MAD4 is a direct target of miR-125b-5p.

<p>(A) The miR-125b-5p responsive element in <i>MAD4</i> gene 3′UTR and the base-paired to miR-125b-5p are showed on the top row of the schematic diagram. The psiCheck2 reporter constructs containing the <i>MAD4</i> 3′UTR (top row) or the mutated one (obtained by deleting the seed sequence of miR-125b-5p as shown in the lower row of the schematic diagram) were assayed. (B) Luciferase reporter assay (taken after 24h transfection in Hek-293 cells) showing that miR-125b-5p represses the luciferase activity of the reporter construct containing the MAD4 responsive element (psiCheck2-MAD4). Deletion of the miR-125b-5p responsive element in the MAD4 3′UTR (psiCheck2-Mut) rescued the luciferase activity. Renilla luciferase activities were normalized to the values of the internal Firefly and to the scrambled transfection control. (C) Detection of the MAD4 and cMyc protein levels in CTCL cells (SeAx and MyLa) after miR-125b-5p overexpression or downregulation. β-actin was the loading control and protein changes are showed as in <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0059390#pone-0059390-g001" target="_blank">Figure 1A</a>. (D) Changes in the relative MAD4 mRNA level detected by qRT-PCR in CTCL cells. (E) Evaluation of the efficient MAD4 silencing in SeAx cells by Western Blot. (F) Decrease of the percentage of the annexin-V positive cells after MAD4 silencing in SeAx cells treated with bortezomib (48h) comparing to scrambled siRNA. (G) A mechanistic model describing the regulative loop involving cMyc and miR-125b-5p in cell resistance to bortezomib. cMyc induces the transcriptional inhibition of miR-125b-5p that directly inhibits MAD4 and induces cMyc accumulation. All quantitative data are shown as mean ± SD (*<i>P</i>< 0.05).