Figure 1.
GATA6 Copy Number Gain Correlates with Intraductal Progression of Pancreatic Cancer.
(A) GATA6 copy number (mean ± SE) in microdissected normal ducts (N = 4), PanIN-1 (N = 13), PanIN-2 (10), PanIN-3 (N = 17) lesions and pancreatic cancer (N = 55). (B) Representative FISH of the nucleus of a neoplastic cell within a PanIN3 lesion with >11-fold GATA6 amplification (right) compared to the nucleus of a neoplastic cell from a different PanIN3 lesion without copy number gain of GATA6 (left). GATA6 probe was labeled with red and chromosome 18 centromere probe (18 Cent) was labeled with green. The sections were counterstained with DAPI to highlight nuclei. (C) Correlation of GATA6 mRNA expression and copy number in microdissected samples of normal, PanIN and cancer tissue. (D) GATA6 immunolabeling of two pancreatic cancer tissues with GATA6 copy number gain compared to two cancers without copy number gain. Increased copy number is highly associated with nuclear labeling of GATA6 protein. (E) Kaplan Meier survival curve illustrating the relationship of GATA6 copy number gain (≥2.3 copies per haploid genome) to overall survival in patients with surgically resected pancreatic cancer.
Table 1.
Genetic Features of Matched Pancreatic Intraepithelial Neoplasia and Cancer Samples.
Figure 2.
Effects of GATA6 Knockdown on Cell Growth in vitro and in vivo.
(A) Total protein was extracted from AsPC1-GATA6sh and A13A-GATA6sh cells and mock shRNA controls and analyzed by Western blot for relative levels of GATA6 protein relative to actin. (B) Real-time PCR for GATA6 expression in AsPC1-GATA6sh and A13A-GATA6sh cells. (C) These cells were also analyzed for cell proliferation at different time points, (D) cultured in soft agar and the number of colonies at 2 weeks counted, and (E) analyzed by flow cytometry to determine the percent of cells in G2/M phase. (F) Representative xenograft formation in vivo (above) and after explantation (lower) of AsPC1 control and GATA6sh cells at 8 weeks postinjection. (G) Average tumor volume (mean ± SE) of these same xenografts at 8 weeks postinjection. Similar results were noted for A13A-GATA6sh cells (data not shown). With exception of flow cytometry that was performed in duplicate, all experimental data shown represents the summary three independent experiments. *, p<0.05; **, p<0.01; ***, p<0.001.
Figure 3.
Effects of GATA6 Overexpression on Cell Growth in vitro and in vivo.
(A) Real-time PCR for GATA6 expression in Panc1-mock, Panc1-GATA6 and Panc1-mGATA6 cells. (B) Panc1-mock, Panc1-GATA6 and Panc1-mGATA6 cells were either analyzed for cell proliferation or (C) cultured in soft agar and the number of colonies at 2 weeks counted. (D) Representative xenograft formation in vivo (above) and after explantation (lower) of Panc1-mock, Panc1-GATA6 and Panc1-mGATA6 cells at 8 weeks postinjection. (E) Average tumor volume (mean ± SE) of these same xenografts at 8 weeks postinjection. All experimental data shown represents the summary three independent experiments. *, p<0.05; **, p<0.01; ***, p<0.001.
Figure 4.
GATA6 Overexpression Correlates with Canonical Wnt Signaling.
(A) Wnt signaling activity in AsPC1-GATA6sh and A13A-GATA6sh cells based on TOPFLASH assay. Luciferase activity is represented as the ratio of OT to OF levels in cells with GATA6 knockdown relative to that of mock-transfected cells. (B) Wnt signaling activity in Panc1-GATA6 and HPNE-GATA6 cells determined by TOPFLASH assay. Luciferase activity is represented as the ratio of OT to OF levels in GATA6 transfected cells relative to that of mock-transfected cells. (C and D) Panc1-GATA6 cells were transiently transfected with ß-catenin or mock shRNA and (C) cell proliferation or (D) colony formation determined. All experimental data shown represents the summary three independent experiments. *, p<0.05; **, p<0.01. (E) Immunolabeling patterns of GATA6 and ß-catenin protein in two representative PDAC tissues. Arrows indicate nuclear labeling of both GATA6 and ß-catenin in serial sections of the same cancer tissue. By contrast, the PDAC sample with low GATA6 expression also shows no expression of ß-catenin.
Figure 5.
(A) Venn diagram indicating the number of dysregulated genes identified by microarray analysis of AsPC1-GATA6sh (left circle, blue) and A13A-GATA6sh cells (right circle, yellow). The green cross-area indicates commonly dysregulated genes and includes DKK1. (B) Real-time PCR confirming DKK1 overexpression in AsPC1-GATA6sh and A13A-GATA6sh cells. (C) Detection of secreted DKK1 protein in conditioned media in AsPC1-GATA6sh and A13A-GATA6sh cells. Secreted DKK1 protein levels are indicated by using absorbance OD450. (D) Chromatin immunoprecipitation assay confirming binding of GATA6 to the DKK1 promoter. Non-immune IgG and whole genome derived gDNA are used as negative and positive controls, respectively. (E) EMSA assay confirming binding of GATA6 to putative GATA binding sites #2 and #3. The mutant sequence mGATA-#3 did not generate any detectable binding. Nuclear Extr, nuclear extract; GATA6-P, a positive control probe derived from the TFF2 promoter containing a GATA6 binding site; GATA-#2, probe containing putative GATA binding site No. 2; GATA-#3, probe containing putative GATA binding site No. 3; mGATA-#3, probe containing a mutated putative GATA binding site No. 3; refer to methods for additional details (F) Effect of GATA6 expression on activity of the DKK1 promoter. Data is presented as the ratio of firefly luciferase activity to sea pansy luciferase activity. pGL3 was used as a negative control for background. (G) Real-time PCR for DKK1 mRNA expression in Panc1 cells. When appropriate, all experimental data shown represents the summary three independent experiments. *, p<0.05; ***, p<0.001.
Figure 6.
DKK1 Mediates GATA6 Effects on Wnt Signaling.
(A) Representative DKK1 and ß-catenin immunolabeling in two human pancreatic cancer tissues. Arrows in the bottom left panel indicate location of cancer cells with negative labeling for ß-catenin (all images x400) (B) Panc1, A10.7 and Hs766t cells were transfected with a shRNA against DKK1 and subjected to TOPFLASH assay. Wnt activities are presented as a ratio of OT activity to OF activity. (C and D) Panc 4.14 cells were transfected with a DKK1-expressing vector and subjected to (C) TOPFLASH assay or (D) cultured in soft agar and the number of colonies at 2 weeks counted. Panc1 cells were transfected with mock, GATA6, and/or DKK1 expressing vectors, and cells were analyzed for (E) cell proliferation, or (F) cultured in soft agar and number of colonies at two weeks counted. When appropriate, all experimental data shown represents the summary three independent experiments. *, p<0.05.