Cells
The human CRC cell lines RKO, SW480, HCT116 and LOVO were provided by Stem Cell Bank, Chinese Academy of Sciences (Shanghai, China). Cells were cultured in minimum Eagle’s medium (MEM, Gibco, USA), Mccoy’s 5A medium (Gibco, USA) and Roswell Park Memorial Institute (RPMI) 1640 medium (Gibco, USA) with 10% fetal bovine serum (Gibco, USA). Cells were grown in an incubator at 37˚C with 5% CO2.
Immunofluorescence
Localization of endogenous transgelin in RKO, SW480, HCT116 and LOVO cell lines was determined by immunofluorescence. The primary antibody (anti-transgelin, 1:500, Abcam, USA), secondary antibody (Alexa Flour 594 goat anti-rabbit IgG, 1:500, Invitrogen, USA), and the VECTASHIELD mounting medium (Vector Laboratories, USA)) with 4′,6-diamidino-2-phenylindole (DAPI) were used. The immunofluorescence images were taken and preserved under the laser scanning confocal microscope (63× oil lens, Carl Zeiss, USA).
Transfection
SW480 and RKO cells were cultured in 12-well plates and transfected with pcDNA6/myc-His B-TAGLN-flag plasmid and pcDNA6/myc-His B-flag plasmid (Takara, Japan). In the validation experiment, we transfected the RKO cells with pENTER-TAGLN-Flag and pENTER-Flag control plasmid (Vigene Biosciences, USA). Transfection was conducted using Lipofectamine 2000/ Lipofectamine 3000 (Thermo Fisher Scientific, USA). The cells were harvested at 48 hours after transfection for further analysis.
Immunoblotting
Nuclear and plasma proteins from HCT116, SW480, LOVO and RKO cell lines were extracted using NE-PER Nuclear and Cytoplasmic Extraction Reagents (Thermo Fisher Scientific, USA). The cytoplasmic protein and nuclear protein extracted above were determined for protein concentration. Immunoblotting was carried out with the primary antibody anti-transgelin (1:500, Abcam, USA, or 1:500, R&D, USA), anti-GADPH (1:400, Abcam, USA or 1:500, Cell signaling technology, USA), anti-PARP1 (1:500, Cell signaling technology, USA), anti-Lamin B1(1:1000, Cell signaling technology, USA), anti-flag (1:500, Cell signaling technology, USA) and the secondary antibody (Horseradish Peroxidase (HRP)-conjugated goat anti-rabbit or anti-mouse IgG, 1:30000, Sigma-Aldrich, USA) or IgG Detector (IgG Detector Solution v2, HRP labeled, 1:1000, Takara, Japan). Antibody detection was performed using a chemiluminescence substrate and the protein bands were visualized with Syngene G:BOX Chemi XT4 fluorescence and chemiluminescence gel imaging system (Cambridge, UK).
Immunoprecipitation
RKO and SW480 cells were cultured conventionally and transfected with pcDNA6/ myc-His B-TAGLN-flag and pcDNA6/ myc-His B-flag plasmids. In the validation experiment, RKO cells were transfected with pENTER-TAGLN-Flag and pENTER-Flag control plasmids. After 48 hours, the culture medium was removed. According to the protocol of the Pierce Crosslink Immunoprecipitation Kit (Thermo Fisher Scientific, USA), antibody immobilization, cell lysis, pretreatment of cell lysate with control agarose resin, immunoprecipitation, immunoprecipitation elution, and immunoblotting analysis were performed in sequence. Anti-flag antibody (10ug, Sigma-Aldrich, USA for the subsequent mass spectrometry; 1:50, Cell signaling technology, USA for the validation experiment) and the control rabbit IgG (1:50, Cell signaling technology, USA) were used.
Mass spectrometry
A fraction of the protein samples after immunoprecipitation were handled by SDS-PAGE and silver staining. Another fraction of the samples was loaded for high performance liquid chromatography (EASY-nLC™, Thermo Fisher Scientific, USA) after filtered aided proteome preparation (FASP) and enzymatic hydrolysis. The samples were then analyzed by Q-Exactive Mass Spectrometer (Thermo Finnigan, USA). The mass charge/ratio of peptides and fragments of peptides were collected. Maxquant 1.3.0.5 software was used to retrieve the Uniprot database by using the raw file as source. The search in the database was set up with specific parameters (Enzyme, trpsin; De-Isotopic, True; Max Missed Cleavages, 2; Fixed modifications, Carbamidomethyl (C); Variable modifications, Oxidation (M); First search ppm, 20ppm; Main search ppm, 6ppm; Decoy database pattern, reverse; Min. Reporter PIF, 0.75; Peptides false discovery rate (FDR) ≤0.01; Protein FDR≤0.01).
Bioinformatics
Identification of differential expression genes (DEGs), functional enrichment and signaling pathway enrichment analysis
According to our previous work[9], the relevant cDNA microarray data was obtained using Affymetrix microarray technique. Over-expression of TAGLN in RKO human colon cancer cells led to a total of 256 downstream transcripts that were differentially expressed with at least a 2-fold change (P<0.05). Among these, transcripts without gene symbols, gene database codes and duplicates were excluded. The remaining DEGs were screened for further bioinformatics analysis.
Using the Metascape tool (www.metascape.org/), the screening parameters were set as the following: P < 0.01 or 0.001 (Biological Process), participating genes ≥3, and enrichment factor > 1.5. We conducted functional and signaling pathway enrichment analysis of the DEGs referring to the gene ontology (GO) database, Kyoto Encyclopedia of Genes and Genomes (KEGG) Pathway and Reactome Gene Sets databases.
Construction of the protein-protein interaction (PPI) network, topological analysis and key gene screening
The DEGs were simultaneously translated into proteins while STRING 10.0 (https://string-db.org/) [11] was used for PPI analysis. Subsequently, relevant data was imported into Cytoscape online software (www.cytoscape.org/) [12] and a PPI network was constructed. In this study, CytoHubba plug-ins were used to calculate the degree centrality and intermediate centrality of the DEGs. Those with values that are 2-fold higher than the overall average value were selected as the core genes in the network. In addition, we obtained the core modules by using an MCODE plug-in (k-core=2), and defined the core genes and the genes included in the core modules as key genes. Key genes were further analyzed with Metascape for signaling pathway enrichment in KEGG Pathways and Reactome Gene Sets database using the same parameters mentioned above.
Prediction of the transcription factors for the key genes
The transcription factor (TF) evaluation model within the GCBI tools (https://www.gcbi.com.cn/) was used to predict the TFs for the key genes. Those with medium or high recommendation were selected, and the potential TFs were selected for further analysis. We then compared these potential TFs to the DNA-binding proteins identified in the mass spectrometry analysis.
Nuclear localization signal analysis
The sequences of selected potential TF(s) were obtained from Uniprot database (https://www.uniprot.org/)[13]. The cNLS Mapper (www.nls-mapper.iab.keio.ac.jp/) [14] was used to detect the nuclear localization signal of the potential TF(s).
Statistics
The statistical analysis was carried out by SPSS 20.0 software. The relevant values were expressed as mean ± standard deviation, and the significance of the differences between two groups was determined by Student’s t test. P<0.05 (bilateral) was considered to be statistically significant.