RNA interference-mediated depletion of TIA proteins in HeLa cells. (a) Immunoblot analysis of HeLa cell lysates (5 μg (lane 1; C(1/5)) or 25 μg (lanes 2 to 5)) prepared 72 h after transfection with siRNAs against control (C; lanes 1 and 2), TIA-1 (lane 3), TIAR (lane 4), and TIA-1 plus TIAR (lane 5; 1+R). TIA-1(a/b) and TIAR(a/b) refer to the major isoforms of TIA-1 and TIAR, respectively. The blot was probed with antibodies against TIA-1, TIAR, U2AF65 and α-tubulin proteins, as indicated. Molecular weight markers and the identities of protein bands are shown. The intensities of the different protein bands from immunoblots were quantified by densitometry. The represented values were normalized and are expressed relative to α-tubulin (α-Tub), whose value is fixed arbitrarily to 1, and are means ± standard error of the mean (SEM; n = 10; *P < 0.05). (b) Cytoplasmic mRNAs from post-transfected HeLa cells in (a) were analyzed by RT-PCR. Positions of size markers and the predicted alternatively spliced products are indicated. Quantification of relative levels of TIA-1, TIAR and β-actin mRNAs in the above post-tranfected HeLa cells by real time RT-PCR. The represented values were normalized and are expressed relative to β-actin mRNA, whose value is fixed arbitrarily to 1, and are means ± SEM (n = 3; *P < 0.01).

Identification of a gene cluster associated with the silencing of TIA proteins. (a) Venn diagrams depicting the numbers of genes that were up-regulated (left) or down-regulated (right) by TIA-1 (green), TIAR (red) or TIA-1 plus TIAR (blue) silencing. All data for all genes are presented in Additional data files 2 to 5 and 7 and 8. (b) Gene cluster defining a molecular signature of up- (highlighted in red) and down-regulated (highlighted in green) genes in TIA-1 and TIAR-depleted HeLa cells. The microarray data were analyzed by the Rank Products method [29]. Fold is an average measure of the fold change in differential expression and the false discovery rate (FDR) indicates the expected percentage of false positives. The Probe-set ID is the Unigene probe-set identifier [43].

Silencing of TIA proteins alters both mRNA stability and gene transcription. (a) DNA transcription was inhibited in control and TIA-1/TIAR-depleted HeLa cells by the addition of actinomycin D (Act D; 5 μg/ml) to the culture medium. Cytoplasmic RNA was isolated at various times (0, 3 and 6 h) after the addition of the inhibitor Act D and analyzed by real time RT-PCR. Relative levels of β-actin, PTGS2, GDF15, IL-8, TIMP2 and TNFSF10 mRNAs were determined with the specific primer pairs (Additional data file 11). A representative experiment is shown. Steady-state RNA levels are represented at various times after the addition of Act D. Open and closed circles illustrate relative RNA levels in control (C) and TIA-1/TIAR (1+R)-depleted samples, respectively. (b) Validation of differential expression for selected genes by western blot analysis. Protein extracts (5 μg (lane 1; C(1/5)) or 25 μg (lanes 2 to 8)) derived from post-transfected HeLa cells with control (lanes 1 to 5) and TIA-1 and TIAR (lanes 6 to 8) siRNAs and treated with Act D as in (a) were analyzed with the indicated antibodies. A representative blot is shown. Molecular weight markers and the identities of protein bands are indicated. TIA-1(a/b) and TIAR(a/b) refer to the major isoforms of TIA-1 and TIAR, respectively. (c, d) Effect of the TIA-1/TIAR silencing on transcriptional activation of the PTGS2 (c) and IL-6 (d) gene promoters. A schematic representation of PTGS2 (c) and IL-6 (d) human gene promoters is shown. In both cases, cis-acting consensus sequences are represented by boxes. Control and TIA-1/TIAR-depleted HeLa cells were transiently cotransfected with PTGS2 or IL-6 promoter-driven firefly luciferase constructs together with a β-galactosidase-expressing plasmid (used as a transfection control) as described in Materials and methods. Sixteen hours after cotransfection, cells were cultured in the presence of DMSO or 20 ng/ml PMA for 6 h and assayed for luciferase and β-galactosidase activities. The represented values in (c, d) - the ratio between luciferase relative light units (RLU)/β-galactosidase activity - were normalized and are expressed relative to the control sample treated with DMSO, whose value is fixed arbitrarily to 1, and are means ± standard error of the mean (n = 3; *P < 0.001 in (c); *P < 0.01 in (d)).

Interaction network associated with the gene cluster induced upon suppression of TIA proteins. Visualization of biochemical relationships between gene clusters up-regulated (red) and down-regulated (green) identified by microarray analyses were computed using the Pathway Studio (version 6.0) software and the 'ResNet Mammalian' database developed by Ariadne Genomics, Inc. [40]. The network topology illustrates positive (red boxes) and negative (green boxes)-regulated cell processes based on interactions after inactivation of TIA proteins. The nature of the regulators is depicted schematically as shown in the legend to the right. Continuous and dotted lines illustrate direct and indirect protein-protein interactions, respectively. The thickness of blue lines is proportional to the number of interactions reported in the literature.

Knockdown of TIA proteins activates an acute inflammation-promoting transcriptome. (a) Heat map representation depicting microarray data for genes that are differentially regulated by siRNA-mediated reduction of TIA-1 and TIAR. Mean fold-change values from three independent replicates for each biological condition tested are given for 538 probe sets (horizontal lanes; out of a total of 47,400 sets) that were detectable above background showing at least a 1.5-fold change in expression and were statistically different (P < 0.05) in the two experimental conditions tested. Red and green indicate up- and down-regulation, respectively, relative to control siRNA-transfected HeLa cells. The color scale to the right indicates the magnitude of the fold change (base 2 logarithm) for a particular transcript. (b) Total number of genes whose expression was up- or down-regulated in TIA-1 and TIAR-depleted cells. (c, d) Graphic pie representations of the distribution of up- (c) and down-regulated (d) genes (P < 0.01) using the GO biological process category. (e, f) Graphic pie representations of the distribution of up- (e) and down-regulated (f) genes (P < 0.05) using the KEGG pathway database. In all cases (c-f), percentages shown reflect the portion of total genes that are associated with the biological functions and pathways indicated.

Validation of microarray-predicted changes by quantitative RT-PCR (qRT-PCR). (a) Differential expression of 26 genes, 10 up-regulated (bottom bars), 12 down-regulated (top bars) and 4 unchanged (middle bars), was verified by real time RT-PCR using specific primer pairs (Additional data file 11). The represented values were normalized and are expressed relative to β-actin, whose value is fixed arbitrarily to 1, and are means ± standard error of the mean (SEM; n = 3; *P < 0.001; **P < 0.01; ***P < 0.05). (b) Differential expression of eight genes, five up-regulated (IL-6, IL-8, AREG, PTGS2 and EREG), two down-regulated (TNFSF10 and CD24) and one unaffected (β-actin), from cytoplasmic RNA isolated from HeLa cells transfected with either control (white bars), TIA-1 (grey bars), TIAR (dark grey bars) or TIA-1 plus TIAR (black bars) siRNAs was verified as above. The represented values were normalized and are expressed relative to β-actin, whose value is fixed arbitrarily to 1, and are means ± SEM (n = 3; *P < 0.001; **P < 0.01).

Depletion of TIA proteins activates cell proliferation and anchorage-independent growth. (a) Post-transfected HeLa cells with siRNAs against control (C; white diamond), TIA-1 (1; grey square), TIAR (R; grey circle) or TIA-1 plus TIAR (1+R; black circle) were seeded in six-well plates and the number was counted on the days indicated for 6 days. Each time point represents the means ± standard error of the mean (SEM; n = 3 to 6; *P < 0.01). (b) Cells grown for 6 days as in (a) were monitored by methyl thiazolyl tetrazolium (MTT) assays. The represented values were normalized and are expressed relative to control (C), whose value is fixed arbitrarily to 1, and are means ± SEM (n = 3 to 5; *P < 0.001). (c) Analysis of cell-cycle phases by flow cytometry after propidium iodide staining. The data are means ± SEM (n = 3; *P < 0.01; **P < 0.05). (d) Representative photographs of colonies generated from transfected HeLa cells. One week after seeding 2,000 cells in a 6-well plate, the cells were fixed in paraformaldehyde (5%), stained with crystal violet (0.01%) and quantified by light microscopy. The represented values are the means ± SEM from three independent experiments performed in duplicate for at least eleven independent colonies (*P < 0.001). A colony was defined as a cell cluster when containing at least 86 cells. Scale bars represent 200 μm. (e) Soft-agar colony formation. In each 6-well plate, 5,000 post-transfected HeLa cells as above were seeded into soft-agar matrix and incubated for 14 days. The cell number was counted under a light microscope after staining with crystal violet as before. The represented values are means ± SEM for three independent experiments performed in duplicate for at least 36 counted fields (*P < 0.001; **P < 0.01). In this case, a colony was defined as a cell cluster containing at least 4 cells. Scale bars represent 100 μm.