Fig 2: AGO2:3′UTR binding clusters are associated with both up- and down-regulation of gene expression. (A) Distribution of gene expression changes for genes with significant AGO2:RNA-binding clusters. (B) Distribution of genes with more than one cluster. (C–F) Plot of gene expression changes versus cluster significance (C) AGO1, (D) AGO2, (E) AGO1/2 and (F) AGO1/2/3 knockout cells. For inclusion in analysis we required peaks to possess a P value <0.05 and a >4-fold enrichment in read number for wild-type verse AGO2 knockouts. See also Supplemental Figure S4.

Fig 3: miR-145-5p enhances Ago2 mRNA translation.a, b RIP (RNA Immuno Precipitation) assays performed in control (CTR) and miR-145-5p overexpressing (145-5p) MDA-MB-231 cells using antibodies directed to Ago2, Ago1, or IgG as negative control. mRNA abundance of Ago2 was evaluated by RT-qPCR and normalized over GAPDH mRNA; a representative experiment of a biological duplicate is shown. c Cytoplasmic extracts from MD-MB-231 cells overexpressing miR-145-5p and the relative control cells were loaded on 15–50% sucrose gradients and fractions measured by absorbance at 253 nm (left panel). Fraction density decreases from left to right; a representative profile out of two independent biological replicates is showed. Right panel: pooled fractions of RNA associated with polysomes, monosomes and free RNA were analyzed by RT-qPCR. The percentage of Ago2 mRNA enrichment, normalized over GAPDH mRNA is reported

Fig 4: Differential nuclear localization of Ago1 and Ago2.(A and B) Protein levels of Ago1, Ago2, RNA Polymerase II (RNAP II), and tubulin were detected in whole cell lysate (WCL), nuclear (Nuc), or cytoplasmic extracts (Cyto) from PC-3 cells (A) and in Nuc or Cyto fractions from RWPE-1 cells (B) using protein-specific antibodies. RNAP II and tubulin served as nuclear and cytoplasmic markers, respectively. (C) Localization of HA-tagged Ago1 (HA-Ago1) and Ago2 (HA-Ago2) were determined in their respective stable cell lines by immunoblot analysis using an antibody specific to the HA epitope. (D) Depicted is the cell fractionation protocol for isolating chromatin-bound protein. (E) All fractions including WCL from PC-3 cells were evaluated by immunoblot analysis. RNAP II and tubulin served as markers for chromatin-bound and unbound protein, respectively. Factors bound to chromatin are found in fractions P1 and S2. (F) Purified nuclei from stable cell lines expressing HA-Ago1 (PC3-HA-Ago1) or HA-Ago2 (PC3-HA-Ago2) were analyzed by IF. Antibodies specific to Ago1 or Ago2 detected both endogenous and exogenous forms. An antibody specific to the HA epitope visualized only HA-Ago1 or HA-Ago2. DAPI (blue) was used to counterstain nuclei. Representative immunofluorescent images were taken at 1000 X magnification (scale bar: 10 µm). (G) Immunofluorescent signal was quantified on the inner periphery (nuclear periphery) or interior (scattered) of purified nuclei. Results are shown as the percent mean distribution of signal (% nuclei) in 200 nuclei from each IF analysis. Negative controls are included in Figure S2.

Fig 5: Ago1-bound and Ago1-responsive genes are enriched for cancer pathways.(A) Illustrated is a schematic representation defining the AbGs-5 kb gene set as possessing Ago1 peaks within ±5 kb of their TSSs (marked in pink). Indicated are the top enriched pathways (P<0.01) for all AbGs-5 kb genes. KEGG pathway enrichment analysis was performed by using the DAVID bioinformatics tool. (B) Indicated are the top enriched pathways (P<0.05) for all up- and downregulated ArGs as determined by KEGG pathway enrichment analysis using the DAVID tool. All underlined pathways are shared by the AbGs-5 kb and ArG gene sets. (C) The schematic illustrates exemplary AbGs and ArGs involved in oncogenic pathways. Major gene nodes for the indicated pathways are represented by genes grouped as Ago1-bound/regulated genes (AbGs/ArGs; pink) or non-Ago1 bound/regulated genes (non-AbGs/ArGs; grey).