Human mammary carcinoma cell lines MDA-MB-231 (HTB-26™) and MCF-7 (AHTB-22™) were obtained from the American Type Culture Collection (Manassas, VA, USA). MDA-MB-231 cells express none of hormone receptors, so they are classified as TNBC and MCF-7 cells express ER. Rabbit anti-phospho-Akt Ser473, rabbit anti-Akt, rabbit anti-phospho-p90RSK Ser380, rabbit anti-phospho-ERK1/2, and rabbit anti-ERK1/2 antibodies were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA). Immobilon-P (#IPVH00010) were purchased from Merck Millipore (Burlington, MA, USA). 10X Phosphate buffered saline (PBS, #EBA-1105) and reveres transcription 5X master mix (#EBT-1511) were purchased from ELPIS-BIOTECH (Daejeon, South Korea). Tri-RNA reagent (#FATRR-001) was purchased from Favorgen (Pingtung, China). Cis-DDP (#C2210000), 2-mercaptoethanol (#M6250), and skim milk (#70166) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Clarity western ECL substrate (#170-5061) and iQ SYBR green supermix (#170-8882AP) were purchased for BIO-RAD (Hercules, CA, USA). EzReprobe (#WSE-7240) and EzRIPA Lysis kit (#WSE-7240) were purchased for ATTO corporation (Tokyo, Japan). Dulbecco's modified Eagle's medium (DMEM, #11963-092), Fetal bovine serum (FBS, #26140-079), Penicillin streptomycin (P/S, #1570-063) and 0.25% Trypsin/EDTA (#25200-072) were purchased from Gibco (ThermoFisher Scientific; Waltham, MA, USA). 3-(4,5-dimethylthiazole-2-yl)-2 and Lipofectamine 3000 (#L3000-008) were purchased from Invitrogen (ThermoFisher Scientific). FMK was purchased from Axon Medchem (#Axon 1848, Reston, VA, USA).

Cell viability was determined using 3-(4,5-dimethylthiazole-2-yl)-2, 5-diphenyltetrazolium bromide assay (MTT, #M6494, Sigma-Aldrich) assay. For the assay, cells were seeded in 96-well at a density of 1 × 10⁴ cells per well. After 24 h of stabilization, the cells were treated with various dose of Cis-DDP for 24 h. Cells were treated with MTT (12 mM) in FBS-free DMEM media and incubated for 2 h at 37℃ and 5% CO₂. The medium was the removed and the formazan precipitate was extracted in DMSO. The absorbance was measured at 540 nm on a microplate reader (TECAN, Männedorf, Switzerland).

The Flow cytometry analysis was used to measure cell viability using a Muse count & viability assay kit (#MCH100102, Merck Millipore) according to the manufacture's protocol. Briefly, MDA-MB-231 and MCF-7 cells were seeded in 6-well plate at a density of 1 × 10⁶ cells/well. For cell proliferation experiment, cells were harvested after 24-36 h treated with Cis-DDP. Cells were directly stained by reagent provided by kit and the cell viability analysis was performed using a Muse cell analyzer (EMD Millipore) and data were analyzed using a MUSE 1.5 analysis program.

Western blotting analysis were performed as described previously (17). Briefly, total protein samples were separated by SDS-PAGE and transferred to nitrocellulose membranes. Western blotting was performed using an each corresponding specific antibody. Polyclonal mouse anti-α-tubulin (Sigma-Aldrich) was used as an internal control.

Plasmids including pcDNA and dominant negative (DN)-p90RSK were transfected with EndoFectin™ Max transfection reagent (#EF014, GeneCopoeia, Rockville, VA, USA) according to the manufacture's protocol. Rat RSK1 (NM031107) was mutated to K94A/K447A to create a kinase dead protein (DN-p90RSK1) with the QuickChange II site-directed mutagenesis kit (#200521, Agilent, Santa Clara, CA, USA) as described previously (17).

The quantitative real-time polymerase chain reaction (qRT-PCR) assay was used to analyze the mRNA expression of E-cadherin, N-cadherin, Snail, and Twist as described previously (17). The relative gene expression was calculated using the 2^−Δct method, and GAPDH was used for normalization. All primer sequences used in qRT-PCR experiments are listed in Table 1.

To knockdown p90RSK in vitro, specific short interference RNA (siRNA) was used (#SC-29475, Santa Cruz Biotechnology, Santa Cruz, CA, USA). A siRNA that does not match any known human coding cDNA was used as a negative control for silencing (Scramble, sc-37007, Santa Cruz Biotechnology). Transfections were performed using Lipofectamine 3000 (Invitrogen) according to the manufacturer's instructions.

Immunofluorescence assay was performed as described previously (17). Briefly, cells were fixed with 4% paraformaldehyde for 15 min at RT and permeabilized with 0.1% Triton X-100 for 20 min. Cells were incubated with the primary antibodies (ZO-1 1:500) overnight at 4℃ followed by incubation with anti-rabbit secondary antibodies (Invitrogen) conjugated with Alexa 488 at a dilution of 1:400 at RT for 1 h. Cell nuclei were counterstained with 40, 6-diamidino-2-phenylindol (DAPI) for 5 min. Slides were mounted with prolong gold antifade mount reagent (#P36930, Invitrogen) and examined with a laser scanning confocal spectral microscope (K1-Fluo, Nanoscope systems, Daejeon, South Korea). Representative images were automatically taken using a SPOT digital camera.

Statistical analysis was performed using GraphPad Prism 5 (version 5.02, GraphPad Software Inc., San Diego, CA, USA). One-way analysis of variance (ANOVA) followed by a Bonferroni multiple comparison was performed. A p value <0.05 was considered significant. All experiments were expressed as the mean ±SEM and were performed independently at least 3 times.

We evaluated Cis-DDP resistance on cell viability in MDA-MB-231 and MCF-7 cells. Cell viability tests were performed by flow cytometry using Muse count & cell viability kit. The serum starved breast cancer cells were treated with various concentrations of Cis-DDP (1, 5, 10, and 20 µg/mL) for 36 hr. Cis-DDP treatment significantly induced cytotoxicity in MCF-7 cells, whereas MDA-MB-231 exhibited the resistance to cell viability after Cis-DDP treatment (Fig. 1).

Previous report demonstrated that overexpression of p90RSK was shown in breast cancer tissues (8). To identify the p90RSK-mediated signaling pathway associated with Cis-DDP resistance to MDA-MB-231 cells, cells were transfected with siRNA of control (si-Cont) or p90RSK (si-RSK1) for 48 hr. As shown in Fig. 2A, Cis-DDP treatment increased activation of p90RSK, whereas depletion of p90RSK inhibited Cis-DDP induced both p90RSK activation and expression. In addition, si-RSK1 transfected cells significantly sensitized cells to Cis-DDP induced cytotoxicity compared with si-Cont transfected cells (Fig. 2B). Next experiment was confirmed the effect of p90RSK activation on Cis-DDP induced cytotoxicity using transfected cells with plasmid of wild type (WT)-p90RSK and DN-p90RSK. WT-p90RSK induced Cis-DDP resistance to cytotoxicity of MDA-MB-231 cells, whereas, DN-p90RSK sensitized the cells to Cis-DDP induced cytotoxicity via deactivation of p90RSK (Fig. 2C). Together, we demonstrated the important role of p90RSK activation in Cis-DDP resistance in TNBC cytotoxicity.

It has been known that EGF signaling pathway was related to cancer cell survival and drug resistance in breast cancer cells (18). Therefore, we examined whether Cis-DDP resistance was associated with up-regulation of EGF expression in MDA-MB-231 cells using qRT-PCR analysis. As shown in Fig. 3A, Cis-DDP increased mRNA expression level of EGF in a dose-dependent manner. Next, we investigated if p90RSK activation is involved in EGF-mediated cancer cell survival signaling pathways using a FMK, which is a p90RSK specific inhibitor. Recombinant protein of human EGF (hEGF) treatment induced phosphorylations of ERK1/2, p90RSK, and Akt at 5 min, and these maximum phosphorylations were seen at 60 min, respectively (Fig. 3B). Interestingly, FMK pretreatment strongly abolished hEGF-induced phosphorylation of p90RSK, but not phosphorylations of ERK1/2 and Akt. It suggests that FMK specifically inhibited only p90RSK activation and Cis-DDP resistance to TNBC is associated with EGF-induced p90RSK activation.

It has been reported that anti-cancer drug resistance was related to EMT (319). We examined the role of p90RSK activation on expression of EMT-associated proteins using a qRT-PCR analysis. MDA-MB-231 cells were transfected si-Cont or si-RSK for 48 h followed by treatment with 0, 10, and 20 µg/ml of Cis-DDP for 6 h. As shown in Fig. 4A-D, Cis-DDP treatment induced EMT by increasing the mRNA expression levels of N-cadherin, Snail, and Twist, while decreasing the expression levels of E-cadherin. The condition of depletion of p90RSK resulted in decreased levels of Cis-DDP-induced EMT markers and increased level of Cis-DDP-inhibited E-cadherin. We further examined epithelial marker, ZO-1 using immunofluorescence analysis and found that Cis-DDP-inhibited ZO-1 expression was recovered by p90RSK deactivated condition (Fig. 4E). Together, our results indicats that p90RSK activation is involved in the regulation of EMT for causing Cis-DDP resistance in TNBC.