53BP1 Sensitizes Breast Cancer Cells to 5-Fluorouracil

Chemoresistance of breast cancer is a worldwide problem for breast cancer and the resistance to chemotherapeutic agents frequently led to the subsequent recurrence and metastasis. In our previous study, we have found that 53BP1 showed a gradual decrease during the progression of breast cancer and loss of 53BP1 was associated with metastasis and poor prognosis in breast cancer. Here we aimed to reveal whether 53BP1 could sensitize breast cancer to 5-Fu. We found that ectopic expression of 53BP1 can significantly sensitize breast cancer cells to 5-Fu while knockdown of 53BP1 conferred the resistance. The in vivo experiments confirmed that overexpression of 53BP1 in combination with 5-Fu markedly inhibited growth of xenotransplanted tumors in nude mice when compared to either agent alone. Furthermore, we demonstrated that 53BP1 regulated the sensitivity to 5-Fu through thymidylate synthase (TS) and dihydropyrimidine dehydrogenase (DPYD). The present studies provide a new clue that combination of 5-Fu and 53BP1 could be a potential novel targeted strategy for overcoming breast cancer chemoresistance.


Introduction
Breast cancer has been estimated to be one of the most commonly diagnosed types of female malignancy around the world. Although mortality rates of breast cancer seem to reduce during the past two decades, incidence rates continue to increase recently [1] and it is estimated about 39,510 women will die of breast cancer in the U.S. in 2012 [2]. Breast cancer is one kind of solid tumors which are sensitive to chemotherapy, thus chemotherapy is an important component in treatment of breast cancer. However chemoresistance is a worldwide problem for breast cancer and the resistance to chemotherapeutic agents frequently led to the subsequent recurrence and metastasis of cancer. Until now, the detailed mechanisms involved in chemoresistance are still largely unknown. Therefore, it is in urgent need to search for novel markers that could predict the response to chemotherapy. 5-Fluorouracil (5-Fu) plays an important role in standard chemotherapy protocols for a variety of solid tumors including breast cancer. But it is limited in clinical application due to the resistance. 5-Fu is antimetabolite inhibitors of de novo purine and pyrimidines syntheses and it is converted intracellular into 59fluoro-29-deoxyuridine by thymidine phosphorylase. Subsequently it is phosphorylated by thymidine kinase into 5-fluoro-29deoxyuridine 59-monophosphate (FdUMP). FdUMP which is the active form of 5-Fu inhibits thymidylate synthase (TS) so as to inhibit DNA synthesis. In addition, 5-FU can be converted into fluoro-5,6-dihydrouracil (FUH2), the inactive form of 5-Fu, by dihydropyrimidine dehydrogenase (DPYD) to lose its function [3]. Also TS and DPYD are reported to be predictive markers for 5-FU in cancers [4,5]. Therefore the expression and activity of TS and DPYD are two major factors in molecular signaling pathway of chemoresistance to 5-Fu.
Human 53BP1 (p53 Binding Protein 1) was first identified by Iwabuchi et al. [6] and it was mapped to chromosomes 15q15-21 [7]. 53BP1 has been reported to be a candidate tumor suppressor by many studies [8][9][10][11]. Our collaborative groups have revealed that tumors with lower 53BP1 had significant poor metastasis free survival. [12]. Our previous studies also have demonstrated that 53BP1 showed a gradual decreased protein levels during the progression of breast cancer and it had lower expression in cancer lesions than in the matched non-tumor lesions. In addition 53BP1 could inhibit cell proliferation and invasiveness of breast cancer through nuclear factor-kappaB pathway [13]. All the above data raise the question whether 53BP1 has the effect on 5-Fu treatment of breast cancer.
In the present study, we aimed to reveal the potential role of 53BP1 in response to 5-Fu and provide a new clue for future clinical treatments of breast cancer patients who are resistant to 5-Fu treatment.

Cell culture and transfection
Breast cancer cell lines MCF-7, MDA-MB-231, MDA-MB-468, and T47D were obtained from American Type Culture Collection (ATCC, Rockville, MD, USA). They were routinely cultured in appropriate medium supplemented with 10% FBS and 100 units of penicillin-streptomycin at 37uC with 5% CO2 in a humidified incubator. The plasmids were constructed and the cells were transfected as previously described [13,14].

Reagents
Antibody against P21, Bax, Histone H2AX, TS and DPYD were purchased from Cell Signaling Technology (Beverly, MA, USA). Antibody against Bcl-2 was from Dako (Carpinteria, CA, USA). Rabbit anti-53BP1 antibody was from Bethyl Laboratories (Montgomery, USA). Signal silence TS siRNA, DPYD siRNA and their control siRNA were purchase from Cell Signaling Technology. Other reagents were from Sigma-Aldrich (St. Louis, MO, USA) unless specifically described.

Western blot analysis
Cells were lysed with radio immunoprecipitation assay (RIPA) buffer (Shennengbocai, Shanghai, China) with protease inhibitors. Equal amount of protein were loaded on a SDS-PAGE gel and transferred to polyvinylidene fluoride membranes (ImmobilonP; Millipore, Bedford, MA). The membrane was incubated with the corresponding primary antibodies overnight at 4uC after blocking with 5% non-fat milk for 1 hour. Subsequently the secondary antibody was added for 1 hour at room temperature. Signals were detected using enhanced chemiluminescence. b-actin was used as the loading control.
Quantitative reverse-transcription PCR analysis (qRT-PCR) RNA was extracted with TRIZOL reagents (TaKaRa, Dalian, China). Briefly, total 0.8 mg RNA was reverse transcribed to cDNA by PrimerScript RT Reagent Kit (TaKaRa, Dalian, China). qRT-PCR was performed with Applied Biosystems StepOne and StepOnePlus Real-Time PCR Systems. GAPDH was used as an endogenous loading control. Relative quantification analysis was performed using the comparative C (T) (2 (2DDCT) ) method. All the experiments were repeated at least three times.

Immunofluorescence staining
The experiments were performed as described previously [16]. The cells were grown on coverslips in the 24-well plates, and then the cells were fixed with 4% paraformaldehyde and permeabilized using 0.1% Triton-X100 in PBS (PBST). Cells were blocked with rabbit anti-53BP1 and anti-H2AX antibody and hodamineconjugated anti-rabbit secondary antibody. The cells were further stained with 4, 6-diami-dino-2-pheny-lindole (DAPI).

Cell cycle analysis by flow cytometry
The cell cycle analysis were done as previously described [15]. Cells were harvested and washed by PBS and then stained with 50 mg/ml propidium iodide for 30 min. For fluorescence-activat-

In vivo experiments
The in vivo experiments were approved by the Animal Care and Use Committee of Shandong University. They were performed as described previously [13]. Briefly, five million cells were injected subcutaneously into each flank of BALB/c nu/nu mice. Then 5-Fu was injected by tail vein three times per week for two weeks at a dose of 30 mg/kg. Tumor diameter was measured with calipers and the tumor volume was calculated by the formula: (width) 2 6length/2. When mice were sacrificed, the formed breast cancer tumors were embedded in paraffin after being fixed in 10% formalin. Then the slides from the paraffin were subjected to terminal nucleotidyl transferase-mediated nick end labeling (TUNEL) assay (Beoytime, Beijing, China).

Statistical analysis
The software SPSS 18.0 (Chicago, IL, USA) was used to analyze the data. Student's t test was used to determine significance between two groups. All experiments were repeated at least three times. The results were shown as mean 6 standard deviation (SD). The results were considered statistically significant when P value,0.05.

Results
High expression of 53BP1 was correlated with 5-Fu sensitivity in breast cancer cell lines In order to explore the effect of 53BP1 on 5-Fu in breast cancer, we first detect the association between 53BP1 expression and 5-Fu sensitivity in 4 breast cancer cell lines. As shown in Fig. 1A and 1B, MDA-MB-231 and MDA-MB-468 cells with lower levels of 53BP1 was more resistant to 5-Fu while T47D and MCF-7 with higher levels of 53BP1 was sensitive to 5-FU. We also detected the the localization of 53BP1 and H2AX foci in breast cancer before and after the treatment of 5-Fu. As shown in Figure 1C and 1D, we found that the MDA-MB-231 and MDA-MB-468 cells had fewer both 53BP1 and H2AX foci than T47D and MCF-7 cells. These results suggested that relative expression of 53BP1 may correlate with 5-Fu sensitivity in breast cancer.  To evaluate the role of 53BP1 conferring sensitivity to 5-Fu in breast cancer cells, we established stable cell lines that overexpressing 53BP1 in MDA-MB-231 and stable MCF-7 cell lines with shRNA-mediated knockdown of 53BP1. The transfection efficiency was confirmed by western blot and quantitative reversetranscription PCR analysis (Figure 2A).
To investigate whether 53BP1 could modulate 5-Fu sensitivity in the above transfected breast cancer cells, we first performed a MTT assay as described earlier [17]. As shown in Figure 2B, MDA-MB-231 cells overexpressing 53BP1 showed enhanced sensitivity to 5-Fu in a time and dosage dependent manner. Consistent with this, we found that it was more resistant to 5-Fu in 53BP1 knockdown MCF-7 cells than its control MCF-7 cells ( Figure 2C). The half maximal inhibitory concentration IC50 assessed by the MTT assay was shown in Table 1.
In addition, we detected the effect on cell cycle of breast cancer cell lines. As shown in Figure 3, 53BP1 could induce the G2/M arrest in both MDA-MB-231 and MCF-7 cells. These results were inconsistent with our findings in ovarian cancer [18]. In our previous studies [13], we have found 53BP1 could inhibit the growth of breast cancer cells. Here we also examined the effects of 53BP1 on doubling time (Dt) in these cells. We found that the Dt for MDA-MB-231 was 9.6 h while the MDA-MB-231 53BP1 cells were 14.4 h. In MCF-7 cells, the Dt for 53BP1 shRNA 1 and shRNA 2 were 13.9 h and 16.8 h respectively while the Dt was

53BP1 sensitizes 5-FU-induced apoptosis in breast cancer cells
As 5-Fu could inhibit cancer cell growth and induce apoptosis, so we detected the cell apoptosis associated protein in 53BP1 transfected breast cancer cells. As shown in Figure 4, the results from western bolt analysis were also in concordance with that of the MTT assay in MDA-MB-231 and MCF-7 cells. The levels of p21 and Bax were increased and Bcl-2 was decreased in 53BP1 transfected MDA-MB-231 cells. Also knockdown of 53BP1 dramatically decreased p21 and Bax and increased the Bcl-2 level.

53BP1 sensitizes breast cancer cells to 5-fluorouracil in vivo
We next investigate the effect of 53BP1 on 5-Fu sensitivity in vivo using nude mice xenograft studies. MDA-MB-231 cells overexpressing 53BP1, MCF-7 cells with 53BP1 knockdown, and their respective control cells were injected subcutaneously into each flank of female nude mice. After establishment of tumors for about a week, 5-Fu was injected by tail vein three times per week for two weeks at a dose of 30 mg/kg. Then the mice were maintained for another three weeks. As shown in Figure 5A and 5B, while overexpressing of 53BP1 or 5-Fu treatment alone significantly inhibited tumor growth, combination of upregulation 53BP1 and 5-Fu treatment resulted in an additive inhibition of tumor growth versus either treatment alone. The similar results were observed in 53BP1 knockdown MCF-7 cells. While the control MCF-7 cells treated with 5-Fu significantly inhibited the tumor growth, there was no significant different in the 53BP1 knockdown MCF-7 cells with or without 5-Fu administration ( Figure 5C and 5D). All the above findings suggested that 53BP1 could sensitize breast cancer cells to 5-Fu in vivo.
Apoptosis induced by 5-Fu in tumor tissues from the vivo experiments was also measured by a terminal nucleotidyl transferase-mediated nick end labeling (TUNEL) assay. As shown in Figure 6A and 6B, when the mice were treated by 5-Fu, there were more TUNEL-positive cells in tumors from 53BP1 overexpressing MDA-MB-231 cells than those of control MDA-MB-231 cells. In addition, we found that the proportation of TUNELpositive cells were less in 53BP1 knockdown MCF-7 tumors treated with 5-Fu than that of MCF-7 control cells ( Figure 6C and 6D). These above data suggested that 53BP1 maybe a novel effective sensitizer to 5-Fu treatment in breast cancer.

53BP1 sensitizes breast cancer cells to 5-fluorouracil through TS and DPYD
Several in vitro and in vivo studies have revealed a strong association between increased TS and DPYD expression and the resistance to 5-FU in various tumors including breast cancer [3,[19][20][21][22]. So we hypothesized that 53BP1 might contribute to 5-Fu sensitivity by regulating TS and DPYD. First we detected the mRNA and protein levels in 53BP1 transfected cells. As shown in Figure 7A, 7B and 7C, TS and DPYD protein expression levels were significantly increased in 53BP1 knockdown MCF-7 cells compared to its control cells, and they were significantly decreased in 53BP1 overexpressing MDA-MB-231 cells compared to MDA-MB-231 control cells.
To further confirm the potential role of TS and DPYD in 53BP1 mediated 5-Fu sensitivity in breast cancer cells, we used TS siRNA and DPYD siRNA to knockdown the levels of TS or DPYD in 53BP1 knockdown MCF-7 cells. The transfection efficiency was confirmed by western blot analysis ( Figure 7D and 7E).
Then we used MTT assay to measure the changes of sensitivity to 5-Fu with the control and TS/DPYD siRNA transfected cells. As shown in Figure 8, both TS and DPYD siRNA knockdown resulted in more sensitive to 5-Fu in 53BP1 knockdown MCF-7 cells. However no change was observed in the control MCF-7 cells after knockdown TS or DPYD. The half maximal inhibitory concentration IC50 assessed by the MTT assay was shown in Table 2. These results revealed that TS and DPYD played an important role the 53BP1 mediated sensitivity to 5-Fu in breast cancer cells.

Discussion
Though many advances have been made in breast cancer systemic treatment, many patients diagnosed with early stage breast cancer eventually die of recurrent disease due to chemoresistance [23,24]. Therefore identify novel markers that could predict the response to chemotherapy will help develop effective targeted strategies to combat breast cancer.
Recently, we and others have identified 53BP1 functioned as a novel tumor suppressor in cancers [9,18,25]. 53BP1 has been shown to inhibit cell proliferation, invasion, metastasis [13] and the high levels of 53BP1 were reported to be associated with better prognosis in cancers [10,12,26]. So this raised the hypothesis that loss of 53BP1 may contribute the treatment failure of breast cancer. However there was no evidence whether it could modulate the sensitivity of chemotherapy in breast cancers. In this manuscript we aimed to reveal the role of 53BP1 in modulating the sensitivity of breast cancers to 5-Fu.
First we investigated the relationship between 53BP1 levels and 5-Fu sensitivity in different breast cancer lines and found that high levels of 53BP1 could sensitize the response to 5-Fu. Then we used RNA interference and ectopic expression strategies to reveal the detailed function of 53BP1 in chemosensitivity. By MTT assay, we found that overexpression of 53BP1 resulted in more 5-Fu induced apoptotic cells and knockdown of 53BP1 attenuated the cell death of 5-Fu. Also 53BP1 could induce the G2/M cell cycle arrest in breast cancer cell lines. In consistent with the data, the in vivo experiments confirmed our in vitro results. We found that when the growth of tumors were inhibited by 5-Fu or overexpression of 53BP1 alone, the volumes of tumors were great significantly reduced when treated with combination of 53BP1 and 5-Fu.  Tumors formed by 53BP1 knockdown MCF-7 cells became more resistant to 5-Fu while the control tumors was dramatic inhibited. The tunnel staining also demonstrated that 53BP1 could additional enhance 5-Fu induced apoptosis while downregulating of 53BP1 conferred the resistance to 5-Fu. The development of drug resistance to 5-Fu impeded the clinical application. Previously studies have found that the effect of 5-Fu was significantly increased when TS was downregulated while the antisense downregulation of TS enhance the efficacy of 5-Fu [27][28][29][30]. Also the DPYD was another predictor of 5fluorouracil toxicity reason that conferring resistance to 5-Fu [31][32][33]. Horiguchi et al [34] also reported that DPYD was a prognostic predictor in breast cancer. So we next detected if thymidine phosphorylase and dihydropyrimidine dehydrogenase were involved in the 53BP1 mediated 5-Fu sensitivity. By detecting the TS and DPYD levels in the 53BP1 transfected cell lines, we found that 53BP1 overexpression could downregulate TS and DPYD at both mRNA and protein levels while knockdown 53BP1 could uprugulate the TS and DPYD levels. In order to confirm the role of TS and DPYD in the sensitivity of 5-Fu, the siRNA TS and siRNA DPYD were used. By using siRNA TS or siRNA DPYD, we found knockdown TS and DPYD reversed the sensitivity to 5-Fu in 53BP1 knockdown MCF-7 cells. However there was no effect on the MCF-7 control cells. These findings suggested that 53BP1 sensitizes breast cancer cells to 5-fluorouracil through TS and DPYD.
In conclusion, we shows that 53BP1 can enhance sensitivity of breast cancer cell lines to 5-Fu. To the best of our knowledge, this is the first study to demonstrate the role of 53BP1 in sensitivity to 5-Fu in breast cancer. Our findings opens a new clue for enhancing the efficacy of 5-Fu in breast cancers. This provides a new concept that the combination of 5-Fu and upregulation of 53BP1 maybe a promising targeted strategy for breast cancer in the future.

Author Contributions
Conceived and designed the experiments: QFY XYL. Performed the experiments: XYL XLK YW. Analyzed the data: XYL XNK SY. Contributed reagents/materials/analysis tools: XLK SY. Wrote the paper: XYL QFY.