Skip to main page content
U.S. flag
See More

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2013 Sep 18;79(6):1169-82.
doi: 10.1016/j.neuron.2013.06.039.

CYFIP1 coordinates mRNA translation and cytoskeleton remodeling to ensure proper dendritic spine formation

Silvia De Rubeis  1 , Emanuela Pasciuto, Ka Wan Li, Esperanza Fernández, Daniele Di Marino, Andrea Buzzi, Linnaea E Ostroff, Eric Klann, Fried J T Zwartkruis, Noboru H Komiyama, Seth G N Grant, Christel Poujol, Daniel Choquet, Tilmann Achsel, Danielle Posthuma, August B Smit, Claudia Bagni
Affiliations

CYFIP1 coordinates mRNA translation and cytoskeleton remodeling to ensure proper dendritic spine formation

Silvia De Rubeis et al. Neuron. .

Abstract

The CYFIP1/SRA1 gene is located in a chromosomal region linked to various neurological disorders, including intellectual disability, autism, and schizophrenia. CYFIP1 plays a dual role in two apparently unrelated processes, inhibiting local protein synthesis and favoring actin remodeling. Here, we show that brain-derived neurotrophic factor (BDNF)-driven synaptic signaling releases CYFIP1 from the translational inhibitory complex, triggering translation of target mRNAs and shifting CYFIP1 into the WAVE regulatory complex. Active Rac1 alters the CYFIP1 conformation, as demonstrated by intramolecular FRET, and is key in changing the equilibrium of the two complexes. CYFIP1 thus orchestrates the two molecular cascades, protein translation and actin polymerization, each of which is necessary for correct spine morphology in neurons. The CYFIP1 interactome reveals many interactors associated with brain disorders, opening new perspectives to define regulatory pathways shared by neurological disabilities characterized by spine dysmorphogenesis.

PubMed Disclaimer

Figures

Figure 1
Figure 1
CYFIP1 Participates in Two Distinct Complexes (A) Analysis of the crystal structure of the WRC. Upper panel: the eIF4E-binding site (light red, aa 733–751 of the human protein) of CYFIP1 (red) is partially occluded by NCKAP1 (green). Lower panels: left, a detail showing the interaction between CYFIP1 (red) and eIF4E (yellow); right, a detail showing that Lys743 is covered by NCKAP1 (green) when CYFIP1 (red) is within the WRC. Lys743 that is crucial for the interaction with eIF4E is highlighted with an arrowhead in both panels. (B) CYFIP1 IP from synaptoneurosomes. Lane 1, input (1/100); lane 2, CYFIP1 IP; lane 3, control IP (rabbit IgGs). Lanes shown belong to the same blot, n = 6. (C) NCKAP1 and eIF4E IPs from synaptoneurosomes. Lane 1, input (1/100); lane 2, NCKAP1 IP; lane 3, control (rIgGs) IP; lane 4, eIF4E IP; lane 5, control (mIgGs) IP. Lanes shown belong to the same blot, n = 3. See also Figure S1 and Table S1.
Figure 2
Figure 2
Upon BDNF Treatment CYFIP1 Shifts between eIF4E and WRC through a Conformational Change Induced by Active Rac1 (A) BDNF changes CYFIP1-eIF4E and CYFIP1-NCKAP1 colocalization in a Rac1-dependent manner. Left: representative dendrites costained for CYFIP1-eIF4E (yellow, upper row) and CYFIP1-NCKAP1 (cyan, lower row; scale bar represents 5 μm). See Figure S2B for single staining (green, CYFIP1; red, eIF4E; blue, NCKAP1). Neurons treated with vehicle or BDNF (100 ng/ml for 30 min) with/without NSC23766 (200 μM for 10 min pretreatment) are shown. Right: percentage of overlap expressed by Mander’s coefficient for CYFIP1-eIF4E (black) and CYFIP1-NCKAP1 (white) (three dendrites from at least ten neurons/condition; one-way ANOVA) with Bonferroni post hoc test; ∗∗p < 0.01, ∗∗∗p < 0.001). Arrowheads indicate colocalization puncta. Bars represent mean ± SEM. (B) BDNF changes CYFIP1-eIF4E and CYFIP1-NCKAP1 interactions through Rac1 activation. Left: CYFIP1 IP from synaptoneurosomes stimulated with BDNF (100 ng/ml for 30 min) with/without NSC23766 (200 μM for 10 min pretreatment). Lane 1, inputs (1/100) for vehicle; lane 2, BDNF; lane 3, BDNF + NSC23766; lanes 4–6, IPs for vehicle, BDNF and BDNF + NSC23766. Right, histogram represents CYFIP1-eIF4E (black) and CYFIP1-NCKAP1 (white) coimmunoprecipitation relative to untreated, control samples (n = 4, paired Student’s t test, ∗p < 0.05). Bars represent mean ± SEM. (C) Active Rac1 reduces the affinity of CYFIP1 for eIF4E. Left panel: m7GTP chromatography on cortical extracts after incubation with exogenous GTP-Rac1 or GDP-Rac1. Lane 1, input (1/100); lane 2, m7GTP chromatography; lanes 3–5, m7GTP chromatography in the presence of 0.45, 0.91, or 1.36 μM of GTP-Rac1, respectively; lanes 6–8, as lanes 3–5 but with GDP–Rac1. Right panel: association of CYFIP1 (black) and FMRP (gray) to eIF4E as percentage to vehicle (n = 6; repeated-measures ANOVA, ∗p < 0.05). Bars represent mean ± SEM. (D) Molecular dynamics simulation of CYFIP1 predicts an alternative conformation. Superposition of the CYFIP1 X-ray structure in the WRC (“planar,” light gray) upon the CYFIP1 structure obtained by clustering the conformations that dominate the end of the 135-ns molecular dynamics simulation (more “globular,” colored). The N-terminal, central, and C-terminal domains of the simulated structure are shown in blue, red, and green, respectively. The arrows indicate the principal movements during the conformational change. (E) mCherry-CYFIP1-EGFP coimmunoprecipitates with NCKAP1 and eIF4E. CYFIP1 was immunoprecipitated with an anti-GFP antibody. Lane 1, mock input (1/100); lane 2, transfected cells input (1/100); lane 3, IP from mock transfected cells; lane 4, IP from transfected cells. (F) Left: sketch showing how FRET reveals a globular CYFIP1 conformation. Right: in neurons CYFIP1 exists in a globular conformation that changes after BDNF treatment. Top to bottom: free EGFP and free mCherry (negative control); tandem mCherry-EGFP (intramolecular FRET); mCherry-CYFIP1-EGFP in neurons treated with the pan-Trk inhibitor k252a (100 nM for 24 hr); mCherry-CYFIP1-EGFP after BDNF (100 ng/ml for 30 min) treatment; mCherry-CYFIP1-EGFP after Rac1 inhibitor NSC23766 (200 μM for 10 min pretreatment) followed by BDNF treatment. At least n = 30, one-way ANOVA with Bonferroni post hoc test, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. Bars represent mean ± SEM. See also Figures S2 and S3.
Figure 3
Figure 3
Rac1 Regulates CYFIP1 and FMRP-Dependent mRNA Translation (A) ARC is upregulated by BDNF in a Rac1-dependent manner. Upper panel: immunoblot analysis of cortical neurons stimulated with BDNF (100 ng/ml for 30 min) +/– cycloheximide (CHX, 60 μM for 30 min) or NSC23766 (200 μM for 10 min pretreatment). Lower panel: ARC protein levels normalized to GAPDH and expressed as percentage to vehicle (n = 4, one-way ANOVA with Holm’s post hoc test, ∗∗∗p < 0.001). Bars represent mean ± SEM. (B) Upregulation of ARC upon BDNF treatment is protein-synthesis-dependent. Upper panel: immunoblot analysis of cortical neurons stimulated with NSC23766 (200 μM for 20 min) or BDNF +/− pretreatment with actinomycin D (Act D) (1 μg/ml for 30 min) or NSC23766 (200 μM for 10 min) as indicated above the lanes. Lower panel: quantification of ARC protein levels normalized to GAPDH and expressed as percentage of vehicle control (n = 3, one-way ANOVA with Holm’s post hoc test, ∗p < 0.05, ∗∗p < 0.01). Bars represent mean ± SEM. (C) Synthesis of ARC is altered in Cyfip1 knockdown, Fmr1 KO, and Fmr1 KO/Cyfip1 knockdown neurons. Upper panel: immunoblot of cortical neurons stimulated with BDNF (100 ng/ml for 30 min) +/– NSC23766 (200 μM for 10 min pretreatment). Lanes 1–3, scrambled shRNA; lanes 4–6, Cyfip1 shRNA, lanes 7–9, scrambled shRNA in WT neurons; lanes 10–12, scrambled shRNA in Fmr1 KO neurons; lanes 13–15, Cyfip1 shRNA in Fmr1 KO neurons. Lower right panels: basal expression of CYFIP1, FMRP, and ARC levels in vehicle-treated neurons. Protein levels were normalized to GAPDH and shown as percentage of scrambled shRNA. Grey, Cyfip1 shRNA; gray stripes, scrambled shRNA in Fmr1 KO; white, Cyfip1 shRNA in Fmr1 KO (n = 6, one-way ANOVA with Holm’s post hoc correction, ∗p < 0.05, ∗∗∗p < 0.001). Lower right panel: activity-induced ARC expression. ARC levels after BDNF +/− NSC23766 expressed as percentage to vehicle-treated neurons. Black, scrambled shRNA; gray, Cyfip1 shRNA; gray stripes, scrambled shRNA in Fmr1 KO; white, Cyfip1 shRNA in Fmr1 KO (n = 6, two-way ANOVA with Holm’s post hoc correction, ∗p < 0.05). Bars represent mean ± SEM. (D) Left panels: ARC level is unaffected in total cortical extracts from Cyfip1+/− mice. Immunoblot of CYFIP1, ARC, and GAPDH in WT (lane 1) and Cyfip1+/− (lane 2) mice. Protein levels were normalized to GAPDH and shown as percentage of WT; white, WT (n = 5); black, Cyfip1+/− n = 3, (Student’s t test, ∗∗∗p < 0.001). Right panels: as in the left panels but in synaptoneurosomes (n = 3, Student’s t test, ∗p < 0.05; ∗∗p < 0.01). Bars represent mean ± SEM.
Figure 4
Figure 4
CYFIP1 Deficiency or Mutations Affecting Interaction with eIF4E or NCKAP1 Alter Synaptic ARC and F-Actin (A) Dissection of the CYFIP1 interactions with NCKAP1 (mutΔ and mutH) or with eIF4E (mutE). Left panel: IP for YFP-CYFIP1 WT or mutants in HEK293T cells silenced for endogenous CYFIP1. Lane 1, input (1/50) from mock-transfected cells; lanes 2–5, input CYFIP1 siRNA with RNAi-resistant CYFIP1 WT, mutH, mutΔ, or mutE (1/50); lane 6, YFP-IP with mock-transfected cells; lanes 7–10, YFP-IP for WT, mutH, mutΔ, or mutE-CYFIP1. Asterisks indicate exogenous CYFIP1. Central panel: quantification of CYFIP1-NCKAP1 and CYFIP1-WAVE1 as percentage of WT (black) for mutH (red) and mutΔ (gray). Right panel: quantification of CYFIP1-eIF4E as percentage of WT (black) for mutE (white), see also Napoli et al. (2008). (B) Upper panels: CYFIP1 deficiency or mutations affecting the interaction with eIF4E or NCKAP1 alter synaptic ARC and F-actin levels. Panels show representative dendritic sections transfected with scrambled or Cyfip1 shRNA (F-GFP, upper panels) and stained for ARC (red, lower panels) in vehicle or BDNF-treated neurons. Spines are highlighted in yellow. Scale bar represents 1 μm. Histogram represents the ARC immunosignal normalized to the spine area for neurons transfected as indicated on the x axis and treated with vehicle (black) or BDNF (white) (at least n = 150, two-way ANOVA with Holm’s post hoc test, ∗∗∗p < 0.001). Lower panels: as above for F-actin (at least n = 50, two-way ANOVA with Holm’s post hoc test, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001). Bars represent mean ± SEM. See also Figure S4.
Figure 5
Figure 5
CYFIP1 Is Required for Correct Spine Morphology (A) Dendritic spines are defective in ex vivo cortical neurons from Cyfip1+/− mice. Cortical neurons were labeled with DiI by diolistic staining on brain slices. Panels show representative dendritic sections; scale bar represents 5 μm. Insets in the lower panel represent magnification of individual spines classified as mature (stubby and mushroom-like) and immature (long thin and filopodia). Scale bar represents 0.5 μm. (B) Dendritic spine morphology in cortical neurons from WT (black) or Cyfip1+/− (white) animals. Distribution of spines as percentage is shown (n = 282 WT; n = 310 Cyfip1+/−; χ2 test, ∗p < 0.05). Bars represent mean ± SEM. (C) Mean spine length in WT and Cyfip1+/− neurons (Student’s t test, ∗∗∗p < 0.001). Bars represent mean ± SEM. (D) Dendritic spines are altered in cultured cortical neurons silenced for Cyfip1. Outline of dendritic shafts from DIV14 primary cortical neurons transfected with scrambled, two Cyfip1 shRNAs (shRNA 319, 315), or shRNA 315 cotransfected with RNAi-resistant CYFIP1 WT, mutΔ, mutH, or mutE. Panels show representative dendritic sections; scale bar represents 5 μm. (E) Dendritic spine morphology of neurons shown in (D), expressed as percentage of mature (in black, stubby + mushroom-like) and immature (in white, long thin + filopodia) spines (at least ten neurons/condition, χ2 test, ∗∗∗p < 0.001). (F) Upper panel: mean spine length of neurons shown in (D) (one-way ANOVA with Bonferroni correction, ∗∗∗p < 0.001). Lower panel: Cumulative probability plots for mean spine length. Bars represent mean ± SEM. See also Figure S5.
Figure 6
Figure 6
The CYFIP1 Interactome in Mouse Cortex and Synapses and Its Relevance for Neuropsychiatric Disorders (A) CYFIP1 interactome, as revealed by MS of the proteins coimmunoprecipitating with CYFIP1 in cortical whole-cell lysate. IP with rabbit IgGs was used as a negative control; n = 3. The identified proteins are listed in Table S2, Table S3. (B) Validation of CYFIP1 interactors by reverse IP. Lane 1, input (1/100); lanes 2–6, specific IPs; lanes 7–8, controls with rabbit and mouse IgGs. See also Figure S6. (C) CYFIP1 interactome is partially RNase sensitive. Lane 1, input (1/50); lane 2, CYFIP1 IP; lane 3, CYFIP1 IP after RNase treatment; lane 4, control IP (rabbit IgGs); n = 5. (D) Outcome of MS analysis of the proteins coimmunoprecipitating with CYFIP1 in cortical synaptoneurosomes. IPs with rabbit IgGs were used as negative control; n = 6. The identified proteins are listed in Table S4. (E) Many CYFIP1 interactors are linked to neurological diseases. Indicated are the percentages of genes related to intellectual disability (ID), autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), schizophrenia (SCZ), major depressive disorder (MDD), and Alzheimer’s disease (AD) (the diagram is approximate, as some genes are related to more than one disease; see also Table S5). (F) Proposed model for the interplay of CYFIP1 complexes in brain. In neurons, CYFIP1 is associated with two distinct protein complexes, the CYFIP1-FMRP-eIF4E, which represses translation of specific mRNAs such as Arc/Arg3.1 mRNA, and the WRC (CYFIP1-NCKAP1-WAVE1-ABI2-HSPC300), which regulates actin remodeling. BDNF signaling activates Rac1, and GTP-Rac1 changes the equilibrium between the two CYFIP1 complexes by inducing a conformational change that releases CYFIP1 from eIF4E and relocates it to active WRC. As a consequence, actin cytoskeleton is remodeled and, concomitantly, the translation of proteins that encode cytoskeleton elements and synaptic function and plasticity is activated. The two processes converge to regulate proper spine morphology, which is compromised by perturbations of CYFIP1 expression or interference with CYFIP1-eIF4E and CYFIP1-WRC.
See this image and copyright information in PMC

References

    1. Aakalu G., Smith W.B., Nguyen N., Jiang C., Schuman E.M. Dynamic visualization of local protein synthesis in hippocampal neurons. Neuron. 2001;30:489–502. - PubMed
    1. Albertazzi L., Arosio D., Marchetti L., Ricci F., Beltram F. Quantitative FRET analysis with the EGFP-mCherry fluorescent protein pair. Photochem. Photobiol. 2009;85:287–297. - PubMed
    1. Bagni C., Tassone F., Neri G., Hagerman R. Fragile X syndrome: causes, diagnosis, mechanisms, and therapeutics. J. Clin. Invest. 2012;122:4314–4322. - PMC - PubMed
    1. Bassell G.J., Warren S.T. Fragile X syndrome: loss of local mRNA regulation alters synaptic development and function. Neuron. 2008;60:201–214. - PMC - PubMed
    1. Bear M.F., Huber K.M., Warren S.T. The mGluR theory of fragile X mental retardation. Trends Neurosci. 2004;27:370–377. - PubMed

Publication types

MeSH terms

Substances