistant, and more studies are necessary for exploring the combination of SBRT with other therapy strategies [44]. Therefore, development of target-based radiosensitization strategies to sensitize cancer cells to RT become attractive therapeutic strategy for the clinical benefit of RCC patients. Significant evidence has revealed the potential of DNA-PKcs in cancer development; thus, various anti-DNA-PKcs strategies have been proposed as either monotherapy or in combination with chemo-and radiotherapy [45]. Here, we found that BPQDs can inhibit the kinase activity of DNA-PKcs and radiosensitize RCC cells in vivo and in vitro.

Conclusions
In summary, we found that BPQDs inhibit DNA-PKcs activity and impair DNA-PKcsmediated NHEJ DNA DSBs repair, resulting in sustained DNA damage in response to IR. BPQDs enhances IR-induced suppression of RCC xenografts growth in vivo, pointing toward a promising BPQDs-based targeted cancer therapy.