SupplementalFigure Legends.
Supplemental Fig S1. SIX2 is overexpressed in breast cancer and correlates with poor prognosis.(A) SIX2 expression is high in human invasive breast carcinoma compared to normal breast (data from Oncomine). (B) Expression of SIX2 in cancers compared to normal tissue in Yu multi-cancer dataset from Oncomine. (C) Examination of SIX2 expression in the TCGA breast dataset revealsa correlation of SIX2 with increased metastatic stage and with patients that died within 5 years. Oncomine™ (Compendia Bioscience, Ann Arbor, MI) was used for analysis and visualization.
Supplemental Fig S2. Six2 KD does not affect proliferation or levels of VEGF-C.(A)Proliferation of 66cl4-NS and 66cl4-Six2 KD cells is not significantly different. Quantification of the BrdU positive cells in 66cl4-NS and 66cl4-Six2 KD lines, 0 and 24 hrs after release from serum starvation . (B) VEGF-C mRNA expression is not altered in control and Six2KD tumors. Six2 and VEGF-C levels were determined by real-time PCR and normalized to Cyclophilin in 66cl4-NS and Six2 KD tumors.
Supplemental FigS3.Six2 overexpression in 4TO7 cells alters cell morphology but does not increase cell proliferation.(A) Six2 expression in 4TO7 cells leads to changes in cell morphology. 4TO7-Six2 cells are smaller andmore rounded when compared to 4TO7-pcDNA control cells. (B) Quantification of the BrdU positive population from 4TO7-pcDNA and 4TO7-Six2 cells at 0 and 24 hrs release from serum starvation.
Supplemental FigS4.Six2 expression in 4TO7 cells increases lung metastasis.Animals injected with 4TO7-pcDNA or 4TO7-Six2 cells were sacrificed and lung metastases were confirmed by histology (H&E staining).Representative images of lungs in 4TO7-pcDNA or 4TO7-Six2 injected mice are shown.
Supplemental FigS5.Six2 overexpression leads to Zeb2 upregulation in 4TO7 and HMLE cells. (A)Microarray analysis for genes related to epithelial-mesenchymal transition (EMT) in 4TO7-pcDNA and 4TO7-Six2 cells. (B) Zeb2 mRNA expression is modestly increased in 4TO7-Six2 expressing cells.Zeb1 and Zeb2 mRNA expression was determined by real-time PCR and normalized to Cyclophilin in 4TO7-pcDNA and 4TO7-Six2 cells. (C) Six2 mRNA expression was measured using real-time PCR in HMLE-pcDNA and HMLE-Six2 cells (upper). Expression of E-cadherin and Zeb2 in HMLE-pcDNA or HMLE-Six2 cells was detected by Western blotting (bottom).**, P<0.05.
Supplemental Fig S6. SIX2 downregulates miR-8 family members, which when restored to SIX2 overexpressing cells, can reverse the SIX2-induced upregulation of Zeb2. (A) MicroRNA expression in 4TO7-pcDNA and 4TO7-Six2 cells. Stable overexpression of Six2 in 4TO7 leads to a decrease in miR200a and miR200b expression as determined using qRT-PCR. (B) 10nM of the different microRNA mimics (miR-200a and miR-200b) were transfected into 4TO7-Six2 cells. 72 hours after transfection, whole cell lysates were collected and Western blotting was performed to detect Zeb2 and -actin.(C) MicroRNA expression in HMLE-pcDNA and HMLE-Six2 cells. Stable overexpression of Six2 in HMLE cells leads to a decrease in miR200a, miR200b and miR200cexpression as determined using qRT-PCR. (D) 10nM of the different microRNA mimics were transfected into HMLE-Six2 cells. 72 hours after transfection, whole cell lysates were collected and Western blotting was performed to detect Zeb2 and -actin.*, P<0.05.***, P<0.001.
Supplemental FigS7. High levels of SIX2 are found in breast cancer cell lines exhibiting E-cadherin hypermethylation and silencing. Expression of E-cadherin and Six2 in breast cancer cell lines was detected by Western blotting. HDAC was used as loading control (top). Methylation value in the listed breast cancer cell lines (table) was retrieved (1) and each value represents the methylationfrom the indicated site to transcription start site (TSS) of CDH1.
Supplemental FigS8. SIX2 expression inversely correlates with CDH1 expression in triple negative and non-triple negative breast cancers. SIX2 and CDH1 expression values were retrieved from an Oncomine microarray data set (Hatzis) and were plotted by expression value divided by triple negative or other biomarker status. Statistical analysis was performed using the Pearson’s test.
Reference
1.Daemen A, Griffith OL, Heiser LM, et al. Modeling precision treatment of breast cancer. Genome Biol 2013; 14:R110.
1