(A) Schematic representation of the Renilla luciferase (Rluc) reporter mRNAs. Sequences of the let-7-binding sites (RL-6xB) and mutated seed sites (RL-6xBMut) are shown below the drawings.
(B) Time course of RL-6xB-pA and RL-6xBMUT-pA mRNA deadenylation as determined by autoradiography. Reporter mRNAs were incubated in the presence or absence of 10 μM cycloheximide, 1 mM hippuristanol, or 10 nM 2′-O-Me oligonucleotide (either anti-let-7a or anti-miR122).
(C) A time course of translation of RL-pA, RL-6xB-pA, and RL-6xB-pA in the presence of anti-let-7 2′-O-Me. Average percentage repression is labeled below each time point. Error bars represent the standard deviation of three independent experiments.
(D) Schematic representation of the 6xB-3′UTR reporter RNA and time course of 6xB-3′UTR and 6xBMUT-3′UTR RNA deadenylation in a Krebs extract as determined by autoradiography. Reporters were incubated in the presence or absence of 10 nM 2′-O-Me oligonucleotide (either anti-let-7a or anti-miR122), and their stability was monitored by autoradiography.
(E) Schematic representation of the 6xB-3′UTR reporter RNAs with either 98As or 150As (*). Time course of 6xB-3′UTR and 6xB-3′UTR* deadenylation in a Krebs extract as determined by autoradiography. Polyadenylated and deadenylated mRNAs are marked on the right of the figure.

(A–D) Time course of RL-pA and RL-6xB-pA translation in rabbit anti-HA- (A), rabbit anti-CAF1- (B), mouse anti-HA- (C), and mouse anti-Ago2-depleted Krebs extracts (D). Average percentage repression is labeled below each time point. Error bars represent the standard deviation of three independent experiments.
(E) 6xB-3′UTR RNA deadenylation in the presence or absence of 10 nM anti-let-7a 2′-O-Me in control (mouse anti-HA) or anti-Ago2-depleted Krebs extract. 6xB-3′UTR RNA deadenylation was followed by autoradiography. Polyadenylated and deadenylated mRNAs are marked on the right of the figure.
(F) 6xB-3′UTR RNA deadenylation in control (rabbit anti-HA) or anti-CAF1-depleted extract in the presence or absence of either 10 nM anti-let-7a 2′-O-Me oligonucleotide, or WT or D40A HA-CAF1 protein.
(G) Wild-type and mutant hook peptides derived from GW182.
(H) 6xB-3′UTR RNA deadenylation in Krebs extract in the presence or absence of either GST or GST hook peptides at concentrations ranging from 0.1 to 2.0 μg per reaction, respectively.

(A) 6xB-3′UTR RNA deadenylation in Krebs extract in the presence or absence of increasing concentrations (0.15, 0.5, 1.0, and 3.0 μg per reaction) of wild-type or mutant GST-eIF4G 41-244.
(B) 6xB-3′UTR RNA deadenylation in mock- or PABP-depleted Krebs extract in the presence or absence of increasing concentrations (25, 50, or 100 ng per reaction, respectively) of either wild-type (lanes 5–7) or M161A PABP (lanes 10–12) and/or wild-type (lanes 8 and 13) or mutant (lanes 9 and 14) GST-eIF4G (41-244).
(C) Quantification of deadenylated bands as a percentage of total RNA in (B) is shown in bar graphs (with standard deviations).

(A) Alignment of GW182 DUF sequences with Paip2 PAM2 motif.
(B) Western blot analysis of GST pulldowns of TNRC6C (1260-1690)-FLAG incubated with various C-terminal (C1 and C2) fragments of GST-PABP and probed with anti-FLAG and anti-GST antibodies.
(C) Western blot analysis of GST pulldowns of GST-PABP incubated with TNRC6C (1260-1690)-FLAG and/or wild-type or mutant PAM2 peptide and probed with anti-FLAG and anti-GST antibodies.
(D) 6xB-3′UTR RNA deadenylation in Krebs extract in the presence of increasing concentrations (1, 10, 50, and 100 μM) of wild-type or mutant (F > A) Paip2 PAM2 peptides.
(E) 6xB-3′UTR RNA deadenylation in Krebs extract, as determined by autoradiography, in the presence of increasing concentrations (0.5, 1.0, and 2.0 μg per reaction) of GST, TNRC6C (1260-1690), or TNRC6C (800-1360).

(A) MuDPIT analysis of Ago1- and Ago2-interacting proteins. Identified proteins are listed along with corresponding peptide coverage for Ago1 and Ago2 coimmunoprecipitations.
(B) Immunoprecipitation of endogenous Ago2 protein from micrococcal nuclease-treated Krebs extract using anti-Ago2 antibody. Immunoprecipitated complexes were subjected to SDS-PAGE and probed with anti-Ago2 antibody, anti-CAF1 antibody, anti-PABP antibody, or anti-eIF4E antibody.
(C) Immunoprecipitation of endogenous CAF1 protein from micrococcal nuclease-treated Krebs extract using anti-CAF1 antibody. Immunoprecipitated complexes were subjected to SDS-PAGE and probed with anti-Ago2 antibody, anti-CAF1 antibody, anti-CCR4 antibody, or anti-eIF4E antibody.
(D) Pulldown of Ago2, CCR4, and CAF1 from micrococcal nuclease-treated Krebs extracts using biotin-conjugated anti-let-7 2′-O-Me oligo-nucleotide and streptavidin Dynabeads. Isolated complexes were subjected to SDS-PAGE and probed with anti-Ago2 antibody, anti-CAF1 antibody, anti-CCR4 antibody, anti-Tob antibody, or anti-eIF4E antibody.

(A) Western blot analysis of Krebs-2 extracts depleted with either GST (Control Extract) or GST-Paip2 (PABP-depleted Extract) probed with anti-PABP antibody and anti-β-actin antibody.
(B) A-capped 6xB-3′UTR RNA incubated in either mock-depleted (lanes 1–2) or PABP-depleted extract (lanes 3–6). PABP-depleted extract was supplemented with recombinant GST, GST-PABP (100 ng, which is the equivalent of roughly 50% of endogenous PABP present in an in vitro reaction), and RNA stability was monitored by autoradiography. Polyadenylated and deadenylated mRNAs are marked on the right of the panel.

(A) Schematic representation of human TNRC6C and GST- and FLAG-tagged recombinant protein fragments. Western blot analysis of GST pull-downs of PABP incubated with GST or various fragments of GST-TNRC6C-FLAG and probed with anti-PABP and anti-GST antibodies.
(B) Schematic representation of human TNRC6C HA-tagged fragments transfected into HEK293 cells. Cell extracts of HEK293 cells, transiently expressing the indicated fusion proteins, were incubated with Anti-HA Affinity Matrix (Roche), and immunoprecipitated proteins were analyzed by western blotting using the indicated antibodies. Inputs represent 1% of the cell extract used for IP. Nontransfected cells served as a control.
(C) Cell extracts of HEK293 cells transiently expressing GST-ΔN1370 were pulled down using glutathione Sepharose resin in the presence or absence of micrococcal nuclease. GST pulldowns were analyzed by western blotting using anti-PABP, anti-eIF4G, and anti-GST antibodies. Non-transfected cells served as a control.

(i) mRNA circularization via eIF4G-PABP interaction stimulates cap-dependent translation by enhancing eIF4E’s binding to the mRNA 5′ cap structure (strong binding [Kahvejian et al., 2005]).
(ii) miRISC binds to its target site in the 3′UTR. GW182 binds to PABP, hypothetically inhibiting its interaction with eIF4G, thereby repressing cap-dependent translation by decreasing eIF4E’s binding to the 5′ cap structure (weaker binding), and sequestering the poly(A) tail into the vicinity of CAF1 and CCR4 deadenylases (illustrated by an arrow) to facilitate deadenylation of the mRNA. The interaction between CAF1/CCR4 and Ago2 is probably indirect through other proteins (depicted as question marks).