Advertisement

Constitutively activated STAT3 promotes cell proliferation and survival in the activated B-cell subtype of diffuse large B-cell lymphomas

B. Belinda Ding, J. Jessica Yu, Raymond Y.-L. Yu, Lourdes M. Mendez, Rita Shaknovich, Yonghui Zhang, Giorgio Cattoretti and B. Hilda Ye

Data supplements

Article Figures & Data

  • Figure 1

    High-level STAT3 expression and activation are preferentially associated with the BCL6-low ABC-DLBCL. (A) IL-6 production, expression, and activation of STAT3 are inversely correlated with BCL6 in DLBCL cell lines. Protein expression of BCL6, MTA3, total and phosphorylated forms of STAT1, STAT3, STAT5, and STAT6 was analyzed by Western blot. GAPDH levels were used as loading control. The COO status of the cell lines are labeled on the top. The titers of secreted IL-6 as measured by ELISA are given at the bottom of the panels. (B,C) Representative IHC staining for BCL6 and CD20/PY-STAT3 in serial sections. (B) Staining of a GCB case that is positive for BCL6 but negative for PY-STAT3. (C) Staining of a non-GCB case that has very few BCL6-positive malignant cells but is strongly positive for PY-STAT3. Note strongly positive endothelial cell nuclei as endogenous positive controls for PY-STAT3 staining. Scale bar = 1 μm (40×). (D) Distribution of BCL6 and PY-STAT3 staining in GCB and non-GCB subgroups. The staining was scored as described in the Document S1. The association between BCL6 expression and DLBCL subgroups or PY-STAT3 and the subgroups was tested by chi-square analysis.

  • Figure 2

    STAT3 mRNA is highly expressed in ABC-DLBCL compared with GCB-DLBCL. (A) Northern blot analysis of BCL6 and STAT3 in DLBCL cell lines. The 28S indicates 28S ribosomal RNA used as loading control. Vertical lines have been inserted to indicate a repositioned gel lane. (B) BCL6, STAT3, and cyclin D2 expression patterns based on previously published DLBCL gene expression data.3 The signal values for BCL6 (probe 24429), STAT3 (probe 31469), and cyclin D2 (probe 16858) were retrieved for 72 ABC-DLBCL and 114 GCB-DLBCL samples. The data had been previously log2 transformed and median centered for each gene across the entire sample set. The P values are based on Wilcoxon-Mann-Whitney nonparametric test performed to compare the values between the 2 subgroups. Bars indicate the group means for each column. Mean, group mean of the log2-transformed values; Fold change, linear fold difference between the group means.

  • Figure 3

    Expression of STAT3 and BCL6 are mutually exclusive in normal GC cells. IHC and IF staining of normal tonsil GC for STAT3 and lineage- and differentiation-associated markers. The color for each marker is depicted by the label beside each panel. (A) A low-power image of BCL6 (red) and STAT3 (green) staining; note the STAT3+ cell localizing to the apical light zone. The high-power insets on the right are from an enlarged apical light zone area showing a rare cell (arrow) with nuclear BCL6 and cytoplasmic STAT3, whereas the other STAT3+ cells have no BCL6. (B) Triple staining for BCL6, PRDM1/Blimp-1, and PY-STAT3 shows mutually exclusive nuclear PY-STAT3 and BCL6, as well as PY-STAT3 and PRDM1. Rare double STAT3+PRDM1+ cells are noted (arrow). The rectangle is selected, and the image is split into a single-color panel below. (C) Double staining for STAT3 and PY-STAT3 shows that these 2 signals are always colocalized (shades of yellow color). (D) Double staining for PY-STAT3 and CD3, a pan-T marker, shows that there is no overlap between these 2 markers. (E) Double staining shows that all STAT3+cells are positive for CD20, a pan-B marker. (F) Double staining for PY-STAT3 and CD138/Syndecan, a late-stage plasma cell marker, shows that these 2 markers are mutually exclusive. Scale bar, 4 μm in the low-power image of panel A, 1 μm in other panels.

  • Figure 4

    BCL6 inhibits STAT3 transcription via 2 upstream binding sites. (A) Schematic representation of the human STAT3 promoter region showing 5 potential BCL6 binding sites, location of ChIP primers, and the 3-kb fragment contained in the wt STAT3 luciferase reporter construct. (B) EMSA analysis of BCL6 binding to the 5 candidate BCL6 sites. The sequences for these 5 sites are shown together with the 10-bp core of the consensus BCL6-binding site FB20. (+) and (−) indicate the top or bottom strand of the STAT3 promoter region. Underlines indicate sequences identical to those in the consensus. In the top panel, the 20-bp FB20 probe was 32P labeled and used in the assay with nuclear extracts of Ly1 cells. In the bottom panel, all 5 candidate BCL6 sites and FB20 were labeled and subjected to EMSA analysis with or without the use of excess, cold FB20 probe in competition. (C) BCL6 represses transcription from the 3-kb wt STAT3 reporter in BCL6-negative Mutu III cells. Based on site-directed mutagenesis, repression by BCL6 was attributed to both sites B and D. The activity of each reporter in the absence of BCL6 was normalized to 100. Error bars represent SD.

  • Figure 5

    BCL6 inhibits expression of the endogenous STAT3 gene in DLBCL. (A) BCL6 and MTA3 bind to the STAT3 promoter region in vivo. ChIP assays were performed in Ly1 cells. PCR products were amplified from chromatin fragments immunoprecipitated with either anti-BCL6, anti-MTA3, or control rabbit IgG antibodies. The PCR primers flank either site B or the negative control +20-kb site (see arrows in Figure 4A). H2O, negative control PCR without any template; input, purified total genomic DNA before precipitation. (B) Forced expression of BCL6 suppresses endogenous STAT3. Ly10 cells were transiently transfected with the indicated expression plasmids. Western blot was performed to analyze expression of BCL6 and MTA3 12 hours after transfection, and total STAT3 24 hours after transfection. Similar results were obtained from Ly3 cells (not shown). (C) RNAi-mediated BCL6 knockdown increased STAT3 protein expression. Control (ctrl) and 2 different BCL6 siRNA oligos (nos. 1 and 2) were transiently transfected into Val cells, and cell lysates were prepared at either 24 or 48 hours for protein analysis. In lanes 7 to 10, cells that have been transfected with oligos for 45 hours were either left alone (lanes 7 and 8) or exposed to PMA (20 ng/mL) and ionomycin (0.3 μM) (P + I, lanes 9 and 10) for another 24 hours before harvesting. In lane 11, Ly3 lysates were loaded for comparison. BCL6, total STAT3, and PY-STAT3 proteins were examined by Western blot analysis. GAPDH levels were shown as loading control. Vertical lines have been inserted to indicate a repositioned gel lane.

  • Figure 6

    AG490 treatment reduced STAT3 activation, triggered cell-cycle arrest and apoptosis in ABC-DLBCL cells. (A) Western blot analysis showed that AG490 treatment inhibited STAT3 activation but not Erk1/2 in Ly10 cells. Cells were treated with the indicated drug concentrations and sampled at the indicated time points. Similar results were obtained from Ly3 cells (not shown). (B) MTT assays indicated that AG490 selectively reduced cell growth of ABC-type DLBCL cells (Ly3 and Ly10) without affecting GCB-DLBCL cells (Ly1 and Ly7). (C) Cell-cycle profiles of AG490-treated Ly10 cells were analyzed by PI staining followed by flow cytometry. Cells were treated with the indicated drug concentrations and sampled at the indicated time points. Result shown is representative of 2 independent experiments. (D) Annexin V and 7-AAD staining was used to monitor ongoing apoptosis in Ly1, SUDHL6, Ly3, and Ly10 cells treated for 24 hours with 10 or 20 μM AG490. The percentage of annexin V+7-AAD cells is plotted in the graph.

  • Figure 7

    RNAi-mediated STAT3 knockdown in ABC-DLBCL cells also triggered cell-cycle arrest and apoptosis. (A) STAT3 siRNA (STAT3i, +) and control (−) oligos were transiently transfected into Ly3 and Ly10 cells. At the indicated time points after transfection, total STAT3 and PY-STAT3 expression was examined by Western blot analysis. (B) Proliferation of STAT3i-transfected Ly3 and Ly10 cells was measured by 3H-thymdine incorporation assays. The amount of radioactivity recovered from STAT3i-transfected cells was normalized to that from the control cells defined as 100%. Represented in the graph are the mean plus or minus standard derivation of 3 independent STAT3 siRNA experiments. (C) Western blot showing increased PARP cleavage (▷) and, thus, activation of caspase 3/7 in STAT3i-transfected cells. (D) Western blot analysis demonstrating alterations in selected cell cycle and survival regulators following STAT3 knockdown. Vertical lines have been inserted to indicate a repositioned gel lane.

Supplementary Materials

  • Figure S1

    Supplementary PDF file available online.

  • Table S1

    Supplementary PDF file available online.

  • Table S2

    Supplementary PDF file available online.

  • Table S3

    Supplementary PDF file available online.

  • Document S1

    Supplementary PDF file available online.