Skip to main content

Characterization of the molecular components of synapses in Fragile-X mental retardation syndrome: new insights into the FMRP regulatory mechanisms

Final Report Summary - SYNAPSES FXS (Characterization of the molecular components of synapses in Fragile-X mental retardation syndrome: new insights into the FMRP regulatory mechanisms.)

Proposal nº 275107. Synapses-FXS
Fragile X mental retardation syndrome (FXS) is the most frequent form of inherited intellectual disability and single gene-linked form of autism and it is characterized, at cellular and molecular level, by an increased protein synthesis, spine dysmorphogenesis and dysregulated glutamatergic and GABAergic systems. Dysmorphic spines are a conserved feature of several forms of intellectual disability as well as of different forms of neurodegeneration. Spine dysmorphogenesis is the result of errors in regulatory mechanisms active during synapse formation in early development and later on in life when synapses remodelling is required. FXS is due to mutations or silencing of the fragile X mental retardation gene (FMR1) encoding for the fragile X mental retardation protein (FMRP). FMRP is an RNA binding protein that regulates transport, stability and translation of hundreds of neuronal mRNAs. The dysregulation of a subset of FMRP target mRNAs is probably the major contribution to FXS.
With this project we aimed at identifying the affected molecules at FXS synapses. To achieve this goal we used a total of three different transgenic mouse models: the Fmr1 KO (a mouse model for FXS), the Fxr2 KO (the autosomal homologue of Fmr1) and the PSD-95TAP (a knock-in mouse that contains the Post Synaptic Density-95 protein tagged with a Tandem Affinity Purification Tag, TAP).
Different approaches were used to address our 3 objectives: (i) Identification of the postsynaptic core proteome of hippocampal and cortical excitatory synapses in the Fmr1 knock-out mouse model; (II) Identification of the mRNA/miRNA repertoire in the postsynaptic compartment of the Fmr1 knock-out mouse model; (III) Identification of the regulatory protein complex machinery regulating two well-established FMRP target such as Arc and PSD-95 mRNAs.

1. For the first objective, we used a knock-in PSD-95TAP, a genetically modified mouse that allows to purify the core proteome of the post-synaptic density (PSD), a region located underneath the postsynaptic membrane involved in synaptic structure, transmission and plasticity. These mice were generated in the laboratory of Prof. Seth Grant (Wellcome Trust Sanger Institute) and kindly provided to us. To identify the “native” complex at FXS synapses, the PSD-95TAP mouse model was crossed with the Fmr1 KO. Several high-scale purifications were performed from hippocampi and cortices from the new generated mouse model (PSD-95TAP x Fmr1ko). We were able to verify that some of the “core” components of the PSD, such as NR2A, NR2B, ARC were present in the TAP- purification indicating that the complex was well preserved during the purification steps. Next, samples were analyzed by mass spectrometry (LC-MS/MS). The analysis of the proteomic data and its relevance on the molecular signalling networks dysregulated at FXS synapses is on progress. Of note, we have recently compared the genes from genome-wide association studies and Autism, Schizophrenia and Mood Disorders databases with the available dataset of the brain FMRP transcriptome and identified several FMRP mRNA targets associated with above-mentioned disorders. These studies further support the importance of our studies on FXS synapses (Fernandez et al., 2013).

2. For the second objective, we first set up a complex procedure that allowed us to isolate synaptoneurosomes from the hippocampus. This new protocol (De Rubeis et al., 2013) is based on previously published methods and modified to biochemically isolate “sealed” synapses the pre- and post-synaptic compartments. Importantly, these synaptoneurosomes are able to respond to synaptic activation under glutamatergic and neurotrophic tyrosine kinase receptor type stimulations. Furthermore, using RNA from these “virtually pure synaptoneurosomes” and a commercially available microRNA array we report on a new group of synaptically expressed miRNAs and show that four of them are associated to FMRP and upregulated at FMRP-deficient synapses. Furthermore, some of the predicted microRNA mRNA targets are predicted targets for FMRP. This study provides a novel dataset of synaptically localized miRNAs and supports the crucial role of local dysregulation of mRNA metabolism at FXS synapses (manuscript in preparation).

3. For the third objective, we aimed at identifying the regulatory complex (ribonucleoproteins, RNPs) regulating the two well-established synaptic FMRP targets: PSD-95 and Arc mRNAs. It has been previously shown that in absence of FMRP, PSD-95 and Arc mRNA metabolism is dysregulated. We have initially isolated and identified by mass spectrometry analysis, the RBPs that bind to the 3’ untranslated region of PSD-95 mRNA. Among the 16 RNPs bound to PSD-95 mRNA, we have identified FMRP, as expected from previous studies, and its paralog the Fragile X mental retardation syndrome-related protein 2 (FXR2P). Of note, FXR2P was previously shown to be involved in adult neurogenesis and spine morphogenesis. We found that absence of FXR2P leads to a decreased translation of PSD-95 mRNA in the hippocampus implying a role for FXR2P as translation activator. Remarkably, PSD-95 mRNA fate and the equilibrium towards translational activation or repression, is the result of the FXR2P-FMRP interaction. Altogether these findings show, a combinatorial regulation of RBPs on a neuronal mRNA that ultimately affects fine-tuning of brain function (Fernandez et al., in preparation).

4. In addition to the three aims described above, we also studied the role of the Cytoplasmic FMRP interacting protein 1 (CYFIP1) on synapses remodelling. In human, deletions, duplications and copy number variants (CNVs) in CYFIP1 are associated to schizophrenia and autism. We showed that CYFIP1 orchestrates two molecular cascades, protein translation and actin polymerization, each of which is necessary for correct spine morphology in neurons. Importantly, absence of CYFIP1 at synapses mimicked the FXS synapses (De Rubeis et al., 2013).