Transcriptome profiling reveals Silibinin dose-dependent response network in non-small lung cancer cells

Cell culturing

Human lung adenocarcinoma cell lines H1650, H1975, A549, H838, H2030 were obtained from ATCC and cultured in DMEM (Dulbecco’s Modified Eagle Medium) medium supplemented with 10% fetal calf serum (FCS) and 100 U/ml penicillin G and 100 µg/ml streptomycin sulfate at 37 °C in a humidified 5%CO₂ incubator. To ensure the ample number of cell count, cells were cultured three days prior to cell seeding. When the culture plates became confluent, NSCLC cells were detached and collected in a sterile Falcon tube with a pre-warmed medium for transfection. Cells were subjected to centrifugation (at 1000 rpm and room temperature for 5 min) and the supernatant was aspirated. The cells were resuspended in the medium for transfection. The number of cells/ml was determined by using Neubauer counting chamber and cells were diluted to get a concentration of 12000 cells/100 µl of the medium. After careful optimization of transfection efficiency, 12000 cells/well in 100 µl was found to show good transfection rate.

Silibinin and Withaferin-A compounds

Silibinin (C₂₅H₂₂O₁₀, mol. weight 482.44) and Withaferin-A (C₂₈H₃₈O ₆, mol. weight 470.60) both were purchased from Sigma-Aldrich (Germany). Both compounds were dissolved in dimethylsulfoxide (DMSO; Sigma-Aldrich) to make a stock solution. The final concentration of DMSO in the culture medium did not exceed 0.5% which has no detectable effects on cells.

CellTiter-Blue cell viability assay

Cells seeded into 96-well plates in sextuplicate at a density of 15 × 10³ cells per well. Twenty-four hours after seeding, the cells were treated with various concentrations of Withaferin-A and Silibinin for totally 24 h. To measure the viability of cells, CellTiter-Blue^® Viability Assay (Promega, G8081) was applied according to the manufacturer’s instructions. Incubation with the dye for 60 min was followed by measurement of the fluorescence with the infinite F200 pro Reader (TECAN). A blank well without cells was measured to determine the background. After subtraction of the background, cell dilution series enabled due to a direct correlation of the signal intensities with the cell number, the verification of absolute cell numbers.

Determination of IC50 from CTB measurements

The half maximal inhibitory concentration (IC50) was determined from series of dose-dependent CTB measurements using following basic steps. First, the raw CTB intensity measurements (I_i) were normalized by the intensity level (I₀) of the reference (untreated) probe: (1) ${\displaystyle n{I}_i=100\frac{I_i}{I_0}.}$Subsequently, the average pattern of normalized dose response (anI_i) from N different measurements was calculated: (2) ${\displaystyle an{I}_i=\frac{1}{N}\underset{j=1..N}{\sum }n{I}_i\left(j\right).}$Finally, the IC50 value of a particular cell line and measurement condition (e.g., duration of treatment) was determined from the average normalized pattern of IC50 dose response by fitting the Hill’s sigmoid function (Spiess & Neumeyer, 2010; Sebaugh, 2011) (3) ${\displaystyle anI\left(D\right)=min+\frac{max-min}{1+{\left(\frac{D}{IC50}\right)}^H},}$where D is the drug dose, min and max are the minimum and maximum values of anI, IC50 and H are the IC50 and the Hill’s coefficient values that are determined from the fit. The nonlinear least square fit of the Hill’s equations was performed automatically using the MATLAB R2019b (The Mathworks, Inc.). To characterize relative differences in viability response (V_i) among five NSCLC cell lines (i = 1..5), normalized IC50 values (nV_i) (4) ${\displaystyle n{V}_i=\frac{V_i-V_{\min }}{V_{\max }-V_{\min }}\in \left[0,1\right]}$and their ranks (R_i) (5) ${\displaystyle R_i=\text{rank}\left(V_i,\left[V_1,V_2,V_3,V_3,V_4,V_5\right]\right).}$were calculated.

Determination of total cell count using quantitative image analysis

Quantitative image analysis was performed to acquire the total cell count. Immediately after the measurement of the fluorescence signal, the medium is aspirated from 96well plate containing NSCLC cells and washed with PBS solution twice to ensure the complete removal of dead cells. Since dead cells do not attach to the surface and freely float in the medium, washing twice with PBS removes approximately all the dead cells from the well. After washing, 70 µl cell fixation solution (PFA 3% + Hoechst dye) was added to each well for the fixation and nuclei staining of cells by a single step. Cells were incubated at 4 °C for 48 h. Wells were washed with PBS twice after incubation to remove unbound Hoechst dye and analyzed under a microscope (Olympus FV 3000) for counting the stained nuclei. Finally, the cell counting was performed manually with the help of images obtained from the microscope.

In vitro cell migration assay

For measurement of cell motility, 2D cell migration assay from (Marwitz et al., 2016) was used. Cells were seeded in 24-well plate (Zell Kontakt 3231-20) at a density of 2,000–5,000 cells/well (depends from cell lines). Twenty-four hours after seeding the cells were treated with various concentrations of Withaferin-A and Silibinin for a total of 24 h and after stained with Hoechst (Sigma) for 1 h. Images were acquired in 20 min intervals for 48 h using environment-controlled microscope (IX81, Olympus) equipped with an UPlanSApo 10 ×/0.4 objective lens (Olympus). Nine positions per well (3 ×3 grid) were imaged and stitched with a ImageJ plugin (Preibisch, Saalfeld & Tomancak, 2009). Single cell tracking was performed with ImageJ Mtrack2 plugin. Speed of each tracked cell was calculated by dividing total travelled distance by total time, for which cell was tracked. Persistence was calculated by dividing the distance between the first and the last point, where the cell was tracked, by total travelled distance. Resulting number was multiplied by the square root of time, for which cell was tracked divided by maximal possible tracking time, in order to penalize cells, which were tracked for a shorter period of time.

Immunofluorescence and microscopic imaging

Cells were fixed for 5 min in 3.7% PFA in PBS at RT, permeabilized with ice-cold 0.1% TritonX-100 in PBS for 5 min, blocked for 30 min with Blocking solution (3% BSA, 2.5% FCS) and stained with primary antibodies to vimentin and actin (Abcam, Cambridge, UK) for 1 h. After staying cells were washed 3 times in PBS and stained with secondary antibodies, Phalloidin-BODIBY (Thermo Fisher) and Hoechst (Sigma) for 30 min. Imaging was performed using the confocal laser scanning microscope (Olympus FV 3000) with 64x oil immersion lens to visualize the actin and vimentin network as well as the nucleus in the cells.

Gene expression data

Gene expression (GE) data of H1650, H1975, A549, H838, H2030 cell lines from GSE47206 (El-Chaar et al., 2014) were used for computational analysis of correlation between transcriptomic and IC50 profiles of NSCLC cells. Analysis was performed only for transcripts with the detection significance of p < 0.05 (P,M) in all five NSCLC cell lines. Trancripts with the non-significant level of detection (A) were excluded from analysis.

GO enrichment analysis

Gene ontology enrichment analysis was performed using STRING v11 (Szklarczyk et al., 2019) with the significance level of p < 0.05 after multiple testing correction (Benyamini & Hochberg, 1995) for each functional classification framework (GO, KEGG, InterPro, etc.).

Transfection of NSCLC cells

All siRNA used for knockdown of BIRC5, FOXM1, BRCA1 was purchased from Ambion^®ThermoFisher Scientific, see Table 3.

Transfection experiments were carried out by lipofection technique using Lipofectamine^®2000 (Invitrogen), with transient expression. Two types of transfection methods were performed in the optimization of transfection efficiency: reverse and forward transfection.

Forward and reverse transfection method

Forward transfection is a conventional method of transfection where cells are first seeded for 24 h and stored at 37 °C and 5% CO₂. After incubation, cells are transfected as per company’s (Invitrogen) protocol. In reverse transfection, unlike conventional transfection, the genetic material is coated on the bottom of well plates prior to the seeding of NSCLC cells into multi-well plate. As the order of adding genetic material is reverse compared to the conventional method it is named as reverse transfection method. Multi-well plates are coated with required siRNA to be transfected along with gelatin, fibronectin and transfection reagent (Lipofectamine^®2000) stored at 4 °C for 28 h and stored in dry place. Later, NSCLC cells were seeded into these wells, incubated for 28 h to attain transfected cells.

Detection of siRNA mediated gene silencing

Estimation of transfection efficiency of the NSCLC cells was performed by blocking PLK1 (polo-like-kinase-1) enzyme using RNA silencing technique via small interfering RNA (siRNA). Expression of PLK1 has several functions like the initiation of mitosis, cytoplasmic separation, mitotic spindle formation and membrane formation in telophase of mitosis by phosphorylating mitotic kinesin-like protein-1. Silencing of PLK1 leads to mitotic cell arrest and the centrosomes lose the ability to nucleate microtubules. This effect can be observed under a microscope with star shape nuclear phenotype. Presence of star shape nucleus indicates that the PLK1 was successfully silenced that in turn confirms that the cells are transfectable.

Immunoblotting

The confirmation of knockdown efficiency was carried out using Western Blot analysis. Harvesting and samples for analysis were prepared by removing supernatant from the cells followed by addition of 500 µl of PBS. Cells were resuspended and transferred to 1.5 ml for centrifugation for 5 min at 1,600 RPM. After centrifugation, PBS was removed without disturbing cell pellet. 25 µl of 1x Laemlli buffer containing DTT (final concentration 100 mM) was added and incubated for 10 min at 95 °C. Proteins in the sample were separated by SDS-PAGE in a 10% resolving gel and transferred onto methanol-activated PVDF membranes by semi-dry blotting technique using transfer buffer at a constant current of 130 V until the sample reaches the resolving gel. As soon as the sample reaches resolving gel electrophoresis is run at 90V. The membrane was blocked with 5% milk powder, 0.5% BSA in PBST for 1 h at room temperature. After blocking, the membrane was incubated with specific primary antibody in 5% (w/v) milk powder/PBST overnight at 4 °C. To remove the unbound primary antibody three subsequent washings for 10 min with PBST were performed. Then the membrane was incubated with the horseradish peroxidase (HRP) coupled secondary antibody in 5% (w/v) milk powder, 0.5% BSA in PBST for 1 h at room temperature. Three subsequent PBST washing were performed to ensure complete removal of the secondary antibody. Detection of proteins was carried out by using ECL chemiluminescent immunodetection system with different exposure time using INTAS Fluoreszenz u. ECL Imager. Subsequently, quantitative analysis of Western Blot images was performed using ImageJ according to a standard protocol (http://www.openwetware.org/wiki/Protein_Quantification_Using_ImageJ). Figure S5 shows exemplary analysis of transient knockdown of target proteins FOXM1 and BIRC5 in A549 and H838 NSCLC cells using small-interfering RNA (siRNA) vs β-actin as a loading control.