FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2988339731> ?p ?o ?g. }

• W2988339731 abstract "Full text Figures and data Side by side Abstract eLife digest Introduction Results Discussion Materials and methods References Decision letter Author response Article and author information Metrics Abstract Synapses and nuclei are connected by bidirectional communication mechanisms that enable information transfer encoded by macromolecules. Here, we identified RNF10 as a novel synaptonuclear protein messenger. RNF10 is activated by calcium signals at the postsynaptic compartment and elicits discrete changes at the transcriptional level. RNF10 is enriched at the excitatory synapse where it associates with the GluN2A subunit of NMDA receptors (NMDARs). Activation of synaptic GluN2A-containing NMDARs and induction of long term potentiation (LTP) lead to the translocation of RNF10 from dendritic segments and dendritic spines to the nucleus. In particular, we provide evidence for importin-dependent long-distance transport from synapto-dendritic compartments to the nucleus. Notably, RNF10 silencing prevents the maintenance of LTP as well as LTP-dependent structural modifications of dendritic spines. https://doi.org/10.7554/eLife.12430.001 eLife digest Brain activity depends on the communication between neurons. This process takes place at the junctions between neurons, which are known as synapses, and typically involves one of the cells releasing a chemical messenger that binds to receptors on the other cell. The binding triggers a cascade of events inside the recipient cell, including the production of new receptors and their insertion into the cell membrane. These changes strengthen the synapse and are thought to be one of the ways in which the brain establishes and maintains memories. However, in order to induce these changes at the synapse, neurons must be able to activate the genes that encode their component parts. These genes are present inside the cell nucleus, which is located some distance away from the synapse. Studies have shown that signals can be sent from the nucleus to the synapse and vice versa, enabling the two parts of the cell to exchange information. Synapses that communicate using a chemical called glutamate have been particularly well studied; but it still remains unclear how the activation of receptors at these “glutamatergic synapses” is linked to activation of genes inside the nucleus at the molecular level. Dinamarca, Guzzetti et al. have now discovered that this process at glutamatergic synapses involves the movement of a protein messenger to the nucleus. Specifically, activation at synapses of a particularly common subtype of receptor, called NMDA, causes a protein called Ring Finger protein 10 (or RNF10 for short) to move from the synapse to the nucleus. To leave the synapse, RNF10 first has to bind to proteins called importins, which transport RNF10 into the nucleus. Once inside the nucleus, RNF10 binds to another protein that interacts with the DNA to start the production of new synaptic proteins. Further work is required to identify the molecular mechanisms that trigger RNF10 to leave the synapse. In addition, future studies should evaluate the levels and activity of RNF10 in brain disorders in which synapses are known to function abnormally. https://doi.org/10.7554/eLife.12430.002 Introduction Understanding how local synaptic events are translated into changes in gene expression is a crucial question in neuroscience. Synapses and nuclei are efficiently connected by bidirectional communication routes that enable transfer of information (Fainzilber et al., 2011; Panayotis et al., 2015) and regulate the transcription of genes associated with long-term structural changes of neuronal excitability (Karpova et al., 2012). NMDAR activation plays a key role in this regard (Hardingham and Bading, 2010). NMDARs are heteromeric ionotropic channels that are essential for excitatory neurotransmission (Paoletti et al., 2013; Sanz-Clemente et al., 2013). NMDARs differ in their subunit composition and, in the forebrain, they can be either di- or tri-heteromeric tetramers consisting of two GluN1 and either one or two GluN2A or GluN2B subunits (Paoletti et al., 2013). GluN2A-containing NMDARs are mainly localized at the postsynaptic membrane, at the core of the postsynaptic density (PSD), while the GluN2B-containing NMDARs are also prominently present at extrasynaptic sites (Hardingham and Bading, 2010). In addition, numerous reports have indicated that GluN2A- and GluN2B-containing NMDARs may play different roles in the modulation of synaptic plasticity and in central nervous system (CNS) disorders (Paoletti et al., 2013). These differences in receptor function are probably linked to the cytoplasmic C-tail of the two subunits, which are fairly different and contain specific motifs that bind to PSD-associated scaffolding proteins and to proteins involved in the downstream signal transduction of receptor activation (Sanz-Clemente et al., 2013; Sun et al., 2016). It is generally accepted that rises in synaptic, somatic and nuclear Ca2+ rapidly regulate gene expression by Ca2+-sensing mechanisms (Bading, 2013). Recent work showed that following a fast genomic response to sustained rises in Ca2+, a considerably slower process that depends on the nuclear import of proteins released from synapses couples local synaptic events to more specific gene expression programs (Jordan and Kreutz, 2009; Ch'ng and Martin, 2011; Karpova et al., 2012). The classical active nuclear import pathway involves the binding of importin α isoforms to nuclear localization signal (NLS) bearing cargo proteins. Interestingly, neuronal importins are present at synapses in association with NMDARs and the PSD (Dieterich et al., 2008; Jeffrey et al., 2009) and they are able to translocate to the nucleus in response to NMDAR activation (Thompson et al., 2004; Dieterich et al., 2008). Proteomic studies have identified many proteins in purified synaptosomes that contain a NLS domain, and in the past 10 years, an impressive number of potential synaptonuclear messenger proteins have been characterized (Ch'ng et al., 2012; Karpova et al., 2013; Kaushik et al., 2014). Many of these proteins contain a bona fide NLS and, in some cases, their binding with importin α was shown to be essential for long-distance transport as importin α can serve as an adaptor for nuclear trafficking of cargo in association with a dynein motor (Karpova et al., 2012; Ch'ng et al., 2012). In recent studies, it was shown that Jacob, a Caldendrin-binding protein abundantly expressed in limbic brain and cortex (Dieterich et al., 2008), encodes and transduces the synaptic and extrasynaptic origin of GluN2B-containing NMDAR signals to the nucleus and elicits the divergent transcriptional responses after activation of these receptors (Dieterich et al., 2008; Karpova et al., 2013). These findings point to the fascinating possibility that specific NMDAR signals are encoded at synaptic sites and decoded in the nucleus by long-distance trafficking of protein messengers. Here, we identified Ring Finger protein 10 (RNF10; Seki et al., 2000) as a novel synaptonuclear protein messenger, localized at the PSD and specifically associated to the cytoplasmic tail of GluN2A but not GluN2B subunit of NMDARs. After stimulation of synaptic GluN2A-containing NMDARs, RNF10 binds to importin α1 for nuclear long-distance transport. Notably, we show that long-term potentiation (LTP) induces nuclear translocation of RNF10, its interaction with the transcription factor Meox2 and the modulation of gene expression. Most interestingly, RNF10-regulated gene expression appears to feed back to synaptic function. Results RNF10 is a neuronal synaptic protein In order to identify new binding partners for GluN2A, we performed a yeast two-hybrid screening using the C-terminal domain (aa 839–1461, without the aa 1462–1464 PDZ-binding sequence) as bait. We obtained a number of positive clones including RNF10 (Seki et al., 2000; Hoshikawa et al., 2008). In Schwann cells, RNF10 has a function in the transcriptional regulation of myelin formation (Hoshikawa et al., 2008). However, very little is known about the neuronal function of this protein (Seki et al., 2000; Lin et al., 2005; Hoshikawa et al., 2008; Malik et al., 2013). Using specific glial (GFAP) and neuronal (MAP2) markers, we confirmed that RNF10 is expressed in both glia (Figure 1A) and neurons (Figure 1B). Interestingly, in neurons RNF10 displayed a nuclear and somatodendritic distribution (Figure 1B). Moreover, transfected GFP-RNF10 displayed a prominent co-localization with PSD-95 and GluN2A at the dendritic spines of hippocampal neurons (Figure 1C). Similarly, analysis of endogenous RNF10 in DIV14 primary hippocampal neurons showed clustered RNF10 immunolabeling along dendrites and most puncta co-localized with GluN2A (Figure 1D, left panels) and PSD-95 (Figure 1D, right panels). PSDs from the rat hippocampus were purified to confirm the subcellular distribution of RNF10 by a biochemical approach. Subcellular fractionation demonstrated that RNF10 is associated with synaptic fractions and that it is prominently present in PSD fractions (Figure 1E). Finally, immunofluorescence analysis of the CA1 region of the adult rat hippocampus revealed the presence of an intense signal for RNF10 in the soma and nuclei together with a punctate staining along MAP2-positive dendrites (Figure 1F). Figure 1 Download asset Open asset RNF10 subcellular distribution in neurons. (A,B) Mixed primary hippocampal cultures (DIV14) immunolabeled with antibodies for RNF10 (red), glial marker GFAP (A; green) or the neuronal marker MAP2 (B; green), and Dapi (blue) to stain the nucleus; scale bar: 20 μm. (C) Dendrite of hippocampal neuron transfected with GFP-RNF10 (DIV7) and immunolabeled at DIV14 for GFP (green), GluN2A (red) and PSD-95 (blue); scale bar: 3 μm. (D) High-magnification confocal images of neuronal dendrites (DIV14) immunolabeled for endogenous RNF10 (green) and GluN2A (red; left panels) or PSD-95 (red; right panels); scale bar: 4 μm. (E) RNF10 and markers of the presynaptic (synaptophysin) and postsynaptic compartment (PSD-95, GluN2A) were analyzed by WB in various subcellular compartments (H: Homogenate fraction; S1: supernatant 1; P1: nuclear fraction; S2: cytosolic fraction 2; P2: crude membrane fraction 2; Syn: synaptosomal fraction; PSD1: Triton Insoluble postsynaptic fraction; PSD2: postsynaptic density fraction). (F) Representative confocal images of adult rat hippocampal CA1 pyramidal layer sections showing immunohistochemical labeling for RNF10 (red), MAP2 (green), and Dapi (blue); scale bar: 40 μm. (G) Confocal images of dendrites from hippocampal neurons (DIV14) transfected at DIV7 with shGluN2A or scramble vector and immunolabeled for GluN2A; scale bar: 4 μm. The histogram shows the quantification of GluN2A integrated density in dendrites (n=7, **p=0.0069 scramble vs shGluN2A; unpaired Student’s t-test). (H) GluN2A silencing induces a reduction of RNF10 enrichment at the glutamatergic synapse. Confocal images of primary hippocampal neurons transfected with pGFP-V-RS-scramble (left panels) or with pGFP-V-RS-shGluN2A (right panels) plasmids and immunolabeled (DIV14) for RNF10 (green) and PSD-95 (red); scale bar: 4 μm. The histogram shows the quantification of RNF10 co-localization with PSD-95-positive puncta (n=30, ***p<0.001; unpaired Student’s t-test). (I) Confocal images of dendrites from hippocampal neurons (DIV14) transfected at DIV7 with shGluN2A or scramble vector and immunolabeled for surface GluN2A (blue) and RNF10 (red); scale bar: 4 μm. https://doi.org/10.7554/eLife.12430.003 Interestingly, GluN2A silencing in primary hippocampal neurons (Figure 1G) induced a significant decrease in RNF10 synaptic levels, as indicated by the reduction of RNF10 co-localization with PSD-95 (Figure 1H). Notably, the remaining dendritic RNF10 in shGluN2A neurons co-localized with the surface GluN2A pool unaffected by the knock down (Figure 1I). However, no modification of RNF10 nuclear level was observed following GluN2A silencing (data not shown; n=30; p=0.5491; shGluN2A vs scramble; unpaired Student’s t-test). RNF10 interacts with the GluN2A subunit of NMDARs Different experimental approaches were used to substantiate the yeast two-hybrid data and to confirm the interaction between RNF10 and GluN2A. Co-immunoprecipitation (co-i.p.) studies performed from hippocampal P2 crude membrane fractions indicated a specific interaction of RNF10 with GluN2A but not with GluN2B subunit of the NMDARs (Figure 2A). No signal for GluN2A or GluN2B was obtained by using anti-synaptophysin as an irrelevant antibody or in the absence of antibody in the co-i.p. assay (Figure 2A). To further validate these findings we performed similar experiments with the GluN2B-associated synapse-to-nucleus messenger Jacob (Dieterich et al., 2008; Karpova et al., 2013). Indeed, the affinity-purified pan-Jacob antibody preferentially co-i.p. GluN2B from rat brain homogenate. Only a very faint band of GluN2A was detected in the complex with Jacob (Figure 2B), which might potentially represent synaptic tri-heteromeric GluN1/GluN2A/GluN2B NMDARs. Figure 2 Download asset Open asset RNF10 interaction with GluN2A-containing NMDARs. (A) Co-immunoprecipitation (co-i.p.) assay performed in P2 crude membrane fractions by using antibodies against PSD-95, RNF10, synaptophysin (Syn) and GluN2A. WB analysis shows the levels of GluN2A (left panel) and GluN2B (right panel) in the co-immunoprecipitated material. No ab lane: control lane in absence of antibodies during the co-i.p. assay. (B) Jacob is a part of the GluN2B receptor complex. Affinity purified pan-Jacob antibodies co-immunoprecipitate GluN2B. (C) Co-i.p. assay performed by using an anti-RNF10 antibody from COS-7 cell extracts transfected with HA-GluN1 and GFP-GluN2A or GFP-GluN2B. WB analysis was performed by using anti-GFP and anti-RNF10 antibodies. No ab lane: control lane in absence of antibodies during the co-i.p. assay. (D) COS-7 cells expressing RNF10 were transfected with HA-GluN1 and GFP-GluN2A or GFP-GluN2B constructs and immunolabeled for GFP (green), GluN1 (blue), Dapi (cyan) and endogenous RNF10 (red); scale bar: 10 μm. (E) In situ detection of proximity between RNF10 and GluN2A (red) along MAP2 (green; left panels) or GFP-positive (green; right panels) dendrites. In control experiments (-), primary hippocampal neurons were labeled with only RNF10 primary antibody and thus only unspecific PLA signals are generated; scale bars: 5 μm (MAP2) and 3 μm (GFP). (F, G) COS-7 cells expressing RNF10 were transfected with GFP-GluN2A (right panels) or GFP-GluN2B (left panels) constructs and immunolabeled for GFP (green), Dapi (cyan) and endogenous RNF10 (red); scale bar: 10 μm. The histogram shows the quantification of RNF10 integrated density (i.d.) expressed as cytoplasm/nucleus ratio (n=10; ***p<0.001; one-way ANOVA, followed by Bonferroni post-hoc test). https://doi.org/10.7554/eLife.12430.004 In addition, cell lysates from COS-7 cells transfected with HA-GluN1 and GFP-GluN2A or GFP-GluN2B constructs were immunoprecipitated with anti-RNF10 and immunoblotted for GFP. The results confirmed that GluN2A interacts with RNF10, whereas GluN2B failed to associate with RNF10 (Figure 2C). To corroborate these results, we transfected HA-GluN1 and either GFP-GluN2A or GFP-GluN2B constructs in heterologous cells (COS-7) endogenously expressing RNF10. Immunofluorescence studies revealed the presence of RNF10 aggregates with a high co-localization degree with GluN2A but not with GluN2B (Figure 2D; 73.8% ± 2.9% GluN2A/RNF10 vs 31.3.% ± 1.7% GluN2B/RNF10; p<0.001, n=15; unpaired Student’s t-test). Finally, RNF10 clustering with GluN2A at synapses was confirmed by proximity ligation assay (PLA). As shown in Figure 2E, a large number of PLA signals were detected when the two antibodies RNF10 and GluN2A were used indicating these two proteins are in close proximity (<40 nm) to each other. As expected, PLA signals were distributed along MAP2-positive dendrites (Figure 2E, left panels) and at the top of GFP-positive dendritic spines (Figure 2E, right panels). For control experiments, only RNF10 primary antibody was used and no PLA signal was generated (Figure 2E). Notably, transfection of COS-7 cells with GluN2A but not GluN2B induces a highly significant redistribution of endogenous RNF10 from the nucleus to the cytoplasm, thus suggesting that GluN2A traps RNF10 outside the nucleus (Figure 2F–G). Overall, these results demonstrate that RNF10 is a component of the excitatory PSD and that it is specifically associated with GluN2A-containing NMDARs. The RNF10 N-terminus binds to the juxta-membrane region of GluN2A We next mapped the binding interface of the interaction between RNF10 and GluN2A using truncation constructs in pull-down and co-i.p. experiments. The pull-down assay performed from adult rat brain tissue using glutathione S-transferase (GST)-GluN2A C-terminal domain fusion proteins (Figure 3A) revealed that RNF10 interacts with the GluN2A cytoplasmic tail (CT; aa 839–1464) but failed to bind the distal part of the GluN2A C-terminus (aa 1049–1464) (Figure 3B, upper panel). In addition, GluN2A(839–991) failed to interact with RNF10 (Figure 3B, lower panel), thus indicating GluN2A(991–1049) is needed for the formation of the GluN2A/RNF10 complex. Figure 3 Download asset Open asset RNF10 N-terminal domain interacts with the juxtamembrane region of GluN2A C-tail. (A) Scheme showing GST-GluN2A fusion proteins used in the pull-down assay. (B) GST and GST-GluN2A fusion proteins were incubated in a pull-down assay with rat hippocampal extracts. WB analysis was performed with RNF10 antibody. (C) Confocal images of COS-7 cells transfected with GFP-GluN2A (1–1049) and immunolabeled for GFP (green), Dapi (cyan) and RNF10 (red); scale bar: 10 μm. (D) Co-i.p. assay performed from lysates of COS-7 cells transfected with GFP-GluN2A or GFP-GluN2A(1–1049). WB analysis was performed by using a GFP antibody (JL-8). (E) GST and GST-RNF10 full-length (FL) fusion proteins were incubated in a pull-down assay with rat hippocampal extracts in presence or absence of calcium (2 mM). WB analysis was performed with GluN2A antibody. (F) GST and GST-GluN2A(839–1464) fusion proteins were incubated in a pull-down assay with rat hippocampal extracts in presence or absence of calcium (2 mM). WB analysis was performed with RNF10 and CaM antibodies. (G) GST and GST-GluN2A(839–1049) fusion proteins were incubated in a pull-down assay with rat hippocampal extracts with CaM (0.1 μM) in the presence or absence of calcium (2 mM). WB analysis was performed with RNF10 and CaM antibodies. (H) Co-i.p. assay performed from lysates of COS-7 cells transfected with GFP, GFP-GluN2A or GFP-GluN2B in the presence or absence of calcium (2 mM)/CaM (0.1 μM). WB analysis was performed by using RNF10 and CaM antibodies. (I) Co-i.p assay performed by using a GFP antibody from lysates of COS-7 cells transfected with RNF10 FL, RNF10(221–802) and GFP-GluN2A. WB analysis was performed by using a Myc antibody. No ab lanes: control lanes in absence of antibodies during the co-i.p. assay. (J) Co-i.p assay performed by using a Myc antibody from lysates of COS-7 cells transfected with RNF10 truncation mutants and GFP-GluN2A. WB analysis was performed by using a GFP antibody (JL-8). (K,L) COS-7 cells expressing RNF10 were transfected with GFP-GluN2A and Myc-RNF10 FL or Myc-RNF10 (221–802) constructs and immunolabeled for GFP (green), Myc (red) and Dapi (blue) (G); scale bar: 10 μm. The histogram (H) shows the quantification of Myc/EGFP co-localization index [n=10; ***p<0.001 RNF10 FL vs RNF10(221–802); unpaired Student’s t-test]. (M) GST and GST-RNF10 fusion proteins were incubated in a pull-down assay with rat hippocampal extracts. WB analysis was performed with GluN2A antibody. https://doi.org/10.7554/eLife.12430.005 To strengthen these results, GFP-GluN2A (1–1049) truncation mutant (bearing a stop codon at aa 1049) was transfected in COS-7 cells to evaluate its capability to interact with endogenous RNF10 (Figure 3C,D). Immunofluorescence assay revealed a high GFP-GluN2A(1–1049)/RNF10 clustering leading also to RNF10 redistribution from the nucleus to the cytoplasm (Figure 3C; cytoplasm/nucleus ratio 0.98 ± 0.04 vs 0.26 ± 0.02 GFP-GluN2A(1–1049) vs untransfected, p<0.001, n=10; unpaired Student’s t-test), similarly to what has been described above for GFP-GluN2A construct (see Figure 2). Accordingly, RNF10 co-immunoprecipitated with both GFP-GluN2A and GFP-GluN2A (1–1049) from COS-7 lysates (Figure 3D), confirming that RNF10 interacts with the GluN2A juxta-membrane aa 839–1049 region. Interestingly, GluN2A(991–1029) domain previously shown to be responsible for a calcium-dependent binding of the NMDAR subunit with Calmodulin (CaM; Bajaj et al., 2014). Based on this consideration, we first evaluated whether calcium (Ca2+) could modulate GluN2A/RNF10 interaction. Using a pull-down assay we found that the presence of free Ca2+ (2 mM) in the buffer significantly reduced GluN2A binding to GST-RNF10 FL (Figure 3E; -37.8% ± 6.2%; p<0.01). Similarly presence of free Ca2+ (2 mM) in the buffer significantly reduced RNF10 binding to GST-GluN2A(839–1464) (Figure 3F; -65.4 ± 8.8%; p<0.05). As previously reported (Bajaj et al., 2014), Ca2+ induced also the binding of CaM to the GluN2A fusion protein (Figure 3F). In addition, co-incubation with exogenous CaM completely prevented the interaction between RNF10 and GluN2A in the pull-down assay (Figure 3G) only in the presence of Ca2+ in the buffer (Bajaj et al., 2014). Similarly, a co-i.p. assay from COS-7 cells transfected with GFP, GFP-GluN2A or GFP-GluN2B confirmed the capability of Ca2+/CaM to disrupt RNF10/GluN2A complex (Figure 3H). The RNF10 protein contains a binding sequence for the transcription factor Mesenchyme Homeobox 2 (Meox2; Meox2 Binding Domain, MBD, aa 101–185; Lin et al., 2005), a Ringer Finger Domain (RFD, aa 225–270) and two putative nuclear localization sequences (NLS1, aa 591–599 and NLS2, aa 784–791). Several Myc-RNF10 mutants were prepared and co-transfected with GluN2A in COS-7 cells in order to identify the RNF10 domain responsible for the interaction with the NMDAR subunit. The RNF10(221–802) construct failed to interact with GluN2A as demonstrated by both co-i.p. (Figure 3I) and co-localization (Figure 3K,L) assays when compared to RNF10 full length (FL). Conversely, all RNF10 truncation mutants bearing the RNF10(1–221) domain co-immunoprecipitated with GFP-GluN2A (Figure 3J). Finally, pull-down assay performed from adult rat brain tissue using GST-RNF10 fusion proteins confirmed that the RNF10 N-terminal region 1–221 is crucial for the binding to GluN2A (Figure 3M). RNF10 silencing induces molecular and morphological modifications of the glutamatergic synapse RNF10 is a member of the Ring Finger Protein family, which has been generally implicated in development, transcriptional regulation, signal transduction, DNA repair and oncogenesis (Saurin et al., 1996). However, the neuronal function of RNF10 is still unknown (Lin et al., 2005; Malik et al., 2013). To understand the role of RNF10 in neurons, we silenced RNF10 expression by using a short hairpin (sh) RNF10 knock-down or a scramble sequence (as control) in primary hippocampal cultures. We tested three different sequences of RNF10 shRNA (shRNF10; see Materials and methods); shRNF10 that led to the highest level of RNF10 downregulation (>90%; TRCN0000041128) was selected and used in all experiments. Confocal imaging of hippocampal neurons transfected with shRNF10 or scramble plasmids demonstrated a significant effect of RNF10 knock-down on dendritic spine morphology (Figure 4A–E). In particular, RNF10 silencing produced a significant reduction in dendritic spine density (Figure 4B) without any effect on dendritic spine length (Figure 4C) or dendritic spine head width (Figure 4D). For a more detailed morphological analysis, dendritic spines were categorized according to their shape (mushroom, thin and stubby) using a validated classification method (Bourne and Harris, 2008). However, no effect of shRNF10 in the proportion of dendritic spine subtypes was observed (Figure 4E). Notably, the effect of RNF10 silencing on dendritic spine density was fully rescued by co-expressing a wild-type human variant of RNF10 resistant to shRNA (see Materials and methods; flag-RNF10; Figure 4A–E). Interestingly, overexpression of RNF10 per se had no effect on dendritic spine morphology (data not shown). We then verified whether the loss of spines following RNF10 protein knock-down was correlated with an altered expression of the main components of the excitatory synapse. To this end, we infected primary hippocampal neurons (Figure 4F, left panels) and organotypic hippocampal slices (Figure 4F, right panels) with pLKO-shRNF10 lentivirus or scramble sequence as a control. As expected, RNF10 silencing produced a significant reduction of RNF10 protein level compared to dissociated neurons and slices treated with scramble construct (Figure 4F). Most importantly, RNF10 silencing resulted in a significant decrease of GluN2A, PSD-95 and the GluA1 subunit of the AMPA receptor (AMPAR) protein levels in the total cell homogenate of pLKO-shRNF10-infected dissociated neurons and organotypic slices (Figure 4F). Figure 4 Download asset Open asset RNF10 silencing induces molecular and morphological modifications of the glutamatergic synapse. (A) Confocal images of primary hippocampal neurons (DIV14) transfected at DIV7 with pGIPZ-scramble, shRNF10 and shRNF10 plus flagRNF10 and immunolabeled for GFP (green); scale bar: 5 μm. (B-E) Histograms showing the quantification of dendritic spine density (B) (n=6–10; *p<0.05, scramble vs shRNF10; ***p<0.001, shRNF10 vs shRNF10 + flagRNF10; one-way ANOVA, followed by Tukey post-hoc test), dendritic spine length (C), dendritic spine head width (D) and dendritic spine type (E). (F) WB analysis from homogenates of primary hippocampal neurons (DIV14) and organotypic hippocampal slices (DIV14) lentivirally infected with pGIPZ-scramble sequence (scramble) as control or with pLKO-shRNF10 (shRNF10). The histogram shows the quantification of the expression levels of GluN2A, GluA1, PSD-95 and GluN2B in shRNF10-infected neurons and slices, normalized on tubulin and expressed as % of scramble (n=6; *p<0.05; **p<0.01; unpaired Student’s t-test). (G) mRNA expression levels of genes associated with synaptic transmission or dendritic spine morphology by real-time PCR from DIV14 organotypic hippocampal slices lentivirally infected (DIV4) with pGIPZ-scramble sequence (scramble) as control or with pLKO-shRNF10 (shRNF10) (n=4, ***p<0.001; **p<0.01; *p<0.05; unpaired Student’s t-test). (H) WB for ArhGef6, ArhGap4, Ophn1 and tubulin from cell lysates of organotypic hippocampal slices infected with pGIPZ-scramble or with pLKO-shRNF10. The histogram shows the quantification of protein levels from shRNF10 samples with respect to pGIPZ-scramble, following normalization on tubulin (n=3, *p<0.05; **p<0.01; unpaired Student’s t-test). https://doi.org/10.7554/eLife.12430.006 To learn more about the effect of RNF10 knock-down on gene expression, we performed a microarray analysis from organotypic hippocampal slices virally infected with pLKO-shRNF10 lentivirus in order to identify RNF10 target genes. Interestingly, heat map of differentially expressed genes (Figure 5A), gene ontology analysis (Figure 5B) and real-time PCR validation (Figure 4G) showed that RNF10 silencing modulated the expression of several genes involved in excitatory synaptic transmission and dendritic spine morphology (Vogt et al., 2007; Michaluk et al., 2011; Ramakers et al., 2012). Furthermore, we confirmed by WB analysis the effect of viral infection with pLKO-shRNF10 on protein levels of some of the RNF10 target genes, such as Ophn1, ArhGap4 and ArhGef6 (Figure 4H). Figure 5 Download asset Open asset Heat map and gene ontology of differentially expressed genes identified in microarray experiments. (A) Heat map of differentially expressed genes. Expression data are reported as log2 and blue color indicates high expression values and red color low expression value. Sc represents the different scramble control (1–4), while Sh represents the shRNA against RNF10 (5–8). (B) Gene ontology analysis of biological processes and enriched pathways analysis of differentially regulated genes in absence of RNF10. https://doi.org/10.7554/eLife.12430.007 Neuronal activity regulates RNF10 synaptonuclear localization The synaptic and nuclear localization in neurons, the presence of two different NLS motifs and its interaction with GluN2A and Meox2 (Lin et al., 2005) suggest that RNF10 could be a novel synaptonuclear protein messenger. To validate this hypothesis, we assessed whether the modulation of neuronal activity affects the subcellular localization of RNF10 and its association with the interacting proteins. We first enhanced synaptic excitatory activity with the GABA-A receptor antagonist Bicuculline (50 μM) in the presence of the K+ channel blocker 4-AP (2.5 mM; Hardingham et al., 2002) ('Bic' treatment). Enhanced excitatory activity significantly decreased RNF10 immunoreactivity along dendrites as compared to untreated neurons, without affecting PSD-95 (Figure 6A) labeling and induced a significant increase in RNF10 nuclear staining (Fig" @default.
• W2988339731 created "2019-11-22" @default.
• W2988339731 creator A5001945408 @default.
• W2988339731 creator A5005693249 @default.
• W2988339731 creator A5006467212 @default.
• W2988339731 creator A5018110934 @default.
• W2988339731 creator A5022198325 @default.
• W2988339731 creator A5023571522 @default.
• W2988339731 creator A5024096995 @default.
• W2988339731 creator A5033599123 @default.
• W2988339731 creator A5043265559 @default.
• W2988339731 creator A5047060848 @default.
• W2988339731 creator A5057692461 @default.
• W2988339731 creator A5066008651 @default.
• W2988339731 creator A5070954835 @default.
• W2988339731 creator A5071023627 @default.
• W2988339731 creator A5085825658 @default.
• W2988339731 creator A5086913593 @default.
• W2988339731 creator A5087488820 @default.
• W2988339731 creator A5088104742 @default.
• W2988339731 creator A5091093273 @default.
• W2988339731 date "2016-02-12" @default.
• W2988339731 modified "2023-09-25" @default.
• W2988339731 title "Author response: Ring finger protein 10 is a novel synaptonuclear messenger encoding activation of NMDA receptors in hippocampus" @default.
• W2988339731 doi "https://doi.org/10.7554/elife.12430.015" @default.
• W2988339731 hasPublicationYear "2016" @default.
• W2988339731 type Work @default.
• W2988339731 sameAs 2988339731 @default.
• W2988339731 citedByCount "0" @default.
• W2988339731 crossrefType "peer-review" @default.
• W2988339731 hasAuthorship W2988339731A5001945408 @default.
• W2988339731 hasAuthorship W2988339731A5005693249 @default.
• W2988339731 hasAuthorship W2988339731A5006467212 @default.
• W2988339731 hasAuthorship W2988339731A5018110934 @default.
• W2988339731 hasAuthorship W2988339731A5022198325 @default.
• W2988339731 hasAuthorship W2988339731A5023571522 @default.
• W2988339731 hasAuthorship W2988339731A5024096995 @default.
• W2988339731 hasAuthorship W2988339731A5033599123 @default.
• W2988339731 hasAuthorship W2988339731A5043265559 @default.
• W2988339731 hasAuthorship W2988339731A5047060848 @default.
• W2988339731 hasAuthorship W2988339731A5057692461 @default.
• W2988339731 hasAuthorship W2988339731A5066008651 @default.
• W2988339731 hasAuthorship W2988339731A5070954835 @default.
• W2988339731 hasAuthorship W2988339731A5071023627 @default.
• W2988339731 hasAuthorship W2988339731A5085825658 @default.
• W2988339731 hasAuthorship W2988339731A5086913593 @default.
• W2988339731 hasAuthorship W2988339731A5087488820 @default.
• W2988339731 hasAuthorship W2988339731A5088104742 @default.
• W2988339731 hasAuthorship W2988339731A5091093273 @default.
• W2988339731 hasBestOaLocation W29883397311 @default.
• W2988339731 hasConcept C125411270 @default.
• W2988339731 hasConcept C163235415 @default.
• W2988339731 hasConcept C169760540 @default.
• W2988339731 hasConcept C170493617 @default.
• W2988339731 hasConcept C185592680 @default.
• W2988339731 hasConcept C2781161787 @default.
• W2988339731 hasConcept C54355233 @default.
• W2988339731 hasConcept C67018056 @default.
• W2988339731 hasConcept C86803240 @default.
• W2988339731 hasConceptScore W2988339731C125411270 @default.
• W2988339731 hasConceptScore W2988339731C163235415 @default.
• W2988339731 hasConceptScore W2988339731C169760540 @default.
• W2988339731 hasConceptScore W2988339731C170493617 @default.
• W2988339731 hasConceptScore W2988339731C185592680 @default.
• W2988339731 hasConceptScore W2988339731C2781161787 @default.
• W2988339731 hasConceptScore W2988339731C54355233 @default.
• W2988339731 hasConceptScore W2988339731C67018056 @default.
• W2988339731 hasConceptScore W2988339731C86803240 @default.
• W2988339731 hasLocation W29883397311 @default.
• W2988339731 hasOpenAccess W2988339731 @default.
• W2988339731 hasPrimaryLocation W29883397311 @default.
• W2988339731 hasRelatedWork W1965176956 @default.
• W2988339731 hasRelatedWork W1994475466 @default.
• W2988339731 hasRelatedWork W2018213389 @default.
• W2988339731 hasRelatedWork W2026855285 @default.
• W2988339731 hasRelatedWork W2032398857 @default.
• W2988339731 hasRelatedWork W2042301393 @default.
• W2988339731 hasRelatedWork W2070475355 @default.
• W2988339731 hasRelatedWork W2147502655 @default.
• W2988339731 hasRelatedWork W2148691225 @default.
• W2988339731 hasRelatedWork W2617486712 @default.
• W2988339731 isParatext "false" @default.
• W2988339731 isRetracted "false" @default.
• W2988339731 magId "2988339731" @default.
• W2988339731 workType "peer-review" @default.