SEARCH

SEARCH BY CITATION

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
sc-12-0067_sm_SupplFigure1.pdf20KSupplemental Figure S1. Expression ectopic RUNX3 and RUNX3R178Q in GIF-14. Western blotting analysis of RUNX3 proteins in parental GIF-14 cells infected with A) iG2- and B) pBOBI- lentiviruses. Immunoblotting of α-tubulin was performed to control for the amounts of protein loaded.
sc-12-0067_sm_SupplFigure2.tif1825KSupplemental Figure S2. P1 and P2 subpopulations have comparable proliferation rates. FACS-fractionated P1 and P2 cells were seeded in 12-well plates at 25,000 cells/well and cell count was performed at indicated days to monitor cell growth. The number of viable cells that adhered 15h post-seeding was counted as cell number at Day 0 (means ± SEM, n=3).
sc-12-0067_sm_SupplFigure3.tif2676KSupplemental Figure S3. P1 and P2 subpopulations show differential expression of Abcb1/Mdr and Abcg2. P1 and P2 subpopulations were fractionated from parental GIF-14 cells based on Hoechst 33342 staining and expanded in culture. Expression levels of three members of the ATP-binding cassette transporters were ascertained by qRT-PCR and normalized data are expressed relative to those of P1 (means ± SEM, n=4).
sc-12-0067_sm_SupplFigure4.pdf1170KSupplemental Figure S4. Formation and long-term culture of GIF-14 spheres, related to Figure 2. A composite phase contrast image of spheres generated from GIF-14 cells shows that they could be maintained in a serum-free medium for an extended period of at least 40 days.
sc-12-0067_sm_SupplFigure5.tif968KSupplemental Figure S5. Increased sphere formation in GIF-14 cells in the presence of Wnt3a-conditioned medium, related to Figure 2. GIF-14 cells were cultured in a standard serum-free medium supplemented with the indicated dilutions of control- or Wnt3a-Cm and 0.5% methyl cellulose to reduce the tendency to form clumps.
sc-12-0067_sm_SupplFigure6A.tif951KSupplemental Figure S6. P2 cells display increased migratory and invasive properties in transwell migration assay. (A) FACS-fractionated P1 and P2 cells were seeded in Transwell inserts at 50,000 cells/insert in serum-free growth media and allowed to migrate for 24h in response to normal growth media with 10% fetal bovine serum. Phase contrast images were captured from 8 different fields for each replicate and the number of migrated cells from three replicates was scored (means ± SEM, n=3). Scale bars, 50μm. (B) The invasive properties of P1 and P2 cells were measured using Transwell inserts coated with matrigel. The cells were stained with 4′,6- diamidino-2-phenylindole (DAPI) after 48h and representative images were shown. Scale bars, 50μm. The data presented are a compilation of the number of invaded and migrated cells from three replicates (means ± SEM, n=3).
sc-12-0067_sm_SupplFigure6B.pdf23KSupplemental Figure S6. P2 cells display increased migratory and invasive properties in transwell migration assay. (A) FACS-fractionated P1 and P2 cells were seeded in Transwell inserts at 50,000 cells/insert in serum-free growth media and allowed to migrate for 24h in response to normal growth media with 10% fetal bovine serum. Phase contrast images were captured from 8 different fields for each replicate and the number of migrated cells from three replicates was scored (means ± SEM, n=3). Scale bars, 50μm. (B) The invasive properties of P1 and P2 cells were measured using Transwell inserts coated with matrigel. The cells were stained with 4′,6- diamidino-2-phenylindole (DAPI) after 48h and representative images were shown. Scale bars, 50μm. The data presented are a compilation of the number of invaded and migrated cells from three replicates (means ± SEM, n=3).
sc-12-0067_sm_SupplFigure6B.tif1161KSupplemental Figure S6. P2 cells display increased migratory and invasive properties in transwell migration assay. (A) FACS-fractionated P1 and P2 cells were seeded in Transwell inserts at 50,000 cells/insert in serum-free growth media and allowed to migrate for 24h in response to normal growth media with 10% fetal bovine serum. Phase contrast images were captured from 8 different fields for each replicate and the number of migrated cells from three replicates was scored (means ± SEM, n=3). Scale bars, 50μm. (B) The invasive properties of P1 and P2 cells were measured using Transwell inserts coated with matrigel. The cells were stained with 4′,6- diamidino-2-phenylindole (DAPI) after 48h and representative images were shown. Scale bars, 50μm. The data presented are a compilation of the number of invaded and migrated cells from three replicates (means ± SEM, n=3).
sc-12-0067_sm_SupplFigure6_P.tif2894KSupplemental Figure S6. P2 cells display increased migratory and invasive properties in transwell migration assay. (A) FACS-fractionated P1 and P2 cells were seeded in Transwell inserts at 50,000 cells/insert in serum-free growth media and allowed to migrate for 24h in response to normal growth media with 10% fetal bovine serum. Phase contrast images were captured from 8 different fields for each replicate and the number of migrated cells from three replicates was scored (means ± SEM, n=3). Scale bars, 50μm. (B) The invasive properties of P1 and P2 cells were measured using Transwell inserts coated with matrigel. The cells were stained with 4′,6- diamidino-2-phenylindole (DAPI) after 48h and representative images were shown. Scale bars, 50μm. The data presented are a compilation of the number of invaded and migrated cells from three replicates (means ± SEM, n=3).
sc-12-0067_sm_SupplFigure7.pdf17KSupplemental Figure S7. P2 cells show greater migratory potential in wound healing assay. FACS-fractionated P1 and P2 cells were cultured in growth medium with 0.5% fetal bovine serum for 12h prior to the generation of scratch wounds by plastic pipette tips. Three wounds were created for each replicate and phase contrast images (2 images per wound) were captured at the indicated time-points. Representative images were shown and the percentage of invaded area was calculated using Tscratch software 1 (means ± SEM, n=3). Scale bars, 50μm.
sc-12-0067_sm_SupplFigure7_P.pdf45KSupplemental Figure S7. P2 cells show greater migratory potential in wound healing assay. FACS-fractionated P1 and P2 cells were cultured in growth medium with 0.5% fetal bovine serum for 12h prior to the generation of scratch wounds by plastic pipette tips. Three wounds were created for each replicate and phase contrast images (2 images per wound) were captured at the indicated time-points. Representative images were shown and the percentage of invaded area was calculated using Tscratch software 1 (means ± SEM, n=3). Scale bars, 50μm.
sc-12-0067_sm_SupplFigure8A.pdf36KSupplemental Figure S8. Immunofluorescent staining of E-cadherin and β- catenin. A) Single cell-derived colonies were generated from purified P1 and P2 cells and immunofluoresent staining of E-cadherin using rat anti-E-cadherin antibody (DECMA-1) was performed. Epifluorescent images of representative P1 and P2 colonies were presented. Scale bars, 50μm. B) Western blotting analysis comparing the expression levels of E-cadherin, Vimentin and α-tubulin (loading control) between two independent fractions of P1 and P2 cells.
sc-12-0067_sm_SupplFigure8B.pdf19KSupplemental Figure S8. Immunofluorescent staining of E-cadherin and β- catenin. A) Single cell-derived colonies were generated from purified P1 and P2 cells and immunofluoresent staining of E-cadherin using rat anti-E-cadherin antibody (DECMA-1) was performed. Epifluorescent images of representative P1 and P2 colonies were presented. Scale bars, 50μm. B) Western blotting analysis comparing the expression levels of E-cadherin, Vimentin and α-tubulin (loading control) between two independent fractions of P1 and P2 cells.
sc-12-0067_sm_SupplFigure9.pdf56KSupplemental Figure S9. Blockade of a constitutively active TGF- β pathway promotes a P1 phenotype in GIF-14. Treatment with the TGF-β inhibitor SB431542 promotes P1/epithelial phenotype in GIF-14 cells. (A) Parental GIF-14 cells were pretreated with SB431542 for 15h prior to the addition of TGF-β1 for another 24h. Representative images were captured by phase contrast microscopy, which showed SB431542 induced an epithelial-like morphology. Scale bars, 50μm. (B) SB431542 treatment reduced the basal and induced expression levels of mesenchymal and EMT- associated markers measured by qRT-PCR. Normalized values are presented relative to those of the untreated sample. (C) GIF-14 cells were subjected to Hoechst 33342 and EpCAM/CD133 staining for flow cytometry after 3 days of SB431542 treatment, revealing a shift to a P1 staining pattern.
sc-12-0067_sm_SupplFigure9B.TIFF955KSupplemental Figure S9. Blockade of a constitutively active TGF- β pathway promotes a P1 phenotype in GIF-14. Treatment with the TGF-β inhibitor SB431542 promotes P1/epithelial phenotype in GIF-14 cells. (A) Parental GIF-14 cells were pretreated with SB431542 for 15h prior to the addition of TGF-β1 for another 24h. Representative images were captured by phase contrast microscopy, which showed SB431542 induced an epithelial-like morphology. Scale bars, 50μm. (B) SB431542 treatment reduced the basal and induced expression levels of mesenchymal and EMT- associated markers measured by qRT-PCR. Normalized values are presented relative to those of the untreated sample. (C) GIF-14 cells were subjected to Hoechst 33342 and EpCAM/CD133 staining for flow cytometry after 3 days of SB431542 treatment, revealing a shift to a P1 staining pattern.
sc-12-0067_sm_SupplFigure9C.pdf18KSupplemental Figure S9. Blockade of a constitutively active TGF- β pathway promotes a P1 phenotype in GIF-14. Treatment with the TGF-β inhibitor SB431542 promotes P1/epithelial phenotype in GIF-14 cells. (A) Parental GIF-14 cells were pretreated with SB431542 for 15h prior to the addition of TGF-β1 for another 24h. Representative images were captured by phase contrast microscopy, which showed SB431542 induced an epithelial-like morphology. Scale bars, 50μm. (B) SB431542 treatment reduced the basal and induced expression levels of mesenchymal and EMT- associated markers measured by qRT-PCR. Normalized values are presented relative to those of the untreated sample. (C) GIF-14 cells were subjected to Hoechst 33342 and EpCAM/CD133 staining for flow cytometry after 3 days of SB431542 treatment, revealing a shift to a P1 staining pattern.
sc-12-0067_sm_SupplFigure10.TIFF1167KSupplemental Figure S10. Expression levels of β-catenin and Desmoplakin in P1 and P2 subpopulations. FACS-fractionated P1 and P2 cells were treated with TGF- β1 for 24 and 48h and analyzed by qRT-PCR. Normalised data are expressed relative to uninduced P1 values (means ± SEM, n=4).
sc-12-0067_sm_SupplFigure11.pdf30KSupplemental Figure S11. Immunofluorescent staining for β-catenin in P1 or P2 colonies. Low magnification images show strong and uniform membrane staining of β-catenin in a typical P1 colony. In contrast, β-catenin staining is variable in P2 colonies. The robust membrane staining that was observed at the center of the colony was lost at the edges. This was accompanied by nuclear β-catenin staining observed frequently in detaching cells at the edge. Scale bars, 200 μm.
sc-12-0067_sm_SupplFigure12A.pdf31KSupplemental Figure S12. TGF- β signaling pathway mediates the interconversion of P1 and P2 phenotypes. (A) P1 and P2 cells were sorted from parental GIF-14 cells and expanded in vitro for 10 days. Hoechst 33342 and EpCAM/CD133 staining was performed on the indicated days and analyzed by flow cytometry. (B) P1 and P2 clonal lines were treated with TGF-β1 and SB431542 for 4 days, respectively. Changes in cell morphology were revealed in the representative phase contrast images. Magnification, 100x and 200x for inserts; scale bars, 50μm. (C) Phase contrast images of representative control and treated P1 and P2 colonies were captured after 3 days of TGF-β1 or SB431542 treatment. Scale bars, 50μm. (D) P1 and P2 clonal lines were treated with TGF-β1 and SB431542, respectively. After 11 days, cells were subjected to Hoechst 33342 and EpCAM/CD133 staining and flow cytometry analysis.
sc-12-0067_sm_SupplFigure12B.pdf52KSupplemental Figure S12. TGF- β signaling pathway mediates the interconversion of P1 and P2 phenotypes. (A) P1 and P2 cells were sorted from parental GIF-14 cells and expanded in vitro for 10 days. Hoechst 33342 and EpCAM/CD133 staining was performed on the indicated days and analyzed by flow cytometry. (B) P1 and P2 clonal lines were treated with TGF-β1 and SB431542 for 4 days, respectively. Changes in cell morphology were revealed in the representative phase contrast images. Magnification, 100x and 200x for inserts; scale bars, 50μm. (C) Phase contrast images of representative control and treated P1 and P2 colonies were captured after 3 days of TGF-β1 or SB431542 treatment. Scale bars, 50μm. (D) P1 and P2 clonal lines were treated with TGF-β1 and SB431542, respectively. After 11 days, cells were subjected to Hoechst 33342 and EpCAM/CD133 staining and flow cytometry analysis.
sc-12-0067_sm_SupplFigure12C_P1.pdf49KSupplemental Figure S12. TGF- β signaling pathway mediates the interconversion of P1 and P2 phenotypes. (A) P1 and P2 cells were sorted from parental GIF-14 cells and expanded in vitro for 10 days. Hoechst 33342 and EpCAM/CD133 staining was performed on the indicated days and analyzed by flow cytometry. (B) P1 and P2 clonal lines were treated with TGF-β1 and SB431542 for 4 days, respectively. Changes in cell morphology were revealed in the representative phase contrast images. Magnification, 100x and 200x for inserts; scale bars, 50μm. (C) Phase contrast images of representative control and treated P1 and P2 colonies were captured after 3 days of TGF-β1 or SB431542 treatment. Scale bars, 50μm. (D) P1 and P2 clonal lines were treated with TGF-β1 and SB431542, respectively. After 11 days, cells were subjected to Hoechst 33342 and EpCAM/CD133 staining and flow cytometry analysis.
sc-12-0067_sm_SupplFigure12C_P2.pdf47KSupplemental Figure S12. TGF- β signaling pathway mediates the interconversion of P1 and P2 phenotypes. (A) P1 and P2 cells were sorted from parental GIF-14 cells and expanded in vitro for 10 days. Hoechst 33342 and EpCAM/CD133 staining was performed on the indicated days and analyzed by flow cytometry. (B) P1 and P2 clonal lines were treated with TGF-β1 and SB431542 for 4 days, respectively. Changes in cell morphology were revealed in the representative phase contrast images. Magnification, 100x and 200x for inserts; scale bars, 50μm. (C) Phase contrast images of representative control and treated P1 and P2 colonies were captured after 3 days of TGF-β1 or SB431542 treatment. Scale bars, 50μm. (D) P1 and P2 clonal lines were treated with TGF-β1 and SB431542, respectively. After 11 days, cells were subjected to Hoechst 33342 and EpCAM/CD133 staining and flow cytometry analysis.
sc-12-0067_sm_SupplFigure12D.TIFF7667KSupplemental Figure S12. TGF- β signaling pathway mediates the interconversion of P1 and P2 phenotypes. (A) P1 and P2 cells were sorted from parental GIF-14 cells and expanded in vitro for 10 days. Hoechst 33342 and EpCAM/CD133 staining was performed on the indicated days and analyzed by flow cytometry. (B) P1 and P2 clonal lines were treated with TGF-β1 and SB431542 for 4 days, respectively. Changes in cell morphology were revealed in the representative phase contrast images. Magnification, 100x and 200x for inserts; scale bars, 50μm. (C) Phase contrast images of representative control and treated P1 and P2 colonies were captured after 3 days of TGF-β1 or SB431542 treatment. Scale bars, 50μm. (D) P1 and P2 clonal lines were treated with TGF-β1 and SB431542, respectively. After 11 days, cells were subjected to Hoechst 33342 and EpCAM/CD133 staining and flow cytometry analysis.
sc-12-0067_sm_SupplFigure13.TIFF1613KSupplemental Figure S13. Suppresses ATP-binding cassette transporters by wildtype RUNX3. Three independent infections were performed where parental GIF-14 cells were infected with the indicated lentiviruses for 6 days prior to TGF-β1 treatment for 24h. GFP-positive cells were enriched by FACS and subjected to qRTPCR. Normalised values are presented relative to the iG2 control untreated values (means ± SEM, n=3).
sc-12-0067_sm_SupplFigure14.pdf25KSupplemental Figure S14. Runx3−/− and Runx3+/+ GIF lines show differential basal expression of epithelial, mesenchymal and EMT-related markers. GIF cell lines were treated with TGF-β1 for 24h and 48h and analyzed by qRT-PCR. Normalised values are expressed relative to GIF-14 uninduced values (means ± SEM, n=3).
sc-12-0067_sm_SupplTable1.pdf52KSupplemental Table S1. Oligonucleotide primers and Taqman probes used for quantitative RT-PCR, related to Figure 2.
sc-12-0067_sm_SupplTable2.pdf33KSupplemental Table S2. Expression levels of various surface markers in GIF-14, P1 and P2 subpopulations, related to Figure 1. GIF-14 cells were co-stained with Hoechst 33342 and antibodies against various stem cell-related surface markers to assess their expression in P1 and P2 cells. EpCAM, CD133/Prominin-1 and CD49f/Integrin α6 were under-represented in the P2 subpopulation.
sc-12-0067_sm_SuppleData.pdf75KSupplementary Data
sc-12-0067_sm_SupplementalMovie1.mov2993KSupplementary Video S1. Time lapse imaging of a representative P1 colony by differential interference contrast (DIC) microscopy, related to Figure 3. P1 colonies were generated by seeding FACS-purified P1 cells at clonagenic densities on glass cover slips. Imaging was performed in a Sykes Moore chamber for a period of 10 h at 3 min intervals.
sc-12-0067_sm_SupplementalMovie2.mov2408KSupplementary Video S2. Time lapse imaging of a representative P2 colony by DIC microscopy, related to Figure 3. P2 colonies were generated by seeding FACSpurified P2 cells at clonogenic densities on glass cover slips. Imaging was performed in a Sykes Moore chamber for a period of 10 h at 3 min intervals.
sc-12-0067_sm_SupplementalMovie3A.mov13721KSupplementary Video S3A and S3B. Time lapse imaging of the transition from the P1 to P2 phenotype following TGF- β1 treatment of P1 colonies (Phase contrast microscopy), related to Figure 4. Single-cell derived P1 colonies generated on a low evaporation 6-well plate were imaged at 100x magnification. Representative P1 colonies of similar sizes were imaged concurrently in two separate wells. Colonies were imaged for 20 h prior to treatment to confirm their epithelial characteristics. Following treatment with 2.5 ng/ml TGF-β1 (S3A) and carrier control (S3B), the colonies were simultaneously imaged at 5 min intervals initially for 44 h, and then for a further 24 h following adjustments in the fields of view to capture the effects of TGF-β1-induced EMT.
sc-12-0067_sm_SupplementalMovie3B.mov14243KSupplementary Video S3A and S3B. Time lapse imaging of the transition from the P1 to P2 phenotype following TGF- β1 treatment of P1 colonies (Phase contrast microscopy), related to Figure 4. Single-cell derived P1 colonies generated on a low evaporation 6-well plate were imaged at 100x magnification. Representative P1 colonies of similar sizes were imaged concurrently in two separate wells. Colonies were imaged for 20 h prior to treatment to confirm their epithelial characteristics. Following treatment with 2.5 ng/ml TGF-β1 (S3A) and carrier control (S3B), the colonies were simultaneously imaged at 5 min intervals initially for 44 h, and then for a further 24 h following adjustments in the fields of view to capture the effects of TGF-β1-induced EMT.
sc-12-0067_sm_SupplementalMovie4A.mov18902KSupplementary Video S4A and S4B. Time lapse imaging of the transition from the P1 to P2 phenotype following TGF- β1 treatment (DIC microscopy), related to Figure 4. P1 colonies generated on glass cover slips were imaged simultaneously by differential interference contrast microscopy at 5 min intervals in two separate Sykes Moore chambers. Following the addition of 2.5 ng/ml TGF-β1 (S4A) or carrier control (S4B), representative P1 colonies were imaged initially for 40 h and, following adjustments in the fields of view, for an additional 44 h. In the TGF-β1- treated sample, colonies were imaged for 32 h prior to treatment to establish their P1 characteristics.
sc-12-0067_sm_SupplementalMovie4B.mov13300KSupplementary Video S4A and S4B. Time lapse imaging of the transition from the P1 to P2 phenotype following TGF- β1 treatment (DIC microscopy), related to Figure 4. P1 colonies generated on glass cover slips were imaged simultaneously by differential interference contrast microscopy at 5 min intervals in two separate Sykes Moore chambers. Following the addition of 2.5 ng/ml TGF-β1 (S4A) or carrier control (S4B), representative P1 colonies were imaged initially for 40 h and, following adjustments in the fields of view, for an additional 44 h. In the TGF-β1- treated sample, colonies were imaged for 32 h prior to treatment to establish their P1 characteristics.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.