Siah1, Siah-1L and the Siah1 mutant with the RING finger deleted that expresses Siah1ΔR were gifts from Prof. SI Matsuzawa, the Burnham Institute, USA [20]. The Tcf/Lef-responsive luciferase reporter gene (Topflash), the negative control with mutated Tcf/Lef binding site (Fopflash), and the Renilla luciferase reporter plasmid (pRL-TK) as an internal control were obtained from Upstate Biotechnology, USA. The human breast cancer cell lines BT-20, MCF7, MD-MBA231, SKBR3 and ZR75-1 were obtained from the American Type Culture Collection (Manassas, VA). To establish stable cell lines, SKBR3 cells (1.5 × 10⁵) were grown on 60 mm² culture dish containing RPMI-1640 medium supplemented with 10% FCS, 1% glutamine, 1% penicillin and streptomycin (Biochrom-Seromed, Germany) at 37°C one day before transfection. Siah1, Siah-1L or Siah1ΔR plasmids were transfected into cells using the FuGENE 6 transfection reagent (Roche, Mannheim, Germany) according to the manufacturer's protocol. Cells were allowed to rest for about 12 to 18 h after transfection, and the medium was changed with 800 μg/mL of G418 (Invitrogen) for 2 weeks. Stable clones were selected for G418 resistance and screened for Sial1 protein level with immunoblotting.

The small interfering RNA (siRNA) duplexes were designed and purchased from Qiagen (Cambridge, MA, USA). The sequences for the Siah1 siRNA were: Siah1 siRNA 5'-AACTCCTGCCTCCTTATGTATTT-3'. Siah1 siRNA-2, 5'-GAUAGGAACACGCAAGCAA-3'; Siah1 siRNA-3, 5'-GUUGCAUGUAGUAACACUA-3'. A control siRNA was used as well (no silencing): 5'-AAGAGCCGTCAGACTGCTACA-3'. Roti-Fectplus was used fro the transient transfections with SIAH1 siRNA, or control siRNA, according to the manufacturer's instructions (Carlroth, Karlsruhe, Germany).

Cells were washed in PBS and lysed in NP40 buffer containing 50 mM Tris-HCl, 1 mM PMSF, 150 mM NaCl, 1% NP-40 and protease inhibitor cocktail (Roche Diagnostics). After determination of the amount of protein in the cell lysates, samples were then resolved by polyacrylamide gel (PAGE) and electrophoretically blotted onto polyvinylidene difluoride membranes as described previously [20]. Membranes were incubated with anti-FLAG-M5 monoclonal antibody (1:500, Sigma Science, Hamburg, Germany). After incubation with a horseradish-conjugated goat anti mouse antibody, the protein content was visualized using enhanced chemiluminescence (Amersham, Freiburg, Germany). Siah1 and Siah2 protein expression was analysed as well using anti-Siah1 (YZ-12, Santa Cruz, CA, USA) and anti-Siah2 (N-14, Santa Cruz, CA, USA) primary antibodies, respectively together with the appropriate secondary antibodies. α-Tubulin monoclonal antibody was used as a control (1:500, Santa Cruz, CA, USA).

Single dose of 0, 2, 4, 6 or 8 Gy with a rate of 650 cGy/min was given using an Elekta Synergy linear accelerator with 6 MV-photons, at room temperature.

Total RNA was extracted from cultured cells and collected at different time points after IR using RNEasy protect mini kit (Qiagen, Germany), according to manufacturer's instructions. cDNA synthesis was performed in a 20 μl reaction volume with 0.5 μg of total RNA using cDNA synthesis kit (Fermentas, USA). The mixture was incubated at 37°C for 1 h followed by 10 min in 70°C for inactivation of reverse transcriptase. The primer sequences for amplification of Siah1 and Siah1L were described previously [21]. The primer sequences for amplification of Siah1ΔR and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were as follow: Siah1ΔR forward primer ACCTCGAAGTGTCCACCATC and reverse primer ACTGCATCATCACCCAGTCA (product size, 547 bp); GAPDH forward primer TGGTCACCAGGGCTGCTT and reverse primer AGCTTCCCGTTCTCAGCCTT (product size, 150 bp). PCR was done using a MyiQ detection system (Bio-Rad). The PCR conditions were 95°C for 5 min followed by 35 cycles of denaturation at 95°C for 30 seconds, annealing at 60°C for 30 seconds, and extension at 72°C for 1 min.

Transfected SKBR3 cells as well as MCF-7 after siRNA for Siah1, including a control siRNA, were plated into 100 mm² dishes and were irradiated (4 Gy). Apoptosis was analyzed 24 h after IR by flow cytometry using Annexin V-FITC apoptosis detection kit (Sigma Aldrich, Munich, Germany). Staining was performed at indicated time points after IR according to the manufacturer's instructions, and flow cytometry was conducted on a flow cytometer (BD FACScan). Statistical analysis was performed using WinMDI V2.9 software. Survival curves were obtained by means of standard colony formation assay. Transfected SKBR3 and MCF-7 cells were irradiated (1 × 10³) and plated onto 25-mL culture flash for 80 to 100 colonies per flash. After 14 days of incubation, colonies were fixed with 10% formalin and stained with crystal violet. Colonies with >50 cells were scored as a surviving colony. Surviving fraction was calculated as: (mean colonies counted)/(cells plated) × (plating efficiency), where plating efficiency was defined as (mean colonies counted)/(cells plated) for unirradiated controls. Experiments were conducted in triplicate and are presented as means ± standard deviation (SD) from three independent experiments. All survival fractions (SF) were fitted into the linear quadratic model. The ID₁₀ and SD for survival curves in SKBR3 cells were determined at 10% cell survival. The sensitizer enhancement ratio (SER) in SKBR3 cells was defined as SER = mean inactivation dose (radiotherapy)/mean inactivation dose (plasmid +radiotherapy), as previously described. SER>1 indicate radiosensitization [24-26].

Transfected SKBR3 cells as wells as MCF-7 cells, upon Siah1 siRNA, were seeded at a density of 1 × 10⁴ cells/well in a 96-well microplate and grown for 6 h and subsequently exposed to 0, 4 and 8 Gy IR. After 24 h of incubation at 37°C, the cell proliferation reagent WST-1 was added (10 μl/well) in SKBR3 cells for 30 min, according to the manufacturer's instructions (Roche, UK). The reaction product was quantified by measuring the absorbance using an ELISA reader (HTS 7000, Perkin-Elmer, Rodgau, Germany) and Software HT-Soft (Perkin-Elmer, USA). The experiments were performed three times in triplicate and presented as means (± SD). Cell viability curves were fitted using the linear quadratic model and the dose resulting in 50% of cell growth inhibition (IC₅₀) was calculated, as previously described [25,26]. The cell viability of MCF-7 cells was measured at 2, 4 and 6 days using the same method, as described above. Cell viability was normalized to the untreated group. A control siRNA group was included as well.

Stable transfected SKBR3 cells (5000) were seeded on cover slips and incubated for 12 h and either unirradiated or irradiated with X-ray (4 Gy). At 3 and 6 h after IR, cells were fixed in cold methanol (-20°C) for 5 min, permeabilized with 1% Triton/PBS for 10 min and then blocked in PBS containing 0.3% Triton X-100, 3% BSA for 1 h. Fixed, permeabilized cells were incubated with in 100 μL mouse monoclonal anti-γH2AX antibody at 1:100 dilution (Upstate Biotechnology, Waltham, MA) for 30 min at room temperature. This was followed by incubation with Alexa 488 conjugated goat anti-mouse secondary antibody (Molecular Probes, Eugene, OR, 1:200 dilutions) for 1 h at room temperature. Finally, cells were counterstained with 4',6-Diamidino-2-phenylindole (DAPI, 0.1 μg/mL). The sections were mounted with fluorescent mounting medium (Dako, Germany). The fluorescent images were examined on an Olympus JMT-2 photomicroscope (Hamburg, Germany). For each treatment condition, γH2AX foci were determined in at least 100 cells.

In vitro cellular invasion was determined the ability of cells to invade a synthetic basement membrane (BD Biosciences, Heidelberg, Germany). Stably transfected SKBR3 cells (2 × 10⁴) and MCF-7 cells (5 × 10⁴), upon Siah1 siRNA, were diluted in 350 μl medium and transferred into precoated Matrigel membrane filters (8 μm pore size). The inserts were placed in the bottom chamber containing RPMI 1640 medium supplemented with 5% FCS. Following 48 h incubation at 37°C, filters were fixed in 100% methanol for 15 min, air dried, and stained with haematoxylin stain for 2 min. Noninvading cells on the upper surface were removed with a cotton swab, whereas invading cells on the underside of the filter were counted using an inverted microscope. All experiments were done in triplicate and a minimum of 10 individual fields at ×10 magnification per filter was counted.

Luciferase reporter assays were done at least in triplicate as described [20]. SKBR3 cells (4 × 10⁴) as well as MCF-7 cells (4 × 10⁴), after siRNA, were grown on 24-well tissue culture plates. In each well, transient DNA transfection was carried out with 50 ng DNA of the Tcf/Lef-responsive luciferase reporter gene (Topflash) or a negative control (Fopflash), 5 ng of the Renilla luciferase reporter plasmid (pRL-TK) as an internal control for the normalization of transfection efficiency and 50 ng of the indicated expression plasmids or an empty control vector. After 24 h, the luciferase and Renilla activity was measured using the dual luciferase assay system (Promega, Mannheim) and the luciferase activity was normalized against the Renilla luciferase activity and are presented as means ± SD from three independent experiments. Together with the Siah1 siRNA group, a control siRNA was tested as well in MCF-7 cells.

Quantitative data were expressed as means ± SD. The significance of differences between the means was assessed using one-way ANOVA or student t-test using SPSS 15.0 software. Statistically significant difference was considered as P < 0.05.

The expression of Siah1 and its splice variant Siah1L in breast cancer cell lines was determined using RT-PCR with specific primers recognizing the common sequences of both siah1 and siah1L. In line with previous report [10,27], MCF7 breast cancer cells exhibited strong expression of Siah1 and Siah1L, whereas BT-20, MDA-MB-231, SKBR3 and ZR75-1 breast cancer lines lacked any expression of Siah1 and Siah1L analysed (Figure 1A). The SKBR3 cells without detectable endogenous expression of Siah1 and Sial1L were chosen to establish stable gene expression. Stable clones were selected for G418 resistance and screened for relative equivalent levels of exogenous expression of Siah constructs detected by using immunoblotting (Figure 1B). Additionally, we performed siRNA for Siah1 in MCF-7 to further investigate its role in radiosensitization. Western blot confirmed blockade of Siah1 in MCF-7. Off-target effects can occur during siRNA [28] and therefore two additional Siah1 siRNAs (siRNA-2 and siRNA-3) were tested as well. They also confirmed blockade of Siah1 protein in MCF-7 cells. Control siRNA was used as well (Figure 2A). Siah2 protein expression was absent in both SKBR-3 and MCF-7 cells (Figure 2B), as previously demonstrated [29].

Whether Siah1 may have an effect on radiation response has not yet been addressed. We therefore investigated the effects of Siah1 and its splice variants on breast cancer cells following IR. We found that overexpression of Siah1 and Siah1L significantly enhanced IR-induced apoptosis in SKBR3 cells compared with the control transfected with empty vector while Siah1 ΔR showed a small but non-significant difference (Figure 3A, control, 44.5 ± 7.9; Siah1ΔR, 53.4.2 ± 2.9; Siah1, 62.3 ± 3.5; Siah1L, 70.9 ± 1.5). Notably, Siah1 and Siah1L increased apoptosis also in unirradiated group (p < 0.05; control, 12.2 ± 9.5; Siah1ΔR, 27.2 ± 13.0; Siah1, 52.3 ± 5.6; Siah1L, 52.5 ± 4.5). Importantly, siRNA for Siah1 resulted in reduced apoptosis in response to IR as compared to the untreated group (Figure 3B, untreated, 31.27 ± 3.9; control siRNA 29.87 ± 4.5; Siah1 siRNA, 17.04 ± 2.1) while in unirradiated group no difference was observed (Fig 3B). To further analyze the radiosensitizing ability of Siah1 and Siah1L, we used clonogenic assay to assess survival of stably-transfected cells after IR. Importantly, overexpression of Siah1L and Siah1 cooperated with IR to reduce clonogenic growth of SKBR3, as compared with irradiated control cells (p < 0.05; Figure 4A). As illustrated in Table 1, a SER₁₀ of 1.87 was observed with Siah1L gain-of-function, indicating a highly cooperative and potentially synergistic fashion between Siah1L and IR. Siah1 induced a rather additive than synergistic radiosensitizing effect (SER₁₀ of 1.26) while no significant difference in survival was noted with Siah1ΔR (Table 1). In addition, Siah1 and Siah1L reduced clonogenic survival also in unirradiated cells, as compared with the control group. Loss-of-function for Siah1 by siRNA demonstrated a significant reduction in the radiosensitivity of MCF-7 cells, even at a dose of 2 Gy, as assessed by clonogenic survival (Figure 4B; p < 0.05). To exclude the possibility of "off-target" effects [28], we tested the impact of Siah1 silencing in radiosensitivity of MCF-7 cells by using two additional siRNAs (Siah1 siRNA-2 and Siah1 siRNA-3). The latter revealed results similar to the main Siah1 siRNA used (Figure 4B), whereby a significant decrease (p < 0.05) in radiosensitivity was detected.

To determine the effect of Siah1 on the cell viability of SKBR3 cells after IR, SKBR3 cells were irradiated with different single doses (0, 4 or 8 Gy) and incubated for 24 h. The overexpression of Siah1 and Siah1L resulted in a significant dose-dependent reduction of the proliferation of unirradiated as well as irradiated cells, as compared with control vector (Figure 5A, p < 0.05). Potentially synergistic enhancement of radiocytotoxicity with a SER₅₀ of 4.65 for cell viability was observed when Siah1L was combined with IR (Table 2). The proliferation assay indicated that the expression of Siah1 also results in increased sensitivity to ionizing radiation but less potently. As shown, Siah1ΔR exhibited no effect with a SER₅₀ of 0.99 (Table 2). Moreover, Siah1 siRNA revealed an increase in MCF-7 cell viability at 2, 4 and 6 days upon suppression of Siah1 activity (Figure 5B; p < 0.05). Control siRNA was used as well.

To gain insight into the molecular mechanisms of radiosensitization of Siah and to investigate its effect on the initial DNA damage response to IR we monitored the fate of double-strand breaks (DSBs) in stable transfected cells after IR (4 Gy) using γH2AX, a landmark marker of DSBs [30,31]. Induction of γH2AX foci was significantly increased in Siah1- and Siah1L-transfected cells at 3 h post-IR compared with cells transfected with empty vector (p < 0.01; Figure 6A). Furthermore, Siah1L prolonged the persistence of γH2AX foci at relatively high levels at 6 h post-IR as compared with control cells, suggesting inhibition of DNA repair (p < 0.01; Figure 6A and 6B).

In an effort to define the significance of expression of Siah1 on the biological behaviour of breast cancer, we performed cell invasion study with a modified Boyden chamber assay. Siah1 and Siah1L gain-of-function in SKBR3 cells, upon transfection, significantly reduced the ability of these cells to invade Matrigel compared with control vector (p < 0.05; Figure 7A). Siah1ΔR transfection displayed a less potent but still significant effect (control, 19.5 ± 2.3; Siah1ΔR, 14.4 ± 3.5; Siah1, 7.8 ± 0.7; Siah1L, 4.6 ± 1.4); (Figure 7A and 7B).

Siah1 siRNA displayed a significant increase of the number of invading MCF-7 cells, as compared with the untreated and the control siRNA group (untreated, 38 ± 4; control siRNA, 40 ± 8; Siah1 siRNA, 147 ± 16) (p < 0.05; Figure 8A and 8B).

Previous studies have demonstrated the role of Siah-1 in degradation of β-catenin and in reduction of activity of Tcf/Lef transcription factor [11,13]. To examine the effect of Siah1 and Siah1L on Tcf/Lef activity in SKBR3 cells, we performed a luciferase reporter assay. In this assay, a reporter plasmid with a mutated Tcf/Lef-binding site was used as negative control. Siah1L gain-of-function significantly decreased Tcf/Lef activity while Siah1 showed a lower but significant effect, as compared with control SKBR3 cells (p < 0.05; Figure 9A). Tcf/Lef activity was not affected by overexpression of Siah1ΔR (control, 20.5 ± 2.2; Siah1, 15.7 ± 0.9; Siah1L, 5.5 ± 2.9; Siah1ΔR, 19.0 ± 1.6) (p > 0.05; Figure 9A). As expected, Siah1 siRNA demonstrated increased Tcf/Lef activity in MCF-7 cells, as assessed by the luciferase assay (Figure 9B, p < 0.05).