AIMFragile X syndrome (Fragile X syndrome, FXS) is one of the most common inheritedmental retardation diseases. Exploring the pathogenesis and therapeutic approach of FXSis an important social and clinical problem. FXS is caused by mutation orhypermethylation of Fmr1, leading to loss of its product FMRP. FMRP is an mRNAbinding protein involved in mRNA aggregation and regulation of transcription, whichmainly function in protein inhibitory expression, thus affecting the neuronal developmentand synaptic plasticity. Recent study showed that loss of FMRP could cause excessivemetabotropic glutamate receptor-dependent translation, resulting in early morphologicalabnormalities in neuronal development and synaptic transmission disorders. In clinic, thepatients of FXS exhibited intelligence inability, seizures, mental retardation and othertypical symptoms. Therefore, FMRP abnormal regulation of protein translation is the keyto the pathogenesis of fragile X syndrome.The abnormal synaptic plasticity of Fragile X syndrome mainly represented indendritic development disorders on the structure and synaptic transmission disorders onthe function. Morphological studies of Fmr1knockout (KO) neurons indicated a decreaseddensity of glutamate AMPA receptor in synapses and thinner spines. Electrophysiologicalrecordings from brain slice found the impaired induction of long-term potentiation (LTP) in the anterior cingulate (ACC) cortical pyramidal neurons. Behavioral test showed thereduced fear memory involved by prefrontal cortex.Astrocytes are the major component of glial cells in the nervous system. Recentstudies have found that astrocytes not only supported neurons, but also possessed someimmunological characteristics. Therefore, it is very important for targeting astrocytes inthe prevention and treatment of neurological diseases. There are a variety ofneurotransmitter receptors expressed by astrocytes, which could release several solublefactors for neuronal survive, development, differentiation, and proper migration. Somestudies have shown a decreased length and increased branches of dendrites in patients ofFXS. When neurons cocultured with Fmr1KO astrocytes, the dendritic development wereabnormal. However, when neurons cocultured with wild-type (WT) astrocytes, theneuronal morphology was significantly improved. These suggest that astrocytes may playan important role in the development of FXS, but the mechanism is still unknown.There are many factors which regulate synaptic plasticity.17β-estradiol (E2), as arepresentative of steroids, has an extensive effect in the reproductive, skeletal andcardiovascular system. In recent years, the role of E2in the nervous system has alsogained a lot of attention. E2can induce LTP through promoting the growth of dendriticspines and the formation of synapses. However, the effect of estrogen in FXS is unclear.Because a lack of FMRP caused mental retardation in FXS, our study focused on the roleof FMRP in learning and memory, and the signaling pathway of estrogen regulation inabnormal synaptic plasticity.In this study, we investigated the synaptic plasticity regulated by astrocytes andestrogen on the structure and function respectively, in order to elucidate the mechanism ofabnormal synaptic plasticity in FXS, which provide new ideas and theoretical basis forexploring potential drug targets and clinical treatment strategies.METHODS1. To investigate the role of astrocytes in neuronal development, the Fmr1wild-type(WT) and knockout (KO) astrocytes were cocultured with neurons respectively. Bygenerating astrocytic conditioned medium (ACM), the difference of dendritic morphology between WT and KO ACM-treated neurons at DIV7was assessed.2. The constructed FMRP expression vector was transfected into Fmr1KO astrocytes.After collecting of ACM from FMRP-transfected KO astrocytes, FMRP expression inastrocytes was important for neuronal development.3. Detecting the concentration of glutamate in WT and KO ACM by HPLC, ROS andMDA were measured in order to characterize the oxidative damage caused byglutamate.4. Neurotrophic factors in the ACMs and cerebral cortex were examined by using ELISAkits, such as NGF,BDNF,NT-3,GDNF and CNTF. To determine whether NT-3isresponsible for the dendritic disorder caused by KO ACM, we added exogenous NT-3into WT ACM to observe the change in neuronal morphology. Meanwile, we analyzedthe effects of neutralizing antibody against NT-3on neuronal dendritic developmentafter KO ACM treatment.5. FMRP has multiple RNA-binding motifs and is involved in translational regulation.To verify that some mRNA are a potential target of FMRP, we performedRNA-Binding Protein Immunoprecipitation (RIP) to examine whether FMRPinhibited some expression in astrocytes.6. Before transplantation in ACC of Fmr1KO mice, the KO astrocytes were infected byshRNA. After ten days of surgery, trace fear conditioning was performed in order toobserve the change of learning and memory.7. After treatment with E2, GluR1trafficking and phosphorylation in cultured neuronswere studied, as well as simultaneous application of mGluR5antagonist DL-AP3.8. To observe the effects of mGluR5antagonist DL-AP3on LTP induction in ACC, bothwhole-cell patch-clamp recordings and Med64-channel recording were performed.9. By using coimmunoprecipitation, the effect of E2on ER-CAV1-mGluR1/5complexes both in WT and KO neurons were observerd.10. By using shRNA knockdown CAV1expression in Fmr1KO neurons, the GluR1distribution in cultured neurons, LTP induction in ACC slices, dendritic spinemorphology and animal behaviors were observed. RESULTS1. Both WT and KO neurons exhibited the abnormal dendritic morphology whencocultured with Fmr1KO astrocytes. The neuronal growth was abnormal in KO ACM,but was improved after transfecting with FMRP expression vector.2. Fmr1KO astrocytes released excessive glutamate through the synthesis and transportmechanisms, resulting elevated oxidative stress in FXS.3. Due to FMRP interacted directly with NT-3mRNA, the NT-3levels of ACM andcerebral cortex in Fmr1KO mice were higher than those in WT mice, whereas otherfour glial-derived neurotrophic factors remained unchanged. The NT-3protein levels,but not mRNA levels were increased in KO astrocytes.4. When exogenous NT-3was added into WT ACM, the dendritic growth performedabnormal as with increased concentrations of NT-3. Conversely, moderateneutralization of NT-3in KO ACM prevents abnormal neuronal growth. Too much ortoo little of NT-3would be toxic to neuronal growth.5. Knockdown of elevated NT-3levels in Fmr1KO astrocytes restores neuronaldendritic growth. After microinjection of NT-3shRNA infected KO astrocytes in ACCregion, the NT-3level was down-regulated, and the deficit of trace fear memory wasrescued in the KO mice.6. Estradiol partially regulates GluR1trafficking and phosphorylation throughmembrane-bound ERs based on the rapid action of E2. However, E2induced GluR1effects and LTP facilitation were absent from Fmr1KO mice.7. Simultaneous application of E2and mGluR5antagonist DL-AP3markedly rescuedthe LTP induction by E2in the Fmr1KO mice, and reversed the deficits in GluR1surface expression and phosphorylation at Ser831in Fmr1KO neurons.8. The impairment of E2-induced facilitation of synaptic potentiation was not attributedto the defect in endogenous E2levels and expression of ERs. ER coulped withexaggerated mGluR1/5, resulting impaired Gq-PLC-PKC signal pathway.9. Using RNA-binding protein immunoprecipitation, we found that FMRP bound with CAV1mRNA to inhibit its translation efficiency. The coimmunoprecipitation showedthe upregulated basal CAV1coupling to ER, leading to the inefficiencies of ER inrelaying the signal from upstream to downstream components in the Fmr1KOneurons.10. Knockdown of CAV1in Fmr1KO neurons significantly increased the GluR1surfaceexpression and phosphorylation at Ser831, and improved the formation ofER-CAV1-mGluR1/5complexes after treatment with E2. While knockdown ofCAV1in Fmr1KO mice, LTP was significantly induced by E2in slices from CAV1shRNA-infected KO mice. Furthermore, E2increased the proportion ofmushroom-shaped dendritic spines in pyramidal neurons, and improved trace fearmemory both in male and OVX mice.CONCLUSION1. It was confirmed that astrocytes played an important role in the fragile X syndrome.Fmr1KO astrocytes released excessive glutamate and NT-3, causing neuronaldendritic growth and developmental disorders.2. We clarified the molecular mechanism of which FMRP deletion caused excessivesecretion of NT-3. The neuronal dendritic morphology and fear conditioning could besignificantly improved by interfering glial-derived NT-3in early development offragile X syndrome3. It was found that the concentration of E2and the expression of ERs were normal inFmr1KO mice. However, estrogen can not modulate the synaptic plasticity of Fmr1KO mice through the membrane non-genomic effects.4. The ER-CAV1-mGluR1/5complexes were excessive coupled in Fmr1KO neurons,inhibiting its signal can rescue the modulation of E2on GluR1effect, dendritic spinemorphology, LTP facilitation and animal behavior. |