All methods were performed in accordance with the relevant guidelines and regulations.
Cell culture. MDA-MB-231 (MM231) and retinal pigment epithelium (RPE1) cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) containing 10% fetal bovine serum (FBS) and 1% Penicillin-Streptomycin. Five types of cells were derived from MM231. Derived cells were MM231 with DMSO (MM231-DMSO), MM231 with paclitaxel for short term (PTX-short) (1µM), MM231 with paclitaxel for long term (PTX-long) (1 nM), MM231 with eribulin for short term (ERI-short) (10 nM), MM231 with eribulin for long term (ERI-long) (0.2 nM). Three types of cells were derived from RPE. Derived cells were RPE with DMSO (RPE-DMSO), RPE with paclitaxel for short term (PTX-short) (1 µM), RPE1 with eribulin for short term (RPE-ERI) (10 nM). PTX-short and ERI-short were incubated for 24hours, while PTX-long and ERI-long were incubated for 60days. All cells were maintained at 37℃ in a 5% CO2 incubator. ERI and PTX were dissolved in DMSO. ERI was stored at -80℃, PTX was stored at -20℃.
Reagents and antibody. ERI was from Fuji Film and PTX was from ADIPOGEN. The following antibodies were used for WB: cGAS (#15102S, 1:1000), STING (#13647, 1:1000), pIRF3 (#4947, 1:1000), Histone H3 (#9715, 1:1000), RAD51 (#D4B10, 1:1000), anti-rabbit IgG (#7074, 1:2000) and anti-mouse IgG (#7076, 1:2000) were from Cell Signaling Technology (CST), IFNβ (#PA5-20390, 1:1000) was from Invitrogen and Vinculin was from Santa cruz (#A1121, 1:1000). The following antibodies were used for IHC: STING (#13647, 1:400) was from Cell Signaling Technology and IFNβ (#PA5-20390, 1:400) was from Invitrogen. The following antibodies were used for IF: cGAS (#79978S, 1:500) was from CST, RAD51 (#ab133534, 1:500) was from abcam. IFNβ (#PA5-20390, 1:400) was from Invitrogen and LAP2 (#8197900, 1:1000) was from BD Biosciences. DNA Damage Detection Kit (#340–09431) was from Fuji Film.
Tubulin Polymelization Assay. The tubulin polymerization experiment was performed as per the reported protocol described in the assay kit (#BK006P). Tubulin protein (3 mg/mL) was incubated with tubulin polymerization buffer in pre-warmed 96-well microtiter plates at 37°C in the presence of 0.5 nM, 5 nM, 10 nM ERI and 1 µM PTX. Then absorbance was monitored continuously for 1 h at 340 nm.
Live cell imaging and image analysis. RPE1 cells stably expressing Histone H2B-RFP or MDA-MB-231 cells stably expressing Histone H2B-mCherry were plated on 4-chaamber 35 mm glass bottom dish at least one day prior to do imaging (#1.5 glass, Cellvis). Cells were treated with DMSO (control), PTX (10 nM), or ERI (10nM) 1 hour prior to imaging. High-temporal live-cell imaging was performed using a Nikon Ti2 inverted microscope equipped with a Hamamatsu Fusion camera, spectra-X LED light source (Lumencor), Shiraito PureBox (TokaiHit) and a Plan Apo 20x objective (NA = 0.75) controlled by Nikon Element software and Metamorph (Molecular Devices). Cells were recorded at 37°C with 5% CO2 in a stage-top incubator using the feedback control to maintain the growing media’s temperature (Tokai Hit, STX model). Image analysis was performed using Nikon Element software. Mitotic stages were determined by nuclear staining. The mitotic duration was measured from nuclear envelope breakdown (NEBD) to anaphase onset. Incidences of multi-nuclei, mitotic slippages and unaligned chromosome were analyzed. The experiments were independently repeated 2–3 times for mitotic duration measurements (total of n = 100), and P-values between variants were calculated by One-Way Anova and two-tailed t-test. P-values < 0.05 were considered significant.
Immunofluorescence (IF). MM231 cell lines(DMSO, PTX-short, PTX-long, ERI-short, ERI-long) and RPE cell lines (DMSO, PTX-short, ERI-short) were plated on the cover glass of 6-well plates at approximately 5×104 cells /well. After cells adhesion, the cells were treated with DMSO or PTX or ERI for 24 h. Then Cells were fixed with 4% paraformaldehyde in PBS for 15 minutes at RT, permeabilized and blocked with 2.5%FBS, 0.2M glycine and 0.1% TritonX-100/PBS overnight at 4℃. The primary antibodies were diluted in PBS with 1% BSA, and incubated 1 hour at RT. The secondary antibodies were dissolved in PBS and incubated 30 min at RT with blocking out light. Cell nuclei were stained with DAPI followed by imaging using a Zeiss AxioCam HRC microscope camera using a 60x objective lens. A DNA Damage Detection Kit was used as the protocol. Characteristic cells were independently counted in each of the 10 fields of vision for each stain and statistically compared by unpaired t test.
Western blotting (WB). Cell lysates were extracted with RIPA bufer (#16488-34, Nacalai) containing protein inhibitor (#A32955, Thermo Fisher). Proteins (20 µg) were resolved by SDS-PAGE using a 15% XV PANTERA Gel, and transferred to Immobilon-P PVDF membranes. After blocking with 5% skim milk for 60 min, except phosphorylated protein blocked with 5% Bovine serum albumin, the membranes were immersed with diluted primary antibody and shaked for overnight at 4°C, followed by shaking with secondary antibody for 1 hour at RT. Proteins were visualized using Chemi-Lumi One Super (#02230) or Chemi-Lumi One Ultra (#11644) which are chemiluminescent substrates. LAS4000 UV mini were used for blot detection. The lysates and the antibodies were washed by using 5% TBST buffer.
Cell fractionation assay. The Thermo Scientific™ NE-PER™ Nuclear and Cytoplasmic Extraction Reagents (#78833) was used for separation of cytoplasmic and nuclear extracts from MM231 cell lines (DMSO, PTX-short, PTX-long, ERI-short, ERI-long). Nuclear components and cellular components were extracted from each cells following the protocol (https://www.thermofisher.com/order/catalog/product/78833). Then we evaluated the differences of cGAS expression between nuclear components and cellular components by WB.
SiRNA transfection. The cGAS specific siRNAs (Hs_C6orfl 50_8, Hs_C6orfl 50_7, Hs_C6orfl 50_6, Hs_C6orfl 50_5, QIAGEN) (5 nmol/L) and negative control siRNA (QIAGEN) (5 nmol/L) were transfected into MM231 cells by using Lipofectamine RNA iMAX Reagent (Invitrogen).
RT-PCR. RNA was extracted from MM231 cell lines according to the manufacture’s protocols of the RNeasy Mini Kit (QIAGEN). The RNA was reverse transcribed to cDNA using SuperScript VILO cDNA synthesis Kit and Master MixRNA (Thermo). PCR reaction solution was prepared using TB Green Fast qPCR Mix (Takara Bio) and human cGAS primers (5'-TGCAAAGGAAGGAAATGGT-3′ and 5′-TTTAAACAATCTTTCCTGCAACA-3′). PCR reactions were performed on an Applied Biosystems 7300. To evaluate the relative expression of proteins, the 2-ΔΔCT method was used to compare with control.35
Proliferation assay. At 24h after siRNA (Hs_C6orfl 50_5, negative control siRNA) transfection into MM231, the cells were detached and seeded at 5×104 cells/ml (set as 0 hour). Then DMSO, PTX, and ERI were added to both MM231 cells and KD-cGAS MM231 cells. Cell proliferation was assessed by measuring absorbance at 450 nm using Cell count Reagent SF (Nacalai Tesque) and a plate reader (Bio-Rad Laboratories).
Tripan blue assay. The cell lysates from MM231 cell lines (DMSO, PTX-short, ERI-short, KD-cGAS-DMSO, KD-cGAS-PTX, KD-cGAS-ERI), which were 24 hours after PTX or ERI treatment, were extracted and diluted equally by tripan blue (Nacalai, #20577-34). The total number of cells and the viable cells were counted by Bio RAD TC20 Automated Cell Counter. The procedure was performed three times independently.
Patients and samples. All clinical samples were obtained from breast cancer patients who assigned to randomized Phase Ⅱ JONIE-3 clinical trial (UMIN000012817). In this multicentre randomised study, 121 patients were diagnosed with invasive breast cancer and performed neoadjuvant chemotherapy (NAC) between December 2013 and April 2016. The patients were randomly assigned to 2 different NAC groups: ERI group (eribulin followed by fluorouracil, epirubicin, and cyclophosphamide; FEC) or PTX group (paclitaxel followed by FEC). The patients of both groups were performed biopsy before and after chemotherapy. In the trial, 115 cases were analyzed for safety. We defined pCR as no invasive residual cancer in the breast. JONIE3 clinical trial protocol was approved by the institutional review board (IRB) of Tokyo Medical University on Dec. 25, 2013 (approval number: SH2588) and subsequently by all participating institutions. No tissues were procured from prisoners or vulnerable groups. Both JONIE3 clinical trial and this study were approved by the IRB at department of medicine, Chiba University (approval number: M10112) and all methods were performed in accordance with the Declaration of Helsinki and domestic relevant guidelines and regulations. Study participant names and other HIPAA identifiers are not included in all texts/figures/tables/images.
Immunohistochemistry (IHC). IHC was performed using tissue samples obtained by JONIE-3 clinical trial.2 We can evaluate 56 clinical samples in out of 115 cases. Four cases were excluded because it was insufficient for evaluation of IHC, 55 cases were excluded because we cannot permit using the samples. For IHC staining, tissue samples were thin sliced at 4 µm thickness. Antigen retrieval was performed by autoclaving for 25 min, and endogenous peroxidase activity was inactivated with 3% hydrogen peroxide. Following nonspecific protein blocking with 5% BSA except for using STING antibody blocked with 5% skim milk, the slides were stained with a cGAS or IFNβ or RAD51 antibody. The sections were incubated overnight with primary antibodies at 4°C. The sections were incubated with primary antibodies overnight at 4°C, and stained with secondary antibodies (DAKO, #k4003112) for 30 min at room temperature, followed by staining with diaminobenzidine for 5 min (Nakalai Tesque). The IHC was evaluated by H-score at hot spot.24 H-score was calculated by adding the percentage of positive cells multiplied by the weighted intensity of staining: H-score= (1×% weak positivity)+ (2×% medium positivity)+ (3×% strong positivity). A positive cGAS-score was defined as beyond 80, a positive STING-score was defined as beyond 120, a positive IFNβ-score was defined as beyond 180 and a positive RAD51-score was defined as beyond 100. The H-scores for each group were calculated and compared using the unpaired t-test. Simple linear regression was used to examine the correlation between the score of cGAS and IFNβ, or cGAS and RAD51.
Statistical analysis. Statistical analysis were performed using Graphpad Prism version 9 statistical software. In all analysis, P < 0.05 was judged statistically significant.