Pin1 and WWP2 regulate GluR2 Q/R site RNA editing by ADAR2 with opposing effects
2011; Springer Nature; Volume: 30; Issue: 20 Linguagem: Inglês
10.1038/emboj.2011.303
ISSN1460-2075
AutoresRoberto Marcucci, James Brindle, Simona Paro, Angela Casadio, Sophie Hempel, Nicholas A. Morrice, Andrea Bisso, Liam P. Keegan, Giannino Del Sal, Mary A. O’Connell,
Tópico(s)Viral Infections and Immunology Research
ResumoArticle16 August 2011Open Access Pin1 and WWP2 regulate GluR2 Q/R site RNA editing by ADAR2 with opposing effects Roberto Marcucci Roberto Marcucci MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Western General Hospital, Edinburgh, UK Search for more papers by this author James Brindle James Brindle MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Western General Hospital, Edinburgh, UK Search for more papers by this author Simona Paro Simona Paro MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Western General Hospital, Edinburgh, UK Search for more papers by this author Angela Casadio Angela Casadio MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Western General Hospital, Edinburgh, UK Search for more papers by this author Sophie Hempel Sophie Hempel MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Western General Hospital, Edinburgh, UK Search for more papers by this author Nicholas Morrice Nicholas Morrice The Beatson Institute for Cancer Research, Glasgow, UK Search for more papers by this author Andrea Bisso Andrea Bisso Laboratorio Nazionale CIB, Trieste, Italy Dipartimento Scienze della Vita, Università di Trieste via L, Giorgeri, Italy Search for more papers by this author Liam P Keegan Liam P Keegan MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Western General Hospital, Edinburgh, UK Search for more papers by this author Giannino Del Sal Giannino Del Sal Laboratorio Nazionale CIB, Trieste, Italy Dipartimento Scienze della Vita, Università di Trieste via L, Giorgeri, Italy Search for more papers by this author Mary A O'Connell Corresponding Author Mary A O'Connell MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Western General Hospital, Edinburgh, UK Search for more papers by this author Roberto Marcucci Roberto Marcucci MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Western General Hospital, Edinburgh, UK Search for more papers by this author James Brindle James Brindle MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Western General Hospital, Edinburgh, UK Search for more papers by this author Simona Paro Simona Paro MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Western General Hospital, Edinburgh, UK Search for more papers by this author Angela Casadio Angela Casadio MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Western General Hospital, Edinburgh, UK Search for more papers by this author Sophie Hempel Sophie Hempel MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Western General Hospital, Edinburgh, UK Search for more papers by this author Nicholas Morrice Nicholas Morrice The Beatson Institute for Cancer Research, Glasgow, UK Search for more papers by this author Andrea Bisso Andrea Bisso Laboratorio Nazionale CIB, Trieste, Italy Dipartimento Scienze della Vita, Università di Trieste via L, Giorgeri, Italy Search for more papers by this author Liam P Keegan Liam P Keegan MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Western General Hospital, Edinburgh, UK Search for more papers by this author Giannino Del Sal Giannino Del Sal Laboratorio Nazionale CIB, Trieste, Italy Dipartimento Scienze della Vita, Università di Trieste via L, Giorgeri, Italy Search for more papers by this author Mary A O'Connell Corresponding Author Mary A O'Connell MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Western General Hospital, Edinburgh, UK Search for more papers by this author Author Information Roberto Marcucci1, James Brindle1, Simona Paro1, Angela Casadio1, Sophie Hempel1, Nicholas Morrice2, Andrea Bisso3,4, Liam P Keegan1, Giannino Del Sal3,4 and Mary A O'Connell 1 1MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Western General Hospital, Edinburgh, UK 2The Beatson Institute for Cancer Research, Glasgow, UK 3Laboratorio Nazionale CIB, Trieste, Italy 4Dipartimento Scienze della Vita, Università di Trieste via L, Giorgeri, Italy *Corresponding author. MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, UK. Tel.: +44 131 467 8417; Fax: +44 131 467 8456; E-mail: M.O′[email protected] The EMBO Journal (2011)30:4211-4222https://doi.org/10.1038/emboj.2011.303 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info ADAR2 catalyses the deamination of adenosine to inosine at the GluR2 Q/R site in the pre-mRNA encoding the critical subunit of AMPA receptors. Among ADAR2 substrates this is the vital one as editing at this position is indispensable for normal brain function. However, the regulation of ADAR2 post-translationally remains to be elucidated. We demonstrate that the phosphorylation-dependent prolyl-isomerase Pin1 interacts with ADAR2 and is a positive regulator required for the nuclear localization and stability of ADAR2. Pin1−/− mouse embryonic fibroblasts show mislocalization of ADAR2 in the cytoplasm and reduced editing at the GluR2 Q/R and R/G sites. The E3 ubiquitin ligase WWP2 plays a negative role by binding to ADAR2 and catalysing its ubiquitination and subsequent degradation. Therefore, ADAR2 protein levels and catalytic activity are coordinately regulated in a positive manner by Pin1 and negatively by WWP2 and this may have downstream effects on the function of GluR2. Pin1 and WWP2 also regulate the large subunit of RNA Pol II, so these proteins may also coordinately regulate other key cellular proteins. Introduction The AMPA class of glutamate-gated ion channel receptors (GluR) are impermeable to calcium if a GluR2 subunit is present in the tetrameric receptor (Hollmann et al, 1991; Verdoorn et al, 1991). This impermeability to calcium results from RNA editing of the GluR2 transcript. The enzyme that catalyses this RNA editing event is a member of the family of adenosine deaminases that act on RNA (ADARs). ADAR2 specifically deaminates an adenosine residue in a glutamine (Q) codon to an inosine that is read as guanosine by reverse transcriptase and the translational machinery. ADAR2 converts the glutamine (Q) codon to an arginine (R) codon with 100% efficiency at the GluR2 Q/R site changing a key residue in the ion channel pore and rendering AMPA receptors assembled with this subunit impermeable to calcium (Sommer et al, 1991). The editing event also regulates AMPA receptor assembly, slowing the passage of the GluR2 subunit through the ER thus ensuring correct receptor assembly (Greger et al, 2003). Failure of RNA editing at this site can lead to neuronal cell death due to the influx of calcium (Higuchi et al, 2000). A decrease in editing at this site has been reported in sporadic ALS motor neurons (Kawahara et al, 2004) and in hippocampal neurons following transient forebrain ischaemia in a rat model of stroke (Peng et al, 2006). Mice that are null mutants for ADAR2 are seizure-prone and die within 3 weeks after birth (Higuchi et al, 2000). Lethality in these Adar2−/− mice can be rescued by knocking-in the edited isoform of GluR2 (GluR2R). This experiment suggests that despite ADAR2 having other transcripts that it edits, the critical site is the Q/R site in GluR2 transcripts. These rescued mice have a normal phenotype, suggesting that the unedited GluR2 isoform does not have an essential biological function. For this deamination event to occur, ADAR2 must recognize and bind to double-stranded (ds)RNA that is formed at the editing site between the edited exon and the downstream intron (Higuchi et al, 1993). Identified transcripts edited specifically by ADAR2 are mostly expressed in the CNS even though the protein is also expressed in other tissues. RNA editing occurs before splicing and ADAR2 localizes to the nucleus. In some cells, ADAR2 accumulates within the nucleolus (Desterro et al, 2003; Sansam et al, 2003); however, this localization is dynamic. When transcripts that can be edited are overexpressed in these cells, ADAR2 relocalizes to the nucleoplasm (Desterro et al, 2003). Until now the only regulator found to influence ADAR2 expression is CREB, which can induce ADAR2 expression in hippocampal CA1 neurons in rat brain (Peng et al, 2006). In this study, we demonstrate that ADAR2 is dynamically regulated post-translationally by the phosphorylation-dependent peptidyl-prolyl cis/trans isomerase Pin1 (peptidyl-prolyl isomerase NIMA interacting protein 1). Pin1 binds to a phosphorylated serine or threonine residue preceding a proline residue and catalyses the cis/trans isomerization of the peptide bond (Lu et al, 1999). This conformational change can have a range of consequences on the function of target proteins, altering catalytic activity, stability or subcellular localization (for review see Lu and Zhou, 2007). Pin1 binds to the amino-terminus of ADAR2 in a phosphorylation-dependent manner. In the absence of Pin1, ADAR2 protein is more labile and is mislocalized to the cytoplasm, where it is unable to edit pre-mRNAs and there is a decrease in editing of the Q/R and R/G sites in endogenous GluR2 transcripts. Pin1 is therefore a positive regulator of ADAR2 editing activity. We also identify a negative regulator of ADAR2 activity, which is WWP2; a HECT (homologous to the E6-AP C terminus) E3 ubiquitin ligase (Pirozzi et al, 1997). WWP2 binds to a conserved PPxY motif in ADAR2 and this interaction results in ubiquitination and subsequent degradation of ADAR2. An increase in the expression of WWP2 results in a decrease in ADAR2 protein level. This report of the post-translational regulation of ADAR2 demonstrates how RNA editing activity is controlled by coordinate action of two regulators. Results Phosphorylation sites near the N-terminus of ADAR2 When human ADAR2 was purified to homogeneity from HeLa cells, enzymatic activity was very labile (O'Connell et al, 1997). However, recombinant human ADAR2 protein purified after overexpression in the yeast Pichia pastoris is active and stable. To determine if the protein is regulated by post-translational modification, we performed mass spectrometry on recombinant ADAR2 purified from P. pastoris and identified two phosphorylated serines near the amino-terminus, serine (S) 26 and S31 (Supplementary Figure S1). Phosphorylation at S26 has been independently verified (Dephoure et al, 2008). The amino-terminal region of ADAR2 is of interest since it has been shown to be important for dimerization of the protein and autoinhibition of catalytic activity (Gallo et al, 2003; Macbeth et al, 2004). ADAR2 interacts with Pin1 The phosphorylated residues near the N-terminus of ADAR2 are within potential recognition motifs (Ser/Thr-Pro) for the phosphorylation-dependent peptidyl-prolyl cis/trans isomerase Pin1, a well-conserved and extremely efficient enzyme for transducing post-translational modifications into conformational changes in key cellular proteins (Lu and Zhou, 2007). To determine whether Pin1 interacts with ADAR2, HEK293T cells were transiently transfected with a construct expressing ADAR2 bearing a FLAG epitope tag at the N-terminus and tetra-His tag at the C-terminus. After 24 h, the cells were harvested, whole cell protein extracts were immunoprecipitated with anti-FLAG monoclonal antibody and analysed by immunoblot detection of the immunoprecipitate with mouse anti-mitotic phosphoprotein monoclonal-2 (MPM-2) antibody (Davis et al, 1983), that recognizes the phosphorylated Pin1 motif (Ser/Thr-Pro) in proteins. As shown in Figure 1A, α-MPM-2 recognizes the FLAG-tagged ADAR2 protein. We mutated T32, as this is the residue that precedes the proline so it may be important for Pin1 binding. When the immunoprecipitation was repeated with alanine (A) substitutions for S26, S26/31 or T32 at the amino-terminus, the antibody recognized ADAR2 less efficiently and loss of binding of the MPM-2 antibody was particularly evident with the triple mutant ADAR2S26A/S31A/T32A (Figure 1A), suggesting that the amino-terminus of ADAR2 harbours phosphorylated S/T-P sites at the amino-terminus that are likely to bind Pin1. Figure 1.The amino-terminus of ADAR2 harbours a Pin1-binding site. (A) The anti-MPM-2 antibody recognizes potential Pin1 sites in ADAR2 purified after overexpression in P. pastoris. Immunoblot analysis with anti-MPM-2 antibody of anti-FLAG immunoprecipitates from lysates of HEK293T cells transfected with FLAG-tagged hADAR2, ADAR2S26A/S31A, ADAR2S26A/S31A/T32A, ADAR2S26A, ADAR2T32A or pcD3. The minor band in the lane with pcD3 is contamination from the neighbouring lane. ADAR input visualized with anti-FLAG antibody, lower panel. (B) Purified ADAR2 binds in vitro to Pin1 immobilized on beads. (Upper left panel) Immunoblot analysis with anti-FLAG antibody of the binding of FLAG-tagged ADAR2, ADAR2S26A, ADAR2S26A/S31A/T32A, ADAR2Δ4–72 bound to GST–Pin1 or GST on glutathione beads. (Lower panel) GST input visualized with anti-GST antibody. (Right panel) Purified ADAR proteins stained with Coomassie. (C) Binding of purified ADAR2 to Pin1 depends on phosphorylation of Pin1 sites on ADAR2. λ phosphatase treatment of lysate from HEK293T cells transfected with ADAR2 for 0 (−), 2 h (+), 3 h (++) prior to incubation with GST–Pin1. Immunoblot analysis of ADAR2 with anti-FLAG antibody. Middle and lower panels are input loading controls. (D) Pin1 binds to ADAR2 in HEK293T cells. Coimmunoprecipitation of ADAR2 and Pin1 performed with anti-FLAG antibody on HEK293T cell lysate cotransfected with HA–Pin1 and either FLAG-tagged ADAR2, ADAR2Δ4–72, ADAR2RRM1–2, ADAR2S26A/S31A/T32A or pcD3. HA–Pin1 was detected with anti-HA antibody. Asterisks represent IgG light chain. (Lower panel) Immunoblot of input proteins with anti-FLAG antibody. (E) Endogenous Pin1 detected with anti-Pin1 antibody after immunoprecipitation with anti-FLAG antibody from cell lysates of HEK293T cells transfected with FLAG-tagged ADAR2, ADAR2S26A/S31A/T32A, ADAR2Δ4–72, ADAR2RRM1–2 or pcD3. (Lower panel) Immunoblot of input proteins detected with anti-FLAG antibody. Asterisks represent IgG light chain. Download figure Download PowerPoint The ability of ADAR2 to bind to Pin1 was next evaluated by in vitro binding assays with GST–Pin1 and recombinant ADAR2 purified from P. pastoris. As shown in Figure 1B (left panel), ADAR2 binds strongly to GST–Pin1 whereas ADAR2 did not interact with the GST beads alone. To map the interaction between Pin1 and ADAR2, ADAR2S26A, ADAR2S26A/S31A/T32A or an N-terminal deletion of ADAR2 from amino acid to 4–72 (Wong et al, 2003) were purified from P. pastoris (Figure 1B, right panel) and similarly tested for interaction with GST–Pin1. The interaction of GST–Pin1 with ADARS26A was slightly weaker than with wild-type ADAR2 and interaction was drastically decreased with the triple mutant ADAR2S26A/S31A/T32A, and totally absent with ADAR2Δ4–72 (Figure 1B, left panel). To determine if the interaction with Pin1 depends on ADAR2 phosphorylation, a transient transfection of ADAR2 into HEK293T cells was performed and the lysate was treated with λ phosphatase followed by a pull-down assay with GST–Pin1 beads. The interaction between ADAR2 and Pin1 was observed and this was abolished with a longer λ phosphatase treatment (Figure 1C). As these experiments were performed in vitro, we then analysed the Pin1 ADAR2 interaction in HEK293T cells by transiently cotransfecting with constructs expressing FLAG-tagged ADAR2 and HA-tagged Pin1. The cells were harvested after 24 h and an immunoprecipitation of the lysate was performed with anti-FLAG monoclonal antibody and the precipitate was detected on an immunoblot with an anti-HA antibody (Figure 1D). Only the wild-type ADAR2 interacted with HA-tagged Pin1. Neither the triple alanine mutant nor ADAR2Δ4–72 interacted with Pin1. In addition, ADAR2 that has mutations in both RNA-binding domains and cannot bind to dsRNA (ADAR2RRM1–2) (Valente and Nishikura, 2007) does not interact with Pin1. Therefore, ADAR2 has to bind to RNA before it can interact with Pin1. In the in vitro binding assays with GST–Pin1 and recombinant ADAR2 purified from P. pastoris (Figure 1B), ADAR2 appears to interact with GST–Pin1 in the absence of dsRNA. However in our experience, it is difficult to eliminate all the dsRNA present in the purified protein fraction from yeast (Gallo et al, 2003) so therefore we presume that this in vitro reaction is also mediated by dsRNA. Similar results were obtained when HEK293T cells were transiently transfected with FLAG–ADAR2 followed by coimmunoprecipitation with endogenous Pin1 (Figure 1E). These results demonstrate that Pin1 binds to ADAR2 in a phosphorylation-dependent manner and that this interaction occurs at the amino-terminal of ADAR2 after it has bound to RNA. Pin1 expression is required for optimal editing at the GluR2 Q/R site Since ADAR2 converts a glutamine (Q) codon to an arginine (R) codon with 100% efficiency at the GluR2 Q/R site in neurons, the important question is whether the interaction between Pin1 and ADAR2 affects editing activity at the critical GluR2 Q/R site. To address this point, we analysed editing of the GluR2 Q/R site in HeLa cells. To increase the level of editing at the Q/R site by ADAR2 in HeLa cells, we transiently cotransfected a plasmid encoding ADAR2 with the GluR2 B13 minigene. The level of editing rose to 100%. We then cotransfected an siRNA specific for Pin1 and editing fell to 53% (Figure 2A). Figure 2.Pin1 is required for efficient editing at the GluR2 Q/R site. (A) DNA sequence chromatograph of the RT–PCR product of the region encompassing the Q/R site (arrow) encoded by the GluR2 B13 minigene transiently cotransfected with ADAR2 (2 μg) in HeLa cells, editing is 100% (left chromatograph). Editing of the Q/R site drops to 53% when an siRNA specific for Pin1 is cotransfected together with plasmids encoding both ADAR2 and the GluR2 B13 minigene (middle chromatograph). Editing is 100% at the GluR2 Q/R site when a control siRNA is cotransfected (right chromatograph). Immunoblot analysis of cell lysate from HeLa cells with either anti-Pin1 or anti-tubulin antibodies (right panel). (B) (Left panel) Sequencing chromatogram of editing by endogenous ADAR2 at the Q/R site of RT–PCR product pools from the GluR2 B13 minigene transcript that has been transiently transfected into HeLa cells. Arrows indicate Q/R editing site in all panels. Immunoblot analysis with anti-Pin1 antibody of HeLa cell extracts that have been cotransfected with GFP in the presence of either Pin1-specific siRNA, no siRNA or HA–Pin1 construct (0.5 μg). Proteins are detected with anti-Pin1 antibody and anti-GFP antibody as a loading control (right panel). (C) Chromatograph of editing of endogenous GluR2 transcript at the Q/R site in neuroblastoma SH-SY5Y cells transfected with ADAR2 (2 μg). Editing is 100% at the Q/R site (left chromatograph). A decrease in editing is observed when an siRNA specific for Pin1 was cotransfected (middle chromatograph). A control siRNA does not affect editing when transfected (right chromatograph). Arrows indicate the Q/R site. Cell lysates of SH-SY5Y cells were analysed by immunoblot with either anti-Pin1 or anti-tubulin (right panel). (D) Chromatograph of editing at the Q/R site of GluR2 B13 minigene transcript in Pin1−/− MEF cells transfected with ADAR2 (2 μg). Editing increased to 100% when Pin1−/− MEF cells were cotransfected with either 0.5 or 1 μg of a construct expressing Pin1. An arrow marks the Q/R editing site in all the chromatographs. Download figure Download PowerPoint We also analysed editing at the Q/R site in the GluR2 B13 minigene transcript (Higuchi et al, 1993) by endogenous ADAR2 and observed it was 60% efficient in this cell line; however, when siRNA specific for Pin1 was cotransfected, the level of editing fell to 46% (Figure 2B). Editing was restored to 74% when Pin1 was overexpressed in HeLa cells. We also examined the effect of reducing Pin1 expression on editing of endogenously expressed GluR2 transcript in a neuroblastoma cell line, SH-SY5Y (Figure 2C). In this cell line, ADAR2 was cotransfected with either a Pin1-specific siRNA or a control siRNA and editing of the endogenous GluR2 transcript was analysed. Again the level of editing dropped from 100% to ∼60% at the Q/R site when there was a reduction in Pin1 expression. We also analysed editing at the R/G site in the GluR2 transcript and found it was 69% but dropped to 45% when siRNA specific for Pin1 was cotransfected whereas editing was 73% when a control siRNA was cotransfected. The reduction in Pin1 expression for this experiment is shown in Supplementary Figure S2. To examine the effect of complete Pin1 elimination, we cotransfected constructs expressing the GluR2 B13 minigene and ADAR2 into an immortalized mouse fibroblast cell line derived from Pin1−/− mice (Figure 2D) (Fujimori et al, 1999). The editing activity at the Q/R site was ∼50% and increased to 100% when a construct expressing Pin1 was reintroduced in these cells. All these experiments strongly suggest that ADAR2 requires Pin1 for maximal editing of the critical Q/Rand R/G sites in GluR2 transcripts. Pin1 has a role in the nuclear localization of ADAR2 Pin1 has many diverse activities within the cell and it can alter the cellular localization of its substrate, as occurs with β-catenin (Ryo et al, 2001). Although ADAR2 has been documented as nuclear, recent evidence demonstrated that in human motor neurons in spinal cord sections, ADAR2 is both nuclear and cytoplasmic (Aizawa et al, 2010). Interestingly, a deletion of the amino-terminal residues 4–72 renders ADAR2 cytoplasmic (Wong et al, 2003) and it has also been demonstrated that this region is required for nuclear localization as it contains a non-canonical NLS within the first 64 amino acids (Desterro et al, 2003). As this deletion removes the Pin1-binding site, we wondered if preventing Pin1 binding also leads to mislocalization of ADAR2. To elucidate this we transiently transfected GFP-tagged ADAR2 into Pin1+/+ and Pin1−/− MEF cells and performed immunofluorescence detection of ADAR2 (Figure 3A). In the absence of Pin1, wild-type ADAR2 is mislocalized in the cytoplasm (Figure 3A, lower panel). Mislocalization of ADAR2 is confirmed when nuclear and cytoplasmic fractionation is performed on Pin1−/− MEF cells transiently transfected with FLAG-tagged ADAR2 (Figure 3D). When Pin1 was reintroduced into these cells, the level of ADAR2 in the cytoplasm was significantly reduced (Figure 3B and D). This effect of Pin1 on the localization of ADAR2 requires Pin1 prolyl-isomerase enzymatic activity as a Pin1S67E mutant that is catalytically inactive was unable to restore ADAR2 localization to the nucleus (Figure 3C). GFP–ADAR2 is localized to the nucleus when Pin1 is present; however, cytoplasmic localization of GFP–ADAR2 increases following cotransfection with catalytically inactive Pin1. Increased FLAG–ADAR2 is also observed in the cytoplasmic fraction of Pin1−/− MEF cells transiently transfected with FLAG-tagged ADAR2 (Figure 3D). Figure 3.Pin1 is required for nuclear localization of ADAR2. (A) ADAR2 is mislocalized from the nucleus to the cytoplasm in Pin1−/− MEF cells. GFP–ADAR2 immunofluorescence in Pin1+/+ and Pin1−/− MEF cells cotransfected with GluR2 B13 minigene and GFP–ADAR2. DAPI staining of nuclei (i, iv), GFP fluorescence of cell (ii, v) and merged (iii, vi). (B) Nuclear localization of ADAR2 is restored in Pin1−/− MEF cells by transfection of HA–Pin1. GFP–ADAR2 (green) direct and HA–Pin1 (red) indirect immunofluorescence detection in Pin1−/− MEF cells cotransfected with GluR2 B13 minigene and (ii) GFP–ADAR2 (green) and (iii) HA–Pin (red). (i) DAPI staining of nuclei. (iv) Merge of all three images. (C) Nuclear localization of ADAR2 depends on catalytic activity of Pin1. GFP–ADAR2 (green) and HA–Pin1S67E (red) in Pin1−/− MEF cells cotransfected with GluR2 B13 minigene and (ii) GFP–ADAR2 (green) and (iii) HA–Pin1S67E (red). (i) DAPI staining of nuclei. (iv) Merge of all three images. All photographs were taken at the same exposure. Scale bar, 10 μm. (D) Nucleo-cytoplasmic fractionation. Immunoblot analysis with anti-FLAG antibody of nuclear and cytoplasmic fractions of Pin1−/− MEF cells transfected with FLAG–ADAR2 (lanes 1 and 2). Pin1 was cotransfected with FLAG–ADAR2 in Pin1−/− MEF cells (lanes 3 and 4). HA–Pin1S67E was cotransfected in Pin1−/− MEF cells (lanes 5 and 6). (Lower panel) Immunoblot of fractionated Pin1−/− MEF cells with tubulin as a marker for cytoplasmic fraction and HP1α for nuclear fraction. Download figure Download PowerPoint As Pin1 recognizes a phosphorylated serine or threonine preceding a proline, we replaced the phosphorylated amino acids as well as the prolines with alanine to determine if all were required for nuclear localization. As expected, the triple mutant FLAG–ADAR2S26/S31A/T32A was present in the cytoplasm (Supplementary Figure S3) and this appears slightly different to ADAR2Δ4–72 that is more localized around the nuclear periphery (Supplementary Figure S4). When the proline mutants were generated; FLAG–ADAR2P27A and FLAG–ADAR2P33A, were transiently transfected into HeLa cells together with HA–Pin1 (Figure 4), FLAG–ADAR2P33A was present in the cytoplasm as detected by immunofluorescence as well as by nuclear and cytoplasmic fractionation (Figure 4B and C) whereas FLAG–ADAR2P27A is nuclear. This implies that the second proline is the critical one, thus the phosphorylation of T32 may be the critical site for Pin1 binding and P33 for isomerization. Notably, this is also the most conserved Pin1 site in vertebrate ADAR2 sequences (Supplementary Figure S1). Figure 4.Proline 33 of ADAR2 is required for the Pin1 effect on ADAR2 nuclear localization. (A) Normal localization of ADAR2 and Pin1. Immunofluorescence of HeLa cells cotransfected with GluR2 B13 minigene, FLAG–ADAR2 and HA–Pin1 stained with (i) DAPI, (ii) anti-HA–Pin1 (green), (iii) anti-FLAG–ADAR2 (red), (iv) Merge of DAP1 and FLAG and (v) merge of all three images. (B) Nuclear localization of ADAR2 depends on Proline 33. Immunofluorescence of HeLa cells cotransfected with GluR2 B13 minigene, FLAG–ADAR2P33A and HA–Pin1 stained with (i) DAPI, (ii) anti-HA–Pin1 (green), (iii) anti-FLAG–ADAR2P33A (red), (iv) Merge of DAP1 and FLAG and (v) merge of all three images. All photographs were taken at the same exposure. Scale bar, 10 μm. (C) Nucleo-cytoplasmic fractionation of wild-type and ADARP27A and ADAR2P33A mutants. Immunoblot analysis with anti-FLAG antibody of nuclear and cytoplasmic fractions of HeLa cells transfected with FLAG–ADAR2 (lanes 1 and 2), ADARP27A (lanes 3 and 4), ADAR2P33A (lanes 5 and 6). (Lower panels) Immunoblot of fractionated HeLa cells with tubulin as a cytoplasmic marker and HP1α as a nuclear marker. Download figure Download PowerPoint Pin1 stabilizes ADAR2 We wanted to elucidate if Pin1 had other effects on ADAR2. As Pin1 has been shown to influence the stability of proteins such as β-catenin (Ryo et al, 2001), NF-κB (Ryo et al, 2003) and p53 (Zacchi et al, 2002; Zheng et al, 2002), we wondered if Pin1 also influences the stability of ADAR2. The level of Pin1 was reduced in HeLa cells by transfecting either pSuper Pin1 siRNA (Rustighi et al, 2009) or a control pSuper LacZ siRNA so that the level of endogenous ADAR2 could be analysed (Figure 5A). A similar experiment was performed in the neuroblastoma cell line SH-SY5Y (Figure 5C). Twenty-four hours after transfection, cycloheximide was added to prevent further protein synthesis, a time course from 0 to 8 h was performed to chase the decay of ADAR2 protein and the samples were analysed by immunoblot analysis to determine ADAR2 levels. In both cell lines, the protein level of ADAR2 decreased when cycloheximide was added; however, the decrease was more dramatic when Pin1 expression was reduced (Figure 5A–C). The stability of the triple mutant ADAR2S26A/S31A/T32A was also analysed after cycloheximide treatment (Figure 5D). As predicted this mutant protein was unstable as it could no longer interact with Pin1. Therefore, Pin1 affects the stability of ADAR2. Figure 5.Pin1 contributes to stability of ADAR2 protein. (A) Knockdown of Pin1 in HeLa cells destabilizes ADAR2 in a cycloheximide time course. HeLa cells were transfected with either pSuper LacZ or pSuper Pin1. Cycloheximide (50 μg/μl) was added to both and a time course from 0 to 8 h was performed. Cell lysates were analysed by immunoblot and the antibodies used were anti-ADAR2 (top panel), anti-tubulin as a loading control (middle panel) and Pin1 (bottom panel). (B) Quantification of (A). (C) Pin1 knockdown destabilization of FLAG–ADAR2 in SH-SY5Y neuroblastoma cells. SH-SY5Y cells were cotransfected with FLAG-tagged ADAR2 and a control siRNA or Pin1-specific siRNA. Cycloheximide (50 μg/μl) was added to both and a time course from 0 to 8 h was performed. Cell lysates were analysed by immunoblot and the antibodies used were anti-FLAG (top panel), anti-Pin1 (middle panel) and GFP as loading control (bottom panel). (D) ADAR2 mutant in the Pin1-binding site is less stable. SH-SY5Y neuroblastoma cells were transfected with FLAG-tagged ADAR2S26A/S31A/T32A and cycloheximide (50 μg/μl) was added and a time course was performed from 0 to 8 h. Cell lysates were analysed by immunoblot and the antibodies used were anti-FLAG (top
Referência(s)