SEARCH

SEARCH BY CITATION

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re-organised for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
emmm201302507.reviewer_comments.pdfPDF document524KReview Process File
emmm201302507-sm-0002-SuppFig-S1.pdfPDF document1346K

Figure S1. ZIP4 induces the expression of miR-373 in pancreatic cancer. (A) Silencing of ZIP4 causes reduced expression of miR-373 in MIA PaCa-2 cell. (p= 0.023, t-test, n= 3).(B) Overexpression of ZIP4 upregulates miR-373 in AsPC-1 cell. Data were expressed as mean ± SD of triplicate values. (p= 0.0025, t-test, n= 3). (C) Scatterplot of miR-373 expression versus ZIP4 expression in pancreatic cancer tissues from two independent cohorts (cohort 1, US cohort, n= 15; cohort 2, Shanghai cohort, n= 25). Pearson correlation = 0.670, p= 0.006 for cohort 1, and Pearson Correlation = 0.641, p= 0.001 for cohort 2, respectively. Residual plots showed no evidence of lack of fit or substantial departures from normality. (D) and (E) IHC staining of ZIP4 in two representative pancreatic cancer tissues (20×), showing the positive staining of ZIP4 in tumour cells, and negative staining in stromal, and normal duct cells. The scale bar is 100 µm. ZIP4 induces the expression of miR-373 in pancreatic cancer. (F) Zinc-dependent upregulation of miR-373 in MIA-ZIP4 cells compared with MIA-V cells. Cells were treated with 4 µM TPEN (Life for 1 h followed by DMEM incubation with different zinc concentration for 24 h. The expression of miR-373 were determined by real time PCR before (control) and after TPEN treatment (from left to right, p= 0.0327, 0.0366, 0.043; t-test, n= 3). (G) Zinc-dependent downregulation of miR-373 in AsPC-shZIP4 cells compared with AsPC-shV cells. The expression of miR-373 were determined by real time PCR as above. Data were expressed as mean ± SD of triplicate values (significant p-value = 0.0462, t-test, n= 3).

Figure S2. The expression of miR-373 is regulated by the zinc finger transcription factor CREB. (A) Densitometry analysis of the WB results in (A). (from top to bottom, significant p-value = 0.0412, 0.0203; t-test, n= 3). Data were expressed as the mean ± SD of triplicate values. The expression of miR-373 is regulated by the zinc finger transcription factor CREB. (B) Predicted CREB binding sites in miR-373 promoter region. (C) Mutation primers of CREB binding sites in miR-373 promoter region. The expression of miR-373 is regulated by the zinc finger transcription factor CREB. (D) Promoter activity. The promoter reporter construct was co-transfected with control plasmid pRL-TK into HEK293 cells, the promoter activity was determined by a chemiluminescence reader (t-test, n= 3). (E), (F) and (G) Mutational analysis of miR-373 promoter. The wild type and mutant promoter constructs were co-transfected with control plasmid pRL-TK into (E) HEK293 (plus mock or ZIP4 vector) (t-test, n= 3), (F) MIA PaCa-2 (t-test, n= 4) or (G) PL45 cells (t-test, n= 6). Data were expressed as mean ± SD of triplicate values.

Figure S3. miR-373 promotes ZIP4-mediated pancreatic cancer cell invasion and proliferation. (A) Cell cycle analysis by flow cytometry. Cells were starved for 48 h and then released by incubation with FBS 10% DMEM medium for 2 h. Then cells were stained with PI (BD Bioscience) before acquisition by a flow cytometer (BD LSRII). The percentage of cells in G1, S and G2/M phases was measured and shown as mean ± SD of triplicate values (for G1 and S phase, p= 0.0015, 0.0013, t-test, n= 3). miR-373 promotes ZIP4-mediated pancreatic cancer cell invasion and proliferation. (B) BrdU incorporation analysis by flow cytometry. Cells were starved for 48 h and then released by incubation with DMEM medium with 10% FBS for 2 h. Cells were treated with 10 µM APC labelled BrdU for 10 mins and 1:50 anti-BrdU Ab (BD Bioscience), then gated by cell width and area during acquisition by a flow cytometer. H1 stands for BrdU negative cells; since both cells in S-phase and G2/M-phase would be BrdU positive by single staining, and (A) has shown that almost no cells were observed in G2/M phase, H2 mainly represents BrdU positive cells in S phase. Representative histograms are shown indicating the BrdU positive cells ratio in MIA-ZIP4-AntiC (Left) and MIA-ZIP4-Anti373 (Right) cells. The quantitative data were also shown (p= 0.032, t-test, n= 3). (C) Apoptosis analysis by Annexin-V assay. Cells were harvested with trypsin and incubated with APC labelled Annexin-V Ab (BD Bioscience). Cells were then doubly stained by Near-IR fluorescent reactive dye (Invitrogen), and excited separately with 633 and 775 nm laser in a BD flow cytometer. H1 stands for live cells, and H2 stands for apoptotic cells (p= 0.1553, t-test, n= 3). All data were shown as mean ± SD of triplicate values. miR-373 promotes ZIP4-mediated pancreatic cancer cell invasion and proliferation. (D) Cell proliferation was measured by MTT assay in MIA-V-AntiC and Anti373 cells. All data were standardised compared with day 0. (E) Wound healing assay in MIA-V-AntiC and Anti373 cells. Cells were treated with 10 μg/ml mitomycin (Sigma) for 2 h. A representative field of wound healing was captured and recorded at 0 and 24 h. All data were shown as mean ± SD of triplicate values (p= 0.5994, t-test, n= 3). Scale bar equals to 500 µm.

Figure S4. miR-373 is required by ZIP4 to promote pancreatic cancer growth and metastasis. Real time RT-PCR showed that miR-373 was significantly increased in (A) AsPC-shZIP4-Pre373 cells (p < 0.0001, t-test, n= 3), and was reduced in (B) MIAZIP4-Anti373 cells (p= 0.0342, t-test, n= 3); data were expressed as mean ± SD of triplicate values. (C) H&E staining with xenograft tumour sections in 4× and 10× magnification, the scale bars are 50 µm (4×) and 200 µm (10×).

Figure S5. TP53INP1, LATS2 and CD44 are target genes of miR-373 controlled by ZIP4. Expression of TP53INP1, LATS2 and CD44 in (A) MIA PaCa-2 (from left to right, p= 0.0086, 0.0108, 0.0013, t-test, n= 3), and (B) AsPC-1 cells (from left to right, p= 0.0043, 0.01987, 0.0107; t-test, n= 3). All data were shown as mean ± SD of triplicate values. Silencing of CREB leads to increased expression of miR-373 downstream target genes in (C) AsPC-1 and (D) MIA PaCa-2 cells. Cells were transfected with siRNA against CREB (Mission siRNA from Sigma) with the XtremeGENE 9 (Roche) following the standard protocol. The total cell lysates were assayed for Western blot at 48 h post transfection. TP53INP1, LATS2 and CD44 are target genes of miR-373 controlled by ZIP4. Predicted miR-373 binding sites in the 3′UTR regions of potential target genes: TP53INP1 (E), LATS2 (F), and CD44 (G). TP53INP1, LATS2 and CD44 are target genes of miR-373 controlled by ZIP4. (H) Mutation primer sequences of 3′UTR regions of miR-373 target genes. The putative miR-373 biding sites were abolished by site directed mutagenesis.

Figure S6. miR-373 target genes mediate ZIP4-induced pancreatic cancer growth. (A) Western blot confirmed the knockdown of TP53INP1 LATS2 and CD44 in both TP53INP1, LATS2, MIA PaCa-2 and BxPC-3 cells (from left to right, p= 0.0122, 0.0051, 0.0002, 0.0203, 0.0016, 0.0153; t-test, n= 3). (B) Cell proliferation in BxPC-shTP53INP1, BxPC-shLATS2 and BxPC-shCD44 cells. Cell proliferation was determined by MTT assay and compared with day 0 value (t-test, from top to bottom, n= 5, 4, 5). (C) Migration of MIA-shLATS2 and BxPC-shLATS2 cells were increased upon the knockdown of LATS2 (from left to right, p= 0.011, 0.027; t-test, n= 3). Data were expressed as the mean ± SD of triplicate values.

Figure S7. Summary of the metastasis. The major symptoms of the nude mice with xenografts of MIA-shTP53INP1, MIA-shLATS2 and MIA-shCD44 cells were summarised and compared to that of the control group MIA-shV.

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.