SETD1A-SOX2 axis is involved in tamoxifen resistance in estrogen receptor α-positive breast cancer cells

Rationale: Approximately 30-40% of estrogen receptor (ER)-positive breast cancer (BC) cases recur after tamoxifen therapy. Thus, additional studies on the mechanisms underlying tamoxifen resistance and more specific prognostic biomarkers are required. In this study, we investigated the role of the SET domain containing 1A (SETD1A), a histone H3-lysine 4 (H3K4) methyltransferase, in the development of tamoxifen resistance in BC. Methods: The relationship between tamoxifen resistance and SETD1A protein level was investigated using resistant cell lines derived from the parent BC cells. Biochemical and molecular assays, such as RNA-sequencing, reverse transcription-quantitative polymerase chain reaction, chromatin-immunoprecipitation, and protein-binding assays, were used to identify the SETD1A target gene in tamoxifen-resistant BC cells. Additionally, the role of SETD1A in cancer stem cells (CSCs) was investigated using CSCs isolated from tamoxifen-resistant BC cells. Comprehensive transcriptome analysis and immunofluorescence staining using clinical datasets and tissue microarray were performed to determine the correlation between the expression of the SETD1A-SRY-box transcription factor 2 (SOX2) pair and recurrence in tamoxifen-treated patients with BC. Results: SETD1A was expressed at higher levels in tamoxifen-resistant BC cells than in primary BC cells. Notably, SETD1A-depleted tamoxifen-resistant MCF-7 cells showed restored sensitivity to tamoxifen, whereas SETD1A overexpression in MCF-7 cells resulted in decreased sensitivity. SETD1A is recruited to the SOX2 gene via its interaction with SOX2, thereby enhancing the expression of SOX2 genes in tamoxifen-resistant BC cells. The growth of tamoxifen-resistant cells and CSCs was effectively suppressed by SETD1A knockdown. In addition, high levels of SETD1A and SOX2 were significantly correlated with a low survival rate in patients with ER-positive tamoxifen-resistant BC. Conclusion: Our findings provide the first evidence of the critical role of the SETD1A-SOX2 axis in tamoxifen-resistant BC cells, implying that SETD1A may serve as a molecular target and prognostic indicator of a therapeutic response in patients with tamoxifen-resistant BC.


Cell proliferation assay
Cells seeded in a 24-well plate (1 × 10 4 cells/well) plate and incubated at 37 °C with 5% CO2 were transferred to IncuCyte® ZOOM system (Essen Bioscience, Ann Arbor, MI, USA), which allowed automated live cell analysis within incubator. Images were captured at 4-h intervals using IncuCyte ZOOM software (version 2013B).

Soft agar assay and 3D cell culture
For soft agar assays, TamR cells expressing either shNS or shSETD1A were plated in DMEM and agarose (0.8 % top agarose and 1 % basic agarose). After 30 days of culture, colonies were stained with crystal violet and observed under a microscope to measure the size and number of each colony.
For 3D spheroid culture, Prosys ® StemFit 3D (SF, Prodizen Inc., Seoul, Korea) was used. TamR cells expressing either shNS or shSETD1A were plated at a density of 1.0 × 10 6 cells/well. After incubation for 24 h, the aggregated cells were observed under a microscope.

Cell migration and invasion assay
Migration assay was conducted using a 24-well Transwell (Costar 3422; Corning Inc., Corning, NY, USA). Invasion assay was performed using BioCoat™ Matrigel® Invasion Chamber (354480, Corning Inc., Corning, NY, USA). Briefly, after transfection, 100 μL of serum-free DMEM and 200 μL of TamR cells (1.5 × 10 5 cells/mL) suspended in serum-free DMEM were added to the upper chamber. The cells were allowed to migrate for 24 h or invade for 48 h in the lower chamber, which contained 750 µL of the medium supplemented with 10 % FBS as a chemoattractant. Then, the cells were stained with 0.1 % crystal violet for 15 min, and washed with PBS. The cells remained in the upper chamber were removed with cotton wool, whereas the cells that had migrated or invaded were imaged with Nikon TS100 stereomicroscope coupled to Canon G10/G11 camera (Canon, Tokyo, Japan). Each insert captures at least three random field images, quantified with the ImageJ program.
The animals were randomly assigned to separate experimental groups (n = 10). No samples or animals were excluded from the analysis. The investigators were not blinded to group allocation.
This study was conducted under ethical approval from the Institutional Animal Ethics Committee of Keyfron Bio Co., Ltd. (Cheongju, Korea)

RNA interference and RT-qPCR
Transfection of siRNA was performed using Oligofectamine (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's protocol. The sequences of siRNAs and shRNAs are listed in Table   S1. TamR cells expressing control shRNA or shRNA targeting SETD1A were generated via lentiviral transfection (Sigma-Aldrich, St. Louis, MO, USA). Inducible human SETD1A shRNA system was generated using mCMV-TurboGFP human SETD1A shRNA (Dharmacon, Lafayette, CO, USA). shRNA expression was induced through the addition of doxycycline (Dox, 0.5 µg/mL) for subsequent assay. Total RNA was isolated from TamR cells using TRIzol reagent (Invitrogen, Carlsbad, CA, USA), and cDNA was synthesized using iScript (Bio-Rad Laboratories, Hercules, CA, USA). RNA concentration was assessed spectrophotometrically using NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). The reverse transcription product (2 μL) was used for real-time qPCR analysis using LightCycler 480 II instrument (Roche, Indianapolis, IN, USA) with the primers listed in Table S2. Relative gene expression levels were normalized to 18S rRNA levels.

Chromatin immunoprecipitation assay
Chromatin immunoprecipitation (ChIP) assay was performed according to a previously described  Table S2.

FAIRE-qPCR
FAIRE-qPCR was performed as previously described [2]. Briefly, after reaching approximately 90% confluence, shRNA-transfected TamR cells were cross-linked with formaldehyde, and extracts were prepared. Results are expressed as the percentage of input chromatin (input DNA). The primer sequences used for FAIRE-qPCR are listed in Table S2.