Theabrownin triggers DNA damage to suppress human osteosarcoma U2OS cells by activating p53 signalling pathway

Abstract Osteosarcoma becomes the second leading cause of cancer death in the younger population. Current outcomes of chemotherapy on osteosarcoma were unsatisfactory to date, demanding development of effective therapies. Tea is a commonly used beverage beneficial to human health. As a major component of tea, theabrownin has been reported to possess anti‐cancer activity. To evaluate its anti‐osteosarcoma effect, we established a xenograft model of zebrafish and employed U2OS cells for in vivo and in vitro assays. The animal data showed that TB significantly inhibited the tumour growth with stronger effect than that of chemotherapy. The cellular data confirmed that TB‐triggered DNA damage and induced apoptosis of U2OS cells by regulation of Mki67, PARP, caspase 3 and H2AX, and Western blot assay showed an activation of p53 signalling pathway. When P53 was knocked down by siRNA, the subsequent downstream signalling was blocked, indicating a p53‐dependent mechanism of TB on U2OS cells (p53 wt). Using osteosarcoma cell lines with p53 mutations (HOS, SAOS‐2 and MG63), we found that TB exerted stronger inhibitory effect on U2OS cells than that on p53‐mut cell lines, but it also exerted obvious effect on SAOS‐2 cells (p53 null), suggesting an activation of p53‐independent pathway in the p53‐null cells. Interestingly, theabrownin was found to have no toxicity on normal tissue in vivo and could even increase the viability of p53‐wt normal cells. In sum, theabrownin could trigger DNA damage and induce apoptosis on U2OS cells via a p53‐dependent mechanism, being a promising candidate for osteosarcoma therapy.

apoptosis on U2OS cells via a p53-dependent mechanism, being a promising candidate for osteosarcoma therapy.

K E Y W O R D S
DNA damage, osteosarcoma, P53, theabrownin, zebrafish 1 | INTRODUCTION Osteosarcoma (OS), a very aggressive intra-osseous neoplasm, is one of the most common primary malignant bone diseases that severely threatens the health of children, adolescents and young adults. 1 It exhibits a predilection to occur in the metaphysis of long bones, and commonly occurs in the distal femur (30%-43%), proximal tibia (15%-23%) or proximal humerus (10%-15%), which represent sites containing the most proliferative growth plates. 2,3 OS is derived from primitive mesenchymal stem cells mutated in the process of differentiation towards osteoblasts, with highly invasive and distant metastatic potential. 4 It is prone to hematogenous metastasis at early onset and after surgery. Approximately 15% to 20% of patients have clinically detectable metastases with more than 85% of occurrence in the lungs. 2,5 The world-wide annual incidence of OS is 3.1 per million for all ages and 4.4 per million for individuals <25 years of age. 6 Approximately 400 new cases of OS are annually diagnosed in the US and 0.2~3/100 000 patients per year (0.8-11/100 000 patients per year in the age of 15-19 years) are reported as overall OS incidence in the EU. 7,8 It becomes the third most common malignancy and the second leading cause of cancer-related deaths in both children and young adults. 8,9 The SEER (Surveillance, Epidemiology and End Results) database revealed that long-term clinical outcomes for children and young adults with OS have changed very little over the past 30 years. 10 At present, surgery with chemotherapy is the first-line treatment. 11 Surgical resection of the primary tumour with adequate margins is an essential component of the curative strategy for OS patients. 12 Nevertheless, surgical results in large bone defects of the affected limb and complex skeletal rebuilding limits its application. 13 The gold standard of chemotherapy regimens includes high doses of cis-platinum, doxorubicin, etoposide, ifosfamide, methotrexate and combinations of these drugs. 14,15 Unfortunately, short-and long-term collateral side effects of the chemotherapy leave both clinicians and patients unsatisfied. 16 Acute toxicities such as cardiac dysfunction, ototoxicity, infertility, myelosuppression, mucositis and secondary malignancies are frequently associated with high doses of chemotherapy, and early or late cardiac failure or sepsis following febrile neutropenia has been the major reasons for deaths caused by chemotherapeutics. 17,18 Moreover, OS may be inherently resistant to chemotherapy or become unresponsive to these drugs, which occurs in 35%-45% of patients. 16,19 Thus, the outcome of chemotherapy remains disappointing with overall 5-year survival rate of only about 20%, [20][21][22] and effective therapies with little side effects are urgently required.
As one of the most traditional and commonly consumed nonalcoholic beverage in the world next to water, tea (fresh leaves of Camellia sinensis (L) O. Kuntze) is used not only for health promotion but also medicinal purposes. It has been reported to possess antioxidant, anti-inflammatory, anti-proliferative and anti-angiogenesis activities which are potentially significant to the prevention and treatment of various forms of diseases, such as cancer. 23,24 Oral intake of tea reduces the risk of many cancer incidences, including breast cancer, liver cancer, oral cancer, etc., with little adverse events. [25][26][27][28] Apart from the chemopreventive effect, tea also exerts chemotherapeutic effects on cancer by inducing apoptosis. 24,[28][29][30] Theabrownin (TB), theaflavin (TF) and thearubigin (TR) are the main components of tea, which determine tea 0 s colour, taste and bioactivity. 31 TB is a reddish-brown material with the highest water solubility. It has significant cholesterol-lowering activity, relieves fatigue and reduces blood lipid levels. 32 TB comprises of a family of macromolecules transformed from polyphenols, and is considered superior to TF or TR in physicochemical and medicinal properties. Previously, we reported that TB possessed strong pro-apoptotic and cell cycle arresting effects on human carcinoma cells, making it a promising candidate for cancer therapy. 33,34 To determine its anti-OS effect, this study employed U2OS cells and performed in vivo and in vitro assays.

| Cell line and culture
The human OS U2OS, SAOS-2, HOS, MG63 cell lines and 1 marrow mesenchymal stem cells (BMSC) were obtained from Shanghai Cell Bank of Chinese Academy of Sciences (Shanghai, China). U2OS, SAOS-2, HOS and MG63 cells were cultured, respectively, in RPMI-1640, McCOY 0 s 5A and MEM (HOS and MG63) medium containing 10%-5% FBS at 37°C in a humidified 5% CO 2 incubator. BMSC was cultured in STEMPRO â hMSC serum free medium at the above condition. All mediums were daily changed and the cells were treated with TB in their logarithmic growth phase.

| Zebrafish
The zebrafish wide-type AB strain was purchased from the China

| TB dose range determination in zebrafish
Totally 180 larval zebrafish (3 dpf) were randomly divided into 6 groups (30 fishes each) and cultured into 6-well plates (Nest Biotech, China) in 3 mL fresh fish water. TB powders were dissolved into the plates at 0, 200, 400, 1000, 1500 and 2000 lg/m, respectively, for 24 hours. Thereafter, fishes were subjected to visual observation under a stereoscopic microscope to record the mortality. The mortality curve was generated using Origin 8.0 (OriginLab, Northampton, MA, USA), and the MNLC and NOAEL of TB was determined.
According to the preliminary studies, the doses at 1/10 NOAEL, 1/3 NOAEL and NOAEL were used as low, middle and high dose of TB, respectively, for subsequent experiments.

| Cell viability assay
Viabilities of TB-treated cells (U2OS, SAOS-2, HOS, MG63, BMSC) was determined by MTT assay as described in our previous report. 34 Cells were seeded on 96-well plates with density of 1 9 10 4 cells/ were selected as the low, medium and high concentrations of TB for following tests.

| Cell morphology and DAPI staining
The TB-treated U2OS cells at 24 hours were washed with phosphate-buffered saline (PBS) thrice and fixed with 4% paraformaldehyde in PBS for 30 minutes at room temperature. Then the cells were permeabilized with 0.5% Triton X-100 in PBS for 10 minutes.
An aliquot of the cells were mounted using ProLong â Diamond Antifade Mountant with DAPI in dark. The unstained and stained cells were observed under a fluorescence microscope (Carl Zeiss, G€ ottingen, Germany). Five coverslips were used as replicates of each group and the apoptotic nuclei of cells were visualized.

| Flow cytometry
TB-induced apoptosis of U2OS cells was determined by flow cytometry using an Annexin-V/PI method, according to the manufacturer's protocol. Briefly, U2OS cells were seeded on 6-well plates with density of 2.5 9 10 5 cells/well for 24 hours and then were treated with TB at low, medium and high concentrations for another 24 hours.
Thereafter, the cells were harvested and washed twice with cold PBS, and then labelled with FITC Annexin V and PI in binding buffer.
Fluorescence intensity of the cells was detected by BD Accuri TM C6 (BD, Franklin Lakes, NJ, USA). The analysis was replicated thrice and the apoptosis rate (%) for each TB treatment was calculated.

| Western blot analysis
After TB treatment, total proteins of the U2OS cells (1.5 9 10 6 ) were extracted using a lysis buffer (50 mmol/L Tris-HCl, pH 7.4, 150 mmol/L NaCl, 1 mmol/L EDTA, 1% Triton, 0.1% SDS) with proteinase inhibitor cocktail (Bimake, Houston, USA) for 30 minutes on ice. The proteins were separated by a denaturing sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE, 6%-12%) and then transferred onto a nitrocellulose membrane (Sartorius Stedim, Goettingen, Germany). The membrane was blocked with 5% non-fat milk for 2 hours, followed by overnight incubation at 4°C with the primary antibodies. Following incubation with peroxidaseconjugated goat anti-rabbit/mouse IgG at room temperature for 2 hours, proteins were visualized using Western Lightning â Plus ECL (Perkin Elmer, Waltham, Massachusetts, USA) and detected using Xfilm (Kodak, Tokyo, Japan) and scanned.

| Statistical analysis
Data were expressed as mean AE SD and subjected to one-way ANOVA, followed by Fisher's least significant difference (LSD) comparison. All analyses were performed using an updated version of DPS software.

| MNLC (maximal-no-lethal-concentration) and NOAEL (no-observed-adverse-effect-level) of TB
The mortality curve of larval zebrafish is shown in Figure 1A

| Anti-OS effect of TB in vivo
Fluorescence intensity was used to assess the U2OS xenograft tumour growth in larval zebrafish. As shown in Figure 1A lower, a xenograft model of OS was successfully established in zebrafish and TB concentrations from low to high doses significantly inhibited the tumour growth with inhibitory rates from 24.6 to 27.3% as compared to untreated controls (all P < .05). Cis-platinum also exerted significant inhibitory effect against the tumour growth (P < .05) at its NOAEL (75 lmol/L), which was no more effective than TB.

| Inhibitory effect of TB in vitro
As shown by MTT assay in Figure 2A   By the TUNEL assay, we visualized and quantified in situ DNA strand breaks and fragments in nuclear chromatin of U2OS cells.

| Pro-apoptotic effect of TB
This assay was performed to verify the pro-apoptotic effects of TB.
As shown in Figure 2C, apoptotic cells with positive fluorescein staining were detected in TB-treated groups. The positive rate was significantly increased from 1.0 AE 1.2% to 88.7 AE 6.2% with increasing concentrations of TB, also indicating a concentrationdependent reaction pattern.
To determine the specificity and safety of TB's pro-apoptotic effects, normal larval zebrafish were treated and AO staining was performed. As shown in Figure 1C, TB induced no apoptosis in normal zebrafish larvae over its effective dose range (2.13 to 21.3 lg/mL) and even at a very high dose (MNLC, 213 lg/mL), whereas cis-platinum exerted significant pro-apoptotic effects in normal cells of zebrafish at its effective dose (NOAEL, 75 lmol/L). Hence, TB exerted proapoptotic effects specifically to tumour cells but not to normal cells, indicating that TB was safer than chemotherapy with cis-platinum.

| Pro-apoptotic targets of TB
Immunofluorescence assay for cleaved-caspase 3 (c-Casp3) and phosphorylated H2AX (p-H2AX) (Ser 139) was performed to study the pro-apoptotic targets of TB on the cellular level. As shown in Figure 2D, the immunopositive rate of c-Casp3 activated cells was significantly increased from 2.9 AE 5.4% to 22.9 AE 5.6% (all P < .01).
As shown in Figure 3

| Molecular mechanism of action of TB
Since c-Casp3 and p-H2AX were activated by TB, the p53 signalling pathway could be deduced as a main regulator of TB's anti-OS effect. To verify this assumption, Western blot assays were performed using antibodies against upstream and downstream signalling components of the p53 signalling pathway. As shown in Figure 5, TB treatment from 0 to 50 lg/mL significantly increased the phosphorylation levels of p53 and its upstream molecules ATM, Chk1, Chk2 (all P < .01). The expression of the downstream proteins Bax, Fas, c-Casp9 and c-Casp8 of P53 were also increased by TB. Moreover, as compared with cis-platinum, TB activated more expressions of p-H2AX, c-PARP, p-ATM and p-P53, indicating that TB has stronger effect than that of the "classical" DNA damaging agent.
TP53-siRNA was used to verify the regulatory role of the p53 signalling pathway in TB-induced apoptosis. As shown in Figure 6, with high dose of TB (50 lg/mL) and TP53-siRNA treatment, p53 was not detectable anymore and the expressions of its downstream molecules c-Casp9, c-Casp8, c-Casp3, P21 and Bax were significantly decreased (all P < .01). Furthermore, c-PARP, the marker of caspase-3 mediated apoptosis, was also inhibited (P < .01), indicating an inhibition of apoptosis by TP53-siRNA.

| Effect of TB on p53-mut OS cells and normal cells
As shown in Figure 7  and HOS cells over its concentration range from 10 to 60 lg/mL, and only a slight inhibitory tendency was seen with TB at 80 lg/mL.
In contrary, TB was found to increase the viability of BMSC from 10 to 80 lg/mL at 24 hours and from 20 to 60 lg/mL at 48 hours, indicating a beneficial effect on the normal cells.

| DISCUSSION
Natural products have attracted growing interest in anticancer therapy due to their advantages of high bioactivity and low toxicity. 35,36 As one of the most commonly used natural products, tea possesses therapeutic potential for treating cancers, including OS. 30,37,38 The components, such as polyphenols, polysaccharide, pigments (TB, TF, and TR), etc., may be responsible for the anticancer activity of tea. 39 To date, a large number of studies have focused on tea polyphenols (mainly catechins) and revealed their chemopreventive and adjuvant effects on cancer. 40,41 However, poor bioavailability, unfavourable side effects, and high costs limited the development and clinical application of catechins. [42][43][44] Apart from polyphenols, TB attracts increasing attention in recent years due to its bioactivities and therapeutic potential. 32,45 In our previous studies, TB exerted strong proapoptotic effects against non-small cell lung cancer via the p53 signalling pathway. 33,34 Using xenograft model and osteosarcoma cells, this study investigated the anti-OS effects and molecular mechanism of TB.
The in vivo data showed that TB exerted a significant anti-OS effect at doses ranging from 1/10 NOAEL to NOAEL levels. The effect was stronger than that of cis-platinum at its NOAEL level. The in vitro data revealed that TB-inhibited proliferation and induced apoptosis of U2OS cells in a concentration-dependent manner.
Through immunofluorescence and Western blot assays, we deter- is responsible for cell apoptosis, the p53 signalling pathway could be assumed mediating the pro-apoptotic mechanism of TB.
To verify this assumption and clarify TB's underlying mechanism, we analyzed the molecules on p53 pathway related and confirmed an activation of p53 signalling pathway in TB-treated U2OS cells.
ATM is a 350 kD serine/threonine kinase that regulates DNA repair. 56 Activation of ATM by autophosphorylation on Ser1981 occurs in response to exposed DNA DSBs, and the activated ATM TB 0 s anti-OS effect was p53-dependent. We found that TB has stronger inhibitory effect on the p53-wt U2OS cells than that on the p53-mut cells, but it could also inhibit SAOS-2 cells (p53 null) obviously at its high concentrations. The result indicated a p53-dependent mechanism on the p53-wt cells and a p53-independent mechanism on the p53-mut cells. Further studies are needed to clarify the latter mechanism of TB.
During recent years, zebrafish xenograft models have been increasingly generated to study malignancies, qualifying this illustra- DNA damage and their decreased traditional repair capacity, leading cancer cells to become exceedingly more dependent on homologous recombination repair as a means of protection from the lethal effect of both spontaneous and therapy-induced DSBs in DNA. 79 Therefore, DNA-damaging regimes, such as TB, could selectively induce tumour cell apoptosis by augmenting genomic instability, and the DNA repair mechanisms would work efficiently in normal cells in contrast to tumour cells.
Although TB exerted better anti-OS efficacy outcomes than cisplatinum, the OS tumour still existed in zebrafish after the treatment.
It reminds us to consider the combination of TB and chemotherapy for enhancing the efficacy and reducing the toxicity in future studies.
Taken together, our results indicated that TB-exerted anti-OS effects as a pro-apoptotic agent through p53 signalling pathway. It can be regarded as an important target for TB on p53-wt cancers.

ACKNOWLEDGEMENT
This study was funded by the Zhejiang Provincial Natural Science

CONF LICTS OF INTEREST
There is no conflict of interest for this work. Thomas Efferth improved the design, draft and revision of this work.

AUTHOR CONTRIBU TI ONS
All listed authors approved the manuscript for publication, and agreed to be accountable for all aspects of this work.