The fragile X protein controls microtubule-associated protein 1B translation and microtubule stability in brain neuron development

Abstract

The fragile X mental retardation protein (FMRP) is a selective RNA-binding protein implicated in regulating translation of its mRNA ligands. The absence of FMRP results in fragile X syndrome, one of the leading causes of inherited mental retardation. Delayed dendritic spine maturation was found in fragile X mental retardation patients as well as in Fmr1 knockout (KO) mice, indicating the functional requirement of FMRP in synaptic development. However, the biochemical link between FMRP deficiency and the neuronal impairment during brain development has not been defined. How FMRP governs normal synapse development in the brain remains elusive. We report here that the developmentally programmed FMRP expression represses the translation of microtubule associated protein 1B (MAP1B) and is required for the accelerated decline of MAP1B during active synaptogenesis in neonatal brain development. The lack of FMRP results in misregulated MAP1B translation and delayed MAP1B decline in the Fmr1 KO brain. Furthermore, the aberrantly elevated MAP1B protein expression leads to abnormally increased microtubule stability in Fmr1 KO neurons. Together, these results indicate that FMRP plays critical roles in controlling cytoskeleton organization during neuronal development, and the abnormal microtubule dynamics is a conceivable underlying factor for the pathogenesis of fragile X mental retardation.

Fig. 1.

FMRP is vigorously up-regulated in the neonatal hippocampus and is selectively associated with the MAP1B mRNA. (A) Immunoblot analysis of FMRP in hippocampus derived from WT and Fmr1 KO littermates. A band is detectable in the Fmr1 KO lysates at the position of FXR1P on a prolonged exposure, likely caused by cross-reaction of the antibody to FXR1P. The blot was reprobed with antibodies against actin and eIF5α as loading controls. (B) Association of MAP1B mRNA with FMRP in the neonatal brain. (Upper) The input and immunoprecipitation of FMRP on immunoblot using WT and Fmr1 KO P7 brain. (Lower) RT-PCR analysis of mRNAs in the total input lysate (T) and in the immunoprecipitated FMRP-mRNP complexes (IP). A reaction without reverse transcriptase (-RT) was carried out as a negative control. The signals derived from MAP1B, MAP1A, and hypoxanthine phosphoribosyltransferase (HPRT) mRNA are marked on the left.

Fig. 2.

Misregulation of the developmentally programmed expression of MAP1B protein in Fmr1 KO hippocampus. (A) Comparable levels of MAP1B mRNA detected in the developing WT and Fmr1 KO hippocampus by RPA. MAP1B mRNA was quantified by a phosphorimager, normalized to the GAPDH mRNA, and illustrated in Lower. (B) Immunoblot analysis reveals a delayed decline of MAP1B protein in the Fmr1 KO hippocampus, most obvious at P10. The signal of MAP1B was quantified by the NIH image software and normalized to that of eIF5α. The average of MAP1B level in each WT litter was defined as 100%. The average of MAP1B level in each litter of the Fmr1 KO mice on the same immunoblot was normalized to that of the WT age-matched control. Results derived from multiple litters were subjected to standard t test. ^*, P < 0.05. (C) Immunofluorescent staining of MAP1B in WT and Fmr1 KO P7 hippocampus. MBP1B is mainly detected in the molecular layer (m) of CA1 and CA3 pyramidal cells and the hilus (arrowhead). An increase of MAP1B staining is more obviously detected in the hilus (arrowhead) and mossy fiber terminals (arrows) in the Fmr1 KO hippocampus. dgc, dentate gyrus granular cell layer.

Fig. 3.

Abnormally elevated MAP1B protein expression in primary cultures of Fmr1 KO cortical neurons. Parallel cultures were raised from embryonic day 16 brain of WT and Fmr1 KO mice, maintained in culture for 3 days before analysis for MAP1B expression. (A) Comparable MAP1B mRNA expression in WT and Fmr1 KO neurons. (Left) A representative image by RPA. (Right) The quantitative analysis by phosphorimager analysis normalizing MAP1B mRNA signal to that of GAPDH. (B) Immunoblot analysis reveals elevated MAP1B protein expression in Fmr1 KO neurons. The signal of MAP1B was normalized to that of house keeping genes in seven independent parallel cultures and statistically analyzed as shown in Left. ^*, P < 0.05 by standard t test. (C) Regulation of MAP1B-P1 by GSK3β kinase. Primary cortical neurons were subjected to mock or 10 mM LiCl treatment for 4 h. MAP1B-P1 was detected by the monoclonal antibody SMI-31 on immunoblot and quantified as described in B.

Fig. 4.

Polyribosome association of MAP1B mRNA in primary cultures of WT and Fmr1 KO neurons indicates misregulated translation. Cytoplasmic extracts were prepared in the presence of MgCl₂ (A and B) or EDTA (C) for linear sucrose gradient fractionation. The sedimentation of translation components, including ribosome subunits (40S and 60S) and the 80S monoribosome and polyribosomes, were monitored by OD₂₅₄ absorption as marked in each panel. MAP1B mRNA and GAP43 mRNA in each fraction were analyzed by RPA, and the corresponding signals are shown in correlation to fractions. (A) WT lysate fractionated on MgCl₂ gradient. (B) Fmr1 KO lysate fractionated on MgCl₂ gradient. (C) WT lysate fractionated on EDTA gradient. (D) Phosphorimager analysis of MAP1B mRNA distribution in A and B.

Fig. 5.

Abnormally increased microtubule stability in Fmr1 KO neurons. (A) Immunofluorescent staining of α-tubulin to visualize the microtubule networks. Cells were treated with 120 nM nocodazole for 2 h or mock-treated before being subjected to immunostaining. The disruption of microtubule polymers in the neuronal soma was scored as described below. Score 0 indicates no disruption of microtubule polymer, as indicated by arrowheads in Upper Left. Scores between 0 and 1 indicate negligible disruption of microtubule with majority of microtubule polymers remain intact. Scores between 1 and 2 indicate significantly disrupted microtubule polymers with obvious broken patches and dots, represented by the solid arrows in Upper Center. Scores larger than 2 indicate severely disrupted and dissolved microtubule networks, represented by open arrows in Upper Right. (Lower) The distribution of WT and Fmr1 KO neurons based on the above scale in mock- and nocodazole-treated cultures. Three hundred randomly selected neurons were subjected to the analysis in each parallel culture. Results from three independent experiments were subjected to statistical analysis. P < 0.05 by one-way ANOVA; ^**, P < 0.01; ^***, P < 0.001 by standard t test when comparing mock- and nocodazole-treated cultures. (B) Immunoblot analysis of acetylated α-tubulin in WT and Fmr1 KO neurons with or without 120 nM nocodazole. The signal of acetylated α-tubulin was quantified by NIH image software and normalized to that of actin. Results from three independent experiments were subjected to statistical analysis. ^*, P < 0.05 by standard t test when comparing mock- and nocodazole-treated WT cells as well as by paired t test when comparing WT and Fmr1 KO cells after nocodazole treatment.