miR-200c regulates induction of apoptosis through CD95 by targeting FAP-1

Abstract

Tumor progression shares many characteristics with the process of epithelial-to-mesenchymal transition (EMT). Cells that have undergone an EMT are known to have an increased resistance to apoptosis. CD95/Fas is an apoptosis-inducing receptor expressed on many tissues and tumor cells. During tumor progression CD95 is frequently downregulated, and tumor cells lose apoptosis sensitivity. miR-200 microRNAs repress both the EMT-inducing ZEB1 and ZEB2 transcription factors. We now demonstrate that miR-200c sensitizes cells to apoptosis mediated by CD95. We have identified the apoptosis inhibitor FAP-1 as a target for miR-200c. FAP-1 was demonstrated to be responsible for the reduced sensitivity to CD95-mediated apoptosis in cells with inhibited miR-200. The identification of FAP-1 as an miR-200c target provides a molecular mechanism to explain both the downregulation of CD95 expression and the reduction in sensitivity of cells to CD95-mediated apoptosis that is observed in the context of reduced miR-200 expression during tumor progression.

Figure 1. Altering Levels of miR-200 Changes Sensitivity of Cells to CD95-induced Apoptosis

(A) Phase contrast pictures showing morphological changes of cells transfected with scrambled pre-miRNA (scr) and pre-miR-200c (3 times over 9 days) or cells stably infected with a retrovirus expressing scrambled oligo (vec) and miR-200c. Levels of mir-200c in retrovirus infected cells were quantified by real-time PCR. (B) Cells shown in A stimulated through CD95. HCT116 cells were transfected with LNA200 seven times prior to stimulation. Cells ectopically expressing miR-200c or cells treated with LNA200 were stimulated with different concentrations of anti-APO-1 for 20 hrs and apoptosis was quantified by MTS assay. (C) CAKI-1 cells transfected as in B were treated with 1 μg/ml anti-APO-1 and 1 μg/ml cycloheximide for 20 hrs. (D). Cells with increased or reduced levels of miR-200 were treated with 2 μg/ml of staurosporine for 20 hrs and cell death was quantified by MTS assay. Asterisks indicate p-values * p<0.05, ** p<0.01, *** p<0.001. Note: different CD95 stimuli were used in the assays shown in A-C because Type I cells are not sensitive to sCD95L, and Type II cells are not sensitive to non-crosslinked anti-APO-1 (Algeciras-Schimnich et al., 2003). The horizontal bars in B represent the mean. Values in graphs A, B and D represent the mean −/+ s.d. from three independent experiments.

Figure 2. FAP-1 Is Expressed in Mesenchymal Cells

(A) Schematic diagram of the 3′UTR of FAP-1 with the location of the putative miR-200b,c/429 target site. (B) Relative expression of FAP-1 protein in epithelial and mesenchymal cells among the NCI60 cells. (C,D) Correlation between expression level of miR-200c (% of max) (C) or let-7d (D) and FAP-1 protein in the NCI60 cells. Expression values were transformed to rank values in both axes, and red straight lines indicate linear regression for transformed rank values. Rho = Spearman Rank Correlation Coefficients between miRNAs and FAP-1 expression, and p values are given. (E) Detection of FAP-1 and E-cadherin by Western blot analysis of CAKI-1 and ACHN cells infected with lentiviruses expressing control shRNA or shRNAs for FAP-1. β-actin was detected to demonstrate equal loading. (F) Real-time PCR analysis of FAP-1 and E-cadherin in CAKI-1 cells expressing control shRNA or FAP-1 shRNAs relative to expression of GAPDH. Values in the graph in F represent the mean −/+ s.d. from three independent experiments.

Figure 3. Identification of FAP-1 as a Target of miR-200c

(A) ACHN, CAKI-1 and SKOV3ip cells were transfected with scrambled pre-miRNA or pre-miR-200c. HCT116 cells were serially treated seven times with LNA scrambled or LNA200. Expression of FAP-1 and E-cadherin was determined by Western blot analysis. (B) HCT116 cells were transfected as in A. FAP-1 mRNA level was determined by semi-quantitative RT-PCR. Actin mRNA was determined as a control. (C) 293T cells were transfected with reporter plasmids containing the 3′UTR of FAP-1 (psiCHECK 3′UTR FAP-1wt) or its corresponding mutant (psiCHECK 3′UTR FAP-1mut) together with either 1 pmol of pre-miRNA-scrambled or pre-miR-200c. Reporter plasmids with a fragment of the 3′UTR of ZEB2 (harboring sites 4, 5 and 6 of the seven mir-200 sites found in the full length) and its mutant (Park et al., 2008) were used as positive controls. Renilla luciferase activity was normalized to firefly luciferase activity. (D) IGROV-1 (miR-200 low) and HCT116 (miR-200 high) cells were transfected with reporter plasmid psiCHECK 3′UTR FAP-1. (E) IGROV-1 cells were transfected with psiCHECK 3′UTR FAP-1wt or its mutant together with 1 pmol of scrambled pre-miRNA, pre-miR-200a or pre-miR-200c. HCT116 cells were transfected with the same reporter plasmids together with 50 nM of LNA scrambled or LNA-200 oligos. * p<0.05, ** p<0.001. Values in C, D and E represent the mean −/+ s.d. from three independent experiments.

Figure 4. Targeting of FAP-1 by miR-200c Regulates Sensitivity to CD95-induced Apoptosis

(A) HCT116 cells infected with control shRNA or FAP-1 shRNA#2 were serially treated seven times with scrambled LNA or LNA-200. Cells were stimulated with different concentration of sCD95L and apoptosis was measured by MTS assay. (B) Real-time PCR analysis of FAP-1, E-cadherin, ZEB1, and ZEB2 in HCT116 cells. The insert shows a Western blot analysis of the same experiment. (C) Flow cytometric analysis of surface CD95 expression of cells with increased or reduced miR-200. Scrambled oligo or miR-200c was introduced in ACHN, CAKI-1, HeyA8, and SF295 cells either transiently or stably. For HCT116 cells, cells were serially treated seven times with scrambled LNA or LNA-200. Percentage of CD95 positivity and mean fluorescence intensity (MFI) were determined. S.D., standard deviation. Values in A and B represent the mean −/+ s.d. from three independent experiments.