Additional file 1 of Targeting FGFR4 abrogates HNF1A-driven metastasis in pancreatic ductal adenocarcinoma

Supplementary Material 1: Supp Fig. 1. A) Quantification of photon flux from bioluminescent images from all harvested livers and lungs inoculated with LacZ or HNF1A overexpressing cells at endpoint (n = 10 mice per group). B) Quantification of photon flux from bioluminescent images from all harvested livers and lungs with control or HNF1A knockdown at endpoint (n = 10 mice per group). C) Percent of blood cells positive for GFP (tumor cells) collected from the cardiac blood. All bar graphs represent the mean and statistical difference was determined by unpaired t-test. Supp Fig. 2. A) Western blotting for HNF1A with and without doxycycline administration before cell inoculation. B) Immunohistochemistry for HNF1A in the liver metastases of mice from each respective arm. C) Western blotting for HNF1A in AsPC-1 cells before inoculation. D) Immunohistochemistry for HNF1A in the liver metastases of mice from each respective arm. E) Representative photographs of harvested livers from both LacZ and HNF1A groups showing increased tumor burden (white tissue) in the HNF1A livers. F) Measured liver weights of all livers after harvest at endpoint (n = 13 mice per group). G) Quantification of the number of lung metastases per H&E stained tissue section. H) Quantification of cells positive for Ki67 immunohistochemistry staining in the liver metastases from 3 representative mice from each group. I) Western blotting for HNF1A in UM53 cells before inoculation. J) Immunohistochemistry for HNF1A in the liver metastases of mice from each respective arm. K) Quantification of cells positive for Ki67 immunohistochemistry staining in the liver metastases from 3 representative mice from each group. All bar graphs represent the mean and statistical difference was determined by unpaired t-test. Supp Fig. 3. A) Correlation of HNF1A and FGFR4 RNA expression in patient liver metastatic tumors from the European Genome Phenome Archive dataset. B) ChIP-sequencing of HNF1A and H3K27ac reads from HepG2 (top) and HNF1A reads from UM5 (bottom) cells. C) Quantitative RT-PCR analysis of FGFR4 RNA levels with overexpression of wild-type (WT) HNF1A and HNF1A with a mutant and inactive transactivation domain (P291) (n = 3 biological replicates). All bar graphs represent the mean ± SEM and statistical difference was determined by two-way ANOVA with Tukey’s multiple comparisons test or one-way ANOVA with Tukey’s multiple comparisons test. Supp Fig. 4. A) Cell viability as normalized to control knockdown 72 h after siRNA-mediated knockdown of FGFR4. B) Cell viability as normalized to control knockdown for colony formation experiments. Cells were transfected with siRNA for 72 h before plating at a low cell density (200 cells/well) and grown for 2 weeks. Resultant colonies were quantified by AlamarBlue viability assay. C) Number of tumorspheres formed after 14 days with siRNA-mediated knockdown of FGFR4. D) Western blotting for HNF1A, FGFR4, E-cadherin, Zeb1, Vimentin, Slug, and Actin 72 h after siRNA-mediated knockdown of FGFR4. E) Cell viability as normalized to LacZ control overexpression 72 h after plating. F) Cell viability as normalized to LacZ control overexpression for colony formation experiments. Cells were plated at a low cell density (200 cells/well) and grown for 2 weeks. Resultant colonies were quantified by AlamarBlue viability assay. G) Number of tumorspheres formed after 14 days with overexpression of FGFR4. H) Western blotting for HNF1A, FGFR4, E-cadherin, Zeb1, Vimentin, Slug, and Actin with overexpression of FGFR4. All bar graphs represent the mean and statistical difference was determined by unpaired t-test. Supp Fig. 5. A) Western blotting for HNF1A and total FGFR4 in AsPC-1, UM5, and UM53 cell lines treated with either DMSO or H3B-6527. B) Western blotting for HNF1A and total FGFR4 in AsPC-1, UM5, and UM53 cell lines treated with either Vehicle or U3-1784. Supp Fig. 6. A) Cell viability of LacZ and HNF1A overexpressing AsPC-1, UM5, and UM53 cells treated with DMSO or H3B-6527 as normalized to LacZ control cells treated with DMSO after 72-h treatment. B) Cell viability for colony formation experiments as normalized to LacZ cells treated with DMSO. Cells were treated with wither DMSO or H3B-6527 for 72 h before plating at a low cell density (200 cells/well) and grown for 2 weeks. Resultant colonies were quantified by AlamarBlue viability assay. C) Number of tumorspheres formed after 14 days with continuous treatment with 1 μM H3B-6527. D) Cell viability of LacZ and HNF1A overexpressing AsPC-1, UM5, and UM53 cells treated with Vehicle or U3-1784 as normalized to LacZ control cells treated with DMSO after 72-h treatment. E) Cell viability for colony formation experiments as normalized to LacZ cells treated with DMSO. Cells were treated with wither DMSO or H3B-6527 for 72 h before plating at a low cell density (200 cells/well) and grown for 2 weeks. Resultant colonies were quantified by AlamarBlue viability assay. F) Number of tumorspheres formed after 14 days with continuous treatment with 5 μg/mL U3-1784. G) Western blotting for HNF1A, FGFR4, E-cadherin, Zeb1, Vimentin, Slug, and Actin 72 h after treatment with H3B-6527 or U3-1784. All bar graphs represent the mean and statistical difference was determined by two-way ANOVA with Tukey’s multiple comparisons test. Supp Fig. 7. A) Normalized quantification and representative images of cell migration in both LacZ and HNF1A overexpressing sublines of UM5 and UM53 cells treated with 1 μM H3B-6527(n = 3 biological replicates). DMSO or H3B-6527 treated cells were plated in transwell migration chambers in serum free media containing DMSO or H3B-6527 and cells that had migrated after 24 h were fixed and stained for counting. B) Normalized quantification and representative images of cell invasion in both LacZ and HNF1A overexpressing sublines of UM5 and UM53 cells treated with 1 μM H3B-6527 (n = 3 biological replicates). DMSO or H3B-6527 treated cells were plated in transwell invasion chambers in serum free media containing DMSO or H3B-6527 and cells that had migrated after 48 h were fixed and stained for counting. C) Cell invasion as normalized to DMSO treated controls for UM53 spheres seeded in a collagen/Matrigel matrix at 0, 24, and 48 h treated with either H3B-6527 or U3-1784. D) Immunohistochemistry for HNF1A (left) and FGFR4 (right) in the liver metastases of mice from each respective arm. E) Quantification of cells positive for Ki67 immunohistochemistry staining in the liver metastases from 3 representative mice from each group. All bar graphs represent the mean ± SEM and statistical difference was determined by two-way ANOVA with Tukey’s multiple comparisons test.