A role for the P-body component GW182 in microRNA function

Abstract

In animals, the majority of microRNAs regulate gene expression through the RNA interference (RNAi) machinery without inducing small-interfering RNA (siRNA)-directed mRNA cleavage. Thus, the mechanisms by which microRNAs repress their targets have remained elusive. Recently, Argonaute proteins, which are key RNAi effector components, and their target mRNAs were shown to localize to cytoplasmic foci known as P-bodies or GW-bodies. Here, we show that the Argonaute proteins physically interact with a key P-/GW-body subunit, GW182. Silencing of GW182 delocalizes resident P-/GW-body proteins and impairs the silencing of microRNA reporters. Moreover, mutations that prevent Argonaute proteins from localizing in P-/GW-bodies prevent translational repression of mRNAs even when Argonaute is tethered to its target in a siRNA-independent fashion. Thus, our results support a functional link between cytoplasmic P-bodies and the ability of a microRNA to repress expression of a target mRNA.

Figure 1. Argonaute proteins interact with GW182, a component of mammalian P-/GW- bodies

(a) Myc-tagged Ago2 protein was expressed in 293 cells. Immunoprecipitates, prepared using a human autoantiserum (18033 serum) that recognizes GW182 or control antibodies, were examined by Western blotting with an anti-myc antibody. (b) Argonaute proteins were visualized using and anti-myc epitope antibody and an Alexa Fluor 488 (green) secondary antibody. GW182 was visualized using a human autoantiserum (IC-6 serum or 18033 serum) and an Alexa Fluor 594 (red) secondary antibody. The scale bar indicates 10 μm.

Figure 2. Suppression of GW182 disrupts P-/GW bodies

(a) GFP-GW182 protein was expressed in HeLa cells. The effect of a co-transfected siRNA against GW182 was examined by western blotting with an anti-GFP antibody. (b) Transfection of siRNA against GW182 in HeLa cells reduces the number and size of the P-/GW- bodies examined by indirect immunofluorescence microscopy using the IC-6 serum which recognizes GW182 protein. (c) Suppression of GW182 expression in HeLa cells reduces the number of Dcp1 foci formation revealed by staining with an anti-Dcp1a antibody. Staining with a human autoantisera that recognizes GW182 (IC-6) is shown for comparison. (d) Suppression of GW182 expression in HeLa cells reduces the Dcp2 foci formation revealed by staining with an anti-Dcp2 antibody. In this case, only the IC-6 antiserum was used to highlight GW182 localization. The scale bar indicates 10 μm.

Figure 3. Suppression of GW182 expression impairs gene silencing

(a) Cells were transfected with a miRNA reporter in the presence or absence of the miRNA mimetic CXCR4 siRNA as indicated. Also included were either control siRNAs or siRNAs that suppress Dcp2, Xrn1 or GW182, as indicated. In all cases, transfection rates were normalized using a co-delivered firefly luciferase plasmid. (b) The experiment was carried out similarly to (a) except that the reporter contained a single perfect binding site for the CXCR4 siRNA.

Figure 4. Localization of Argonaute proteins in the P-/GW- bodies is required for suppression of a tethering reporter in small-RNA independent manner

(a) Localization of λN-Ago fusion proteins in HeLa cells. HA epitope tagged fusions between Lamda N protein and Ago1 or Ago2 were expressed in HeLa cells. These localize to discrete cytoplasmic foci as shown by staining with FITC-conjugated anti-HA. A HIWI fusion protein fails to accumulate in these foci. Mutations within the PAZ domain (PAZ9 or PAZ10) disrupt small RNA binding as previously shown . These same mutations disrupt the discrete localization of the λN-Ago fusion proteins. (b) A Renilla luciferase reporter containing 5 BoxB sites, that bind lamdaN, was transfected into HeLa cells. Various Ago fusion proteins were co-delivered as indicated. Also, in each case Renilla activity was normalized to a co-transfected firefly luciferase plasmid. As a control, An HA-epitope-Ago2 fusion that does not bind the boxB sites within the reporter was used to set the 100% level in the assay. λN-fusions to either Ago1 or Ago2 or to mutant Ago2 proteins carrying 9 or 10 mutations in the PAZ domain were tested for their ability to repress as indicated. As was previously shown lamdaN-HIWI was inert in this assay. (c) PAZ mutant Ago2 proteins retain the ability to associate with GW182 protein. The interaction between the mutants and GW182 was examined by immunoprecipitation with 18033 serum followed by western blotting with an anti-myc antibody. The scale bar indicates 10 μm.