Skip Navigation


Human Molecular Genetics Advance Access originally published online on September 19, 2007
Human Molecular Genetics 2007 16(24):3047-3058; doi:10.1093/hmg/ddm263
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
16/24/3047    most recent
ddm263v2
ddm263v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (15)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Davidovic, L.
Right arrow Articles by Khandjian, E. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Davidovic, L.
Right arrow Articles by Khandjian, E. W.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

The fragile X mental retardation protein is a molecular adaptor between the neurospecific KIF3C kinesin and dendritic RNA granules

Laetitia Davidovic1,2,3, Xavier H. Jaglin1,2,{dagger}, Aude-Marie Lepagnol-Bestel4, Sandra Tremblay1, Michel Simonneau4, Barbara Bardoni3 and Edouard W. Khandjian1,2,*

1 Unité de Recherche en Génétique Humaine et Moléculaire, Centre de recherche Hôpital Saint-François d’Assise, le CHUQ, Québec, Canada G1L 3L5, 2 Département de Biologie Médicale, Faculté de Médecine, Université Laval, Québec, Canada, 3 CNRS UMR6543, Faculté de Médecine, Université de Nice Sophia-Antipolis, 06 107, NICE-France, 4 INSERM U675, IFR2, Faculté de Médecine Xavier Bichat, Université Paris VII, 75 018 Paris, France

* To whom correspondence should be addressed at: URGHM, Centre de recherche Hôpital St François d’Assise, 10 rue de l’Espinay, Québec, PQ, Canada G1L 3L5. Tel: +1 4185254402; Fax: +1 4185254195; Email: edward.khandjian{at}crsfa.ulaval.ca

Received May 25, 2007; Revised August 15, 2007; Accepted September 11, 2007


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 FUNDING
 REFERENCES
 
Fragile X mental retardation 1 protein (FMRP) is an RNA-binding protein whose absence results in the fragile X syndrome, the most common inherited form of mental retardation. FMRP contains multiple domains with apparently differential affinity to mRNA and interacts also with protein partners present in ribonucleoprotein complexes called RNA granules. In neurons, these particles travel along dendrites and axons to translocate mRNAs to specific destinations in spines and growth cones, where local synthesis of neuro-specific proteins is taking place. However, the molecular mechanisms of how RNA granules are translocated to dendrites remained unknown. We report here the identification and characterization of the motor protein KIF3C as a novel FMRP-interacting protein. In addition, using time-lapse videomicroscopy, we studied the dynamics and kinetics of FMRP-containing RNA granules in dendrites and show that a KIF3C dominant-negative impedes their distal transport. We therefore propose that, in addition to modulate the translation of its mRNA targets, FMRP acts also as a molecular adaptor between RNA granules and the neurospecific kinesin KIF3C that powers their transport along neuronal microtubules.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 FUNDING
 REFERENCES
 
The absence of the RNA-binding protein fragile X mental retardation 1 protein (FMRP) is responsible for the fragile X syndrome (FXS), the main hereditary cause of mental retardation (1). In brain, FMRP is present in cytoplasmic ribonucleoparticles (RNPs) associated with somatic and synaptic polyribosomes, strongly suggesting its involvement in the translational control of its mRNA target and particularly in synaptic localized translation (2–5). Once synapses are established, localized translation will control distinct forms of activity-dependant synaptic plasticity, associated with changes in dendritic spine morphology (6). In adult FXS patients and Fmr1 null mice, the loss of FMRP induces abnormal immature-looking dendritic spines (7,8). In addition, Fmr1 null mice display alterations of several forms of activity-dependant synaptic plasticity, cortical long-term potentiation (9), and cerebellar long-term depression (10), further substantiating the essential role played by FMRP in localized mRNA translation. FMRP-containing RNPs are also observed as RNA granules traveling in the dendritic branching (11,12). These RNA granules correspond to complex structures containing mRNA, ribosomes and different RNA-binding proteins (13–17) that achieve the transport, targeting and release of neurospecific mRNAs locally at the synapse. Interestingly, a decrease in the amount of RNA granules and of polyribosomes in dendritic spines were reported in Fmr1 null mice, pointing out a possible role for FMRP in the regulation of the formation of RNA granules (18,19).

FMRP is one of the many components of RNP macromolecular complexes (reviewed in 20). Most of FMRP-interacting partners are RNA-binding proteins, such as its close homologues FXR1P and FXR2P (21). So far, a plethora of studies have focused on the RNA-binding properties of FMRP and several hundreds putative mRNA targets whose translation seem to be regulated by FMRP have been indexed. Among these, very few targets have been studied and validated in the neuronal context (reviewed in 22) and none has been definitely connected to the phenotype observed in patients. This suggests that the functions of FMRP in neurons are not restricted to its RNA-binding properties, a hypothesis based on the fact that FMRP interacts with cytoskeleton-linked proteins such as CYFIP1/2, Lgl and Ran-BPM (20,23), as well as with the motor proteins myosin Va, kinesin I heavy chain (KIF5) and dynein heavy chain (17,24,25). The functional significance of such interactions has been explored only in Drosophila melanogaster non-neural S2 cells (24), and evidence for a direct interaction between FMRP and motor proteins is lacking. In search for novel proteins that interact with FMRP in neurons, we identified the neurospecific kinesin KIF3C and we investigated the role of FMRP as a direct molecular adaptor between KIF3C and RNA granules.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 FUNDING
 REFERENCES
 
FMRP interacts with the neurospecific kinesin KIF3C in the yeast-two-hybrid system and in vitro
In order to identify new proteins that interact with FMRP in neurons, we screened a human fetal brain cDNA library using a yeast-two-hybrid system, with the major FMRP isoform Iso7 as bait. From ~2.8x106 clones screened, 78 positive clones displayed adenine and histidine prototrophy as well as ß-galactosidase activity. PCR and restriction analysis of selected colonies showed that among these, one colony carried a 2.2 kb insert bearing a sequence 100% identical to the 3' end of human neurospecific kinesin KIF3C cDNA (26) that is highly conserved in rodents (27). This insert corresponded to amino acids 403–792 of human KIF3C and contained also a stretch of the 3'-UTR of KIF3C cDNA. The interaction between KIF3C and FMRP was further confirmed by co-transforming the AH109 yeast strain with plasmids coding for the human KIF3C C-ter and human FMRP. Indeed, KIF3C C-ter was found to interact with FMRP, whereas no interaction was detected when plasmids bearing unrelated proteins were used (Table 1). These results showed that FMRP directly interact with KIF3C in the yeast-two-hybrid system.


View this table:
[in this window]
[in a new window]

  		Table 1. FMRP, FXR1P and FXR2P interact with the C-terminal domain of kinesin KIF3C in Yeast

 
Structurally, KIF3C is composed of three domains, which are common to all kinesin family members (Fig. 1A). Human (792 amino acids) and mouse (796 amino acids) KIF3C are evolutionary conserved since their amino acid sequences are 94.9% homologous. Therefore, in the present study, we used the available mouse KIF3C cDNA constructs. The kinesin motor region lies within the N-terminal 380 amino acids, while the central stalk domain contains an {alpha}-helix from residues 455 to 633, probably involved in homo- or heterodimerization (27). The C-terminal domain of KIF3C delineates a globular proline-rich tail domain (amino acids 634–796), which would mediate interactions with other proteins or cargoes (27–29). Interestingly, the original yeast-two-hybrid positive clone encompassed this putative cargo-binding site, since it contained the C-terminal half of human KIF3C (Fig. 1A). As KIF3C diverges mostly from KIF3A and KIF3B through its last 79 amino acids (27), we hypothesized that the interaction domain with KIF3C was likely to be located in this region. To verify the specificity of the interaction between KIF3C C-ter and FMRP, we performed a series of pull-down assays using the mouse KIF3C C-terminal 71 last amino acids (amino acids 725–796). As a negative control, we used its N-terminal stretch (amino acids 1–383), which contains the motor domain that has been proposed to interact with microtubules but not with the cargoes. The ability of KIF3C to interact with FXR1P and FXR2P, the two homologues of FMRP, was also tested. The GST-KIF3C 71 C-ter or GST-KIF3C N-ter immobilized on glutathione-Sepharose beads were incubated with either in vitro translated FMRP, FXR1P, FXR2P or luciferase as a negative control. The three members of the FXR family were shown to bind to the immobilized GST-KIF3C 71 C-ter, whereas luciferase did not (Fig. 1B). No binding to GST-KIF3C N-ter, or GST alone, was observed, confirming the specificity of the interaction. Interestingly, no binding of FMRP to the 71 last amino acids of KIF3B, another member of the KIF3 family was observed (not shown). To verify that the interaction between FMRP and KIF3C was direct and not RNA-dependent, GST-pulldowns were also performed in the presence of RNases A/T1. This treatment did not alter the binding of FMRP to KIF3C, indicating that the interaction occurs at the protein-protein level. To assess the robustness of the interaction between FMRP and GST-KIF3C 71 C-ter, another series of GST-pulldowns was conducted in the presence of different salt concentrations (Fig. 1C). In the absence of salts, the amount of FMRP bound to GST-KIF3C C-ter was three times higher than in the presence of 150 mM NaCl. At 500 mM NaCl and above, only trace amounts of FMRP could be detected.


Figure 1
View larger version (38K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. In vitro interaction of FMRP, FXR1P and FXR2P with KIF3C 71 C-ter in pulldown assays. (A) KIF3C is a neurospecific kinesin with three functional domains (shown here is the murine amino acids numbering). The N-terminus bears the motor domain conserved among all kinesins, the central part (stalk) contains a coil–coil domain responsible for hetero and homodimerization of KIF3C. Finally, the C-terminal domain carries the cargo-docking site. Shown in the second line is the yeast-two-hybrid original clone (amino acids 403–792 of human KIF3C) encoding the C-terminal half of the protein encompassing the dimerization domain and the putative cargo binding site (amino acids 406–792 of mouse KIF3C). N-ter (amino acids 1–383) and 71C-ter (amino acids 725–796) were fused to GST for GST-pulldown experiments. The anti-KIF3C antibody ({alpha}KIF3C) was raised against the 71 last amino acids by Yang and Goldstein (27). (B) Pull-down assays using 1 µg of the fusion protein GST-KIF3C 71 C-ter and the in vitro translated radiolabeled FXR proteins. FMRP, FXR1P and FXR2P bind to GST- KIF3C 71 C-ter but not to its N-terminal domain, indicating that KIF3C binds the FXRP through its last 71 amino acids. (C) The binding of FMRP to KIF3C 71C-ter is sensitive to increasing salt concentrations.

 
These results suggest that KIF3C binds FMRP, FXR1P and FXR2P through its cargo docking domain corresponding to its last 71 amino acids and that this interaction is labile in high-salt conditions.

KIF3C co-localizes with FMRP in the neuronal soma and is present in FMRP containing granules throughout the dendritic branching
To validate in vivo the potential interaction between KIF3C and FMRP, we performed immunofluorescence double labeling on rat brain cerebellar sections using mAb1C3 against FMRP and the affinity-purified anti-KIF3C antibody. In agreement with Yang and Goldstein (27), strong anti-KIF3C staining was detected in the Purkinje cell layer of the cerebellum (Fig. 2). Staining was strong in the cytoplasm of the soma and the proximal dendrites, becoming weaker distally. On the other hand, staining for FMRP was predominant in the soma, and faint signals were detected in the dendritic branching, whereas it appeared below detection levels in distal dendrites, as previously reported (30,31). Analyses of merged images reveal that KIF3C and FMRP are both present in the somatic compartment.


Figure 2
View larger version (76K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. KIF3C and FMRP are present in the somato-dendritic compartment of Purkinje cells of the cerebellum. Epifluorescence micrographs of rat brain cerebellum section carried out using rabbit anti-KIF3C antibodies (green) and anti-FMRP mAb1C3 (red) and counterstained with DAPI (blue). Apical dendrites are highly stained with the anti-KIF3C antibody, whereas only a faint FMRP signal is visible at the somatic branching of neurites. The merged image clearly shows that FMRP and KIF3C are present in the soma of Purkinje cells of the cerebellum. Control experiments using recombinant FMRP and KIF3C to assess the specificity of both antibodies are shown in the lower panels.

 
In order to assess the co-localization of KIF3C and FMRP in the dendritic branching of neurons, we double-stained primary cultures of rat hippocampal neurons with mAb1C3 and anti-KIF3C antibodies. Confocal microscopy analyses revealed strong staining of KIF3C in the soma, as well as in the dendritic arborization. As previously reported for KIF5 (32), most KIF3C freely diffuses in the cytoplasm in a cargo-unbound form, apparently masking the granular cargo-bound form. However, at higher magnification, a granular-like punctuate KIF3C staining was clearly observed in the dendritic branches (Fig. 3A and B). Consistent with previous reports (33), FMRP displays a granular-like pattern in the somatodendritic compartment (Fig. 3A), and at higher magnification, co-localization of FMRP and KIF3C was evident in several granule-like structures (Fig. 3A; see arrowheads in insets). However, this co-localization appears partial, since several puncta appeared positive only for FMRP (Fig. 3A; arrow in inset). We also studied the association of KIF3C with ß-tubulin, an essential structural element of the dendritic cytoskeleton. Figure 3B shows the co-localization of KIF3C with ß-tubulin in the somatodendritic compartment. At higher magnification, KIF3C staining appears as granular puncta displayed along microtubular structures that are revealed by the ß-tubulin staining (Fig. 3B).


Figure 3
View larger version (83K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3. KIF3C co-localizes with FMRP and ß-tubulin in the somatodendritic compartment of primary cultured hippocampal neurons. (A) Confocal microphotographs of 15 DIV primary cultured rat hippocampal neurons labeled with antibodies against KIF3C and FMRP. Low magnification shows co-localization of KIF3C and FMRP in the cytoplasm as well as in dendrites in granule-like structures. FMRP and KIF3C clearly co-localize in the majority of granules (inset, arrow head); however, several granules containing FMRP do not localize with KIF3C (arrows in inset). (B) Double-labeling with antibodies against KIF3C and ß-tubulin showing co-localization of the two proteins. At higher magnification, the presence of granule-like structures stained for KIF3C are detected aligned along microtubules stained for ß-tubulin.

 
These observations illustrate that KIF3C co-localizes with FMRP in granule-like structures in the somato-dendritic compartment of neurons and are in favor for a direct interaction of KIF3C with FMRP in vivo on microtubule tracks.

FMRP and KIF3C co-fractionate in GTP-induced mouse brain microtubule preparations
As a motor protein, KIF3C binds to microtubule structures in order to power the transport of its cargo. To validate the possibility that FMRP and KIF3C interact biochemically with microtubules, we searched for the presence of both proteins on such isolated structures. A fraction highly enriched in microtubules was obtained from mouse brain extracts through two cycles of temperature-dependant polymerization and depolymerization in the presence of GTP, using a standard procedure that has been applied to reveal the presence of FMRP and its homologue FXR1P in mouse testis microtubule extracts (34). Coomassie-blue staining of the proteins present in the final microtubules-enriched fraction (Fig. 4A,) and total homogenate (Fig. 4B) clearly show the enrichment in a 55 kDa protein, while a drastic reduction of total proteins was observed. This 55 kDa protein was identified as ß-tubulin (Fig. 4A). Densitometric analyses of the Coomassie-stained proteins indicated a ~15-fold enrichment of ß-tubulin, while immunoblot analyses to reveal the presence of FMRP, KIF3C and L7 did not show a clear increase of the proteins, indicating that only a fraction of these proteins is recovered with polymerized microtubules (see also below). The presence of the ribosomal protein L7 is indicative of the presence of ribosomes attached to the microtubules. On the other hand, synaptophysin, an integral membrane glycoprotein present in presynaptic vesicles in neurons, as well as the 78 kDa glucose-regulated/immunoglobulin heavy chain binding protein (GRP78/BiP), an endoplasmic reticulum (ER)-localized protein, were not detected in the microtubules-enriched pellet, suggesting that these preparations contain a selective subsets of proteins.


Figure 4
View larger version (53K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4. FMRP and KIF3C resemble with in vitro polymerized brain microtubules. (A) A microtubule-enriched fraction (~2 µg) obtained following two cycles of temperature dependent GTP-induced polymerization and depolymerization and (B) total mouse brain homogenate (~20 µg) were analyzed by SDS–PAGE and the proteins were either stained with Coomassie Blue (CB) or analyzed by immunoblotting to reveal ß-tubulin (Tub), KIF3C (Kif), FMRP, ribosomal protein L7, synaptic vesicle marker synaptophysin (SYP) and the ER marker BiP. Note the strong enrichment in ß-tubulin between (A) and (B) visible by Coomassie staining and by immuno-blotting with anti-Tub. FMRP and KIF3C are present in the microtubule-enriched fraction (A) together with L7, whereas the synaptic vesicle marker synaptophysin (SYP) and the endoplasmic reticulum marker BiP were not detected.

 
Since the experimental conditions used here to induce in vitro polymerization of microtubules could have induced artefactual re-association of FMRP with microtubules, analyses were performed using different experimental conditions that prevent microtubule polymerization (Fig. 5). Using the standard microtubule buffer devoid of non-ionic detergent, considerable amounts of FMRP were recovered in the first pellet (P1) obtained at 12 000 g that contains a subset of FMRP-RNPs associated with polyribosomes attached to the ER (2). The 12 000 g supernatant was then subjected to a microtubule-polymerization step by incubation at 37°C, in the presence of GTP, and polymerized microtubules were obtained after centrifugation at 25 000 g. At this step, the presence of tubulin, together with FMRP and L7, suggests that a population of FMRP-containing RNPs remained attached to polymerized microtubules. Omission of GTP or incubation at 4°C, both regimes known to prevent microtubules-polymerization, yielded trace amounts of ß-tubulin as well as FMRP and the ribosomal L7 marker in the P2 fraction, indicating that these proteins are recovered only in the polymerized microtubule fraction. The 25 000 g supernatant was further centrifuged at 105 000 g to yield a P3 fractions that contain polyribosomes and RNA granules. As expected, this fraction contains the great majority of FMRP (3,4), but low levels of tubulin. Analyses of the final supernatant (Sup) reveal that considerable amounts of soluble tubulin are recovered in the final supernatant in conditions that prevent microtubule-polymerization, while FMRP and L7 are not detected. These data indicate that, in the conditions used to polymerize microtubules, a subset of RNPs containing FMRP and L7 can be recovered associated with these in vitro polymerized structures. Comparing the amount of FMRP recovered in the P2 fraction (polymerized microtubules) with that of P3 containing polyribosomes, we estimate that in the experimental conditions used here, approximately 5–15% of total brain FMRP is recovered in structures associated with microtubules while the great majority remains associated with polyribosomes.


Figure 5
View larger version (53K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 5. FMRP and L7 sediment with polymerized microtubules fractions. Total mouse brain was homogenized in PEM buffer containing GTP or not and incubated at 37°C or at 4°C. Fractions were obtained after differential centrifugations (see Material and Methods) at 12 000 g (P1), 25 000 g (P2) and 105 000 g (P3). Extracts were incubated at 37°C, with or without 1 mM GTP for 30 min, or incubated at 4°C in presence of GTP. Each fraction was re-suspended in a volume normalized to the starting volume and equal volumes (20 µl) were analyzed by immunoblotting, using mAb1C3 against FMRP, anti-L7a against the ribosomal marker L7 and mAbE7 against ß-tubulin.

 
KIF3C co-immunoprecipitates with FMRP in mouse brain microtubule preparations
To further document the in vivo interaction between FMRP and KIF3C, we performed immunoprecipitation studies using the microtubule enriched fraction. As a negative control, we used the mAb1C3 directed against FMRP that fails to immunoprecipitate FMRP (35 and our unpublished observation) and a fortiori KIF3C (Fig. 6, top panels). In contrast, when using mAb7G1-1 antibody directed against FMRP, substantial amounts of FMRP were recovered, especially at high salts concentrations, in agreement with Ceman et al. (35), and KIF3C could be detected in association with FMRP only in buffer containing NaCl concentrations lower than 150 mM in agreement with the GST-pulldown experiments (Fig. 1C). These results indicate that both, in vitro and in vivo, the interaction between FMRP and KIF3C is salt labile. Finally, we used the anti-KIF3C antibody to immunoprecipitate KIF3C (Fig. 6, bottom panels) and could reveal the presence of FMRP, however only at low salt concentrations.


Figure 6
View larger version (74K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 6. FMRP co-immunoprecipitates with KIF3C in microtubule-enriched fractions from mouse brain. Immunoprecipitations were performed using microtubule-enriched fractions derived from GTP-induced mouse brain polymerized microtubules in the presence of increasing salt concentrations. As a negative control, mAb1C3 was used against FMRP and failed to immunoprecipitate FMRP as well as KIF3C. On the other hand, FMRP was immunoprecipitated in the presence of mAb7G1, particularly at high salt concentrations, while faint amounts of KIF3C were detectable in immunoprecipitates performed in the presence of low-salt concentrations. KIF3C was immunoprecipitated in the presence of anti-KIF3C antibodies, independently of the salts conditions, while again FMRP could be detected only if the reactions were conducted under low-salt concentrations. HC IgG, IgG heavy chains reacting with the secondary antibodies.

 
These analyses confirm the protein–protein interaction between the motor protein KIF3C and FMRP and suggest that the two proteins are present in complexes associated with polymerized microtubules.

In vivo dynamics and kinetics of neuronal granules containing GFP-FMRP
To investigate in vivo the dynamics and kinetics of particles containing FMRP, we followed the movements of GFP-FMRP using time-lapse video-microscopy in primary cultured neurons (Fig. 7). Dendritic granules containing exogenous GFP-FMRP had previously been shown to contain Fmr1 mRNA and ribosomal RNA (11,12,36), and we therefore hypothesized that this experimental system would reflect endogenous transport of FMRP-containing granules. In addition, since overexpression of FMRP is known to induce the formation of cytoplasmic stress-granules-like structures (37), cultures were analyzed at early times after transfection to minimize accumulation of GFP-FMRP (see Material and Methods section). In these conditions, we observed that GFP-FMRP displayed a granule-like pattern in the somato-dendritic compartment (Fig. 7), in agreement with previous studies (11,38). GFP-FMRP granules were heterogenous in size, with large granules often encountered in the soma, whereas smaller granules were distributed throughout the dendritic arborization (Fig. 7A). As already reported for neuronal RNA granules (13,16), the majority of the large GFP-FMRP particles remained static, corresponding most likely to either stress granules or mega-granules. In contrast, small granules displayed processive movements in both anterograde and retrograde directions, whereas others displayed back-and-forth oscillatory movements. Time-lapse videomicroscopy was performed only on small GFP-FMRP granules which most likely reflect endogenous FMRP RNA granules. Examples of the decomposition of the movements of two small granules during 140 s are illustrated in Figure 7B, while the distance moved by each granule along the x- and y-axis versus time is presented in Figure 7C. To estimate the speed of the back-and-forth oscillating movements, the mean speed of granules that mainly reflects processive movements were calculated over a period of 15 min. The average rate was 0.0736±0.0017 µm/s, whereas oscillatory movements displayed an average rate of 0.0156±0.0015 µm/s.


Figure 7
View larger version (40K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 7. GFP-FMRP granules movements in primary cultured neurons. (A) GFP-FMRP forms granules in 6 DIV transfected primary cultures of telencephalon neurons. (B) Two sets of granules highlighted in the insets (1 and 2) were studied and their movements in dendrites recorded during 140 s with 20 s intervals. Note that the time-lapse series corresponding to the inset 2 was taken from 300 s up to 440 s; therefore, the initial image of the time-lapse series presented in (B) is different from (A). (C) Distances traveled of the two granules from the start point traced by the arrowheads in panel (B) and their final moved distances along the x-axis (white circles) and the y-axis (black circles). (D) Histogram plotting maximal distances traveled by GFP-FMRP granules in 104 analyzed neurons.

 
Finally, we calculated the maximal distance covered by GFP-FMRP granules in 104 individual neurons. Five sub-populations of granules were defined by their traveled distance (0–99, 100–199, 200–299, 300–399 and ≥400 µm, respectively, Fig. 7D). More than 70% FMRP granules were found between 0 and 300 µm from the cell body, with 40% of granules traveling between 100 and 199 µm. Since the minimal traveled distance was estimated to be 74 µm, this indicates that GFP-FMRP granules transport does not result from passive diffusion from the soma, but are rather actively transported in the dendrites. The maximal distance was estimated to be 750 µm, indicating that granules are able to travel long-range distances.

KIF3C drives the transport of FMRP-containing granules in vivo
To validate the functional involvement of KIF3C in the transport of FMRP granules, we expressed GFP-FMRP alone or in the presence of RFP-KIF3C and of its dominant-negative form lacking the N-terminal motor domain (RFP-KIF3C{Delta}N). We hypothesized that this truncated version of KIF3C that still contains the stalk (dimerization domain) and tail domain (cargo binding domain) could be incorporated into traveling complexes. However, since this truncated version lacks a motor domain, it was expected that it was unable to power the transport of its cargo along microtubules, presumably by disrupting endogenous KIF3C motor activity.

RFP-KIF3C and RFP-KIF3C{Delta}N were expressed, together with GFP-FMRP, in cortical neurons. As shown in Figure 8A, RFP-KIF3C and RFP-KIF3C{Delta}N displayed a cytoplasmic granular pattern comparable to that observed for endogenous KIF3C (Fig. 3). Since time-lapse videomicroscopy experiments revealed that accumulation of KIF3C or KIF3C{Delta}N did not affect significantly the movements and speeds of FMRP granules (not shown), we therefore further analyzed the maximal distance traveled by FMRP granules (Fig. 8).


Figure 8
View larger version (34K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 8. Functional impact of a KIF3C dominant-negative mutant on the transport of GFP-FMRP granules. (A) Neurons (6 DIV) were transfected with both GFP-FMRP and RFP-KIF3C or GFP-FMRP and RFP-KIF3C{Delta}N, or GFP-FMRP alone. (B) The percentage of GFP-FMRP granules transported at different distances was plotted in each condition. Confocal analysis was performed on 104 (GFP-FMRP alone), 106 (GFP-FMRP plus RFP-KIF3C) and 112 (GFP-FMRP plus RFP-KIF3C{Delta}N) transfected neurons, respectively.

 
First, we calculated that the presence of the dominant-negative KIF3C{Delta}N significantly decreased the average maximal distance of granules to 191 µm (n = 112), when compared to GFP-FMRP either alone (250 µm, n = 104, P<0.0001) or together with KIF3C (215 µm, n = 106, P<0.036). The main sub-populations of GFP-FMRP granules that were affected by over-expression of KIF3C or its dominant negative KIF3C{Delta}N corresponded to granules traveling from 0 to 299 µm (Fig. 8B), representing 70% of GFP-FMRP granules (see above). Accumulation of KIF3C increased the sub-populations of granules between 100 and 199 µm (Fig. 8), while those present at 300 µm were reduced. On the other hand, the dominant-negative KIF3C{Delta}N strongly increased the number of FMRP granules proximal to the cell body (0–100 µm range) from 1 to 20%, while those present between 200–299 µm were reduced from 28 to 14% (Fig. 8B). In addition, FMRP granules in the 300–399 µm range and above were also diminished in the presence of KIF3C{Delta}N, indicating that this mutant alters long-range transport of FMRP granules (Fig. 8B). These data demonstrate in vivo the functional implication of the neurospecific kinesin KIF3C in the transport of RNA granules containing FMRP.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 FUNDING
 REFERENCES
 
Translation repression and transport of mRNA within neuronal RNA granules are thought to be tightly coupled processes, probably to prevent ectopic translation of dendritic mRNAs while ‘en route’ to the synapse. Several RNA-binding proteins present in RNA granules are thought to play a dual role in translation repression and transport. This dual role has been proposed for the RNA-binding protein CPEB that binds and regulates the synaptic translation of CamKIIa mRNA, while facilitating at the same time its dendritic transport, apparently via interaction with motor proteins (39), and also for the zipcode-binding protein ZPB1 that targets ß-actin mRNA into dendrites and axons (40). However, no evidence has been provided so far for a direct interaction between RNA-binding proteins with a motor protein. In a similar way, FMRP has been described as a translation repressor in vitro (41,42) and in vivo (37). Moreover, a role for FMRP in the dendritic localization of several mRNA has been suggested (12,43), but its direct role as an adaptor molecule between RNA granules and motor proteins has not been experimentally documented. In the present study, we provide evidence that FMRP and a member of the kinesin II family, the neurospecific kinesin KIF3C, interact directly in vitro and co-localize in granule-like structures in the somato-dendritic compartment of neurons in vivo. Moreover, we show that FMRP and KIF3C can be specifically recovered and co-immunoprecipitated from in vitro polymerized brain microtubules fractions. This direct interaction allows us to speculate that, in addition to regulate translation repression in RNA granules, FMRP behaves also as a molecular adaptor between RNA granules and KIF3C. We estimate that under the experimental conditions used, while the vast majority of FMRP-containing RNPs remains associated with polyribosomes, 5–15% of FMRP-RNPs are associated with microtubule structures, probably representing the fraction of FMRP traveling in RNA granules in the dendritic arborization. Whether one single FMRP molecule can bind its target mRNA and simultaneously interact with KIF3C is not known. Alternatively, we propose that a certain class of FMRP binds only mRNA, while other species behave as molecular adaptors with KIF3C. It should be recalled that FMRP forms homo- and heteromers throughout its protein–protein interacting domain (PPID) and that in such situation it is conceivable that one molecule binds RNA while the second links to the motor protein. It is worth noting that FXR1P, a homolog and interactor of FMRP, regulates the affinity of FMRP towards the G-quartet RNA structure (44). It is therefore possible that post-translational modifications, such as phosphorylation, are the factors that govern these functions.

KIF3C belongs to the family of anterograde motors kinesin II that contains also KIF3A and B. KIF3C is the neurospecific member, being mainly expressed in brain and retina (27,28). On the other hand, KIF3A and KIF3B are the most studied and are thought to function in fast axonal transport of several cargoes via interaction with the kinesin-associated protein KAP3, a non-motor accessory subunit that recognizes the cargoes (45,46). The heterodimer KIF3A/B complex transports vesicles containing fodrin (29) and Kv1 channels (47), as well as axonal RNA granules containing Tau mRNA (48). KIF3A/C complex is associated with axonal vesicles; however, a fraction of KIF3C does not associate with KIF3A, evoking that it could function as a homodimer (27,28). Our study show that, in contrast to KIF3A/B or KIF3A/C complexes which function in axonal targeting of vesicles via the adapter KAP3, the homodimer KIF3C actually transports dendritic RNA granules via direct interaction with FMRP that serves as a direct molecular adaptor between KIF3C and the RNA granules. Indeed, direct binding of cargoes to kinesin heavy chain is thought to drive the transport towards dendrites, whereas binding to kinesins via the accessory protein KAP or kinesin light chain KLC targets cargoes towards the axon (17,49).

Using videomicroscopy approaches, we observed that in primary cultured neurons, GFP-FMRP granules display oscillatory bidirectional movements over short distances as well as processive movements over longer distance. Oscillatory movements of GFP-FMRP granules have a mean speed of 0.0156±0.0015 µm/s, in line with the average speed calculated for GFP-Pur{alpha} granules (0.034±0.025 µm/s; 17). Processive movements have an average speed of 0.0736±0.0017 µm/s, compatible for granules labeled for BC1 untranslated RNA (0.07 µm/s; 50), Arc mRNA (0.08 µm/s; 51) or CaMKII mRNA (0.05±0.03 µm/s; 52). Moreover, the velocity of these RNA granules is comparable to that observed for GFP-Staufen (0.1 µm/s; 13) and GFP-Pur{alpha} granules (0.10–0.12 µm/s; 17), both RNA-binding proteins also detected in RNA granules. These processive movements correspond to long-range dendritic transport of GFP-FMRP RNA granules that is disrupted by overexpression of the dominant negative of KIF3C lacking its N-terminal motor domain, similarly to the dominant-negative KIF5C transporting GFP-Pur{alpha} granules (17). Accumulation of KIF3C{Delta}N had no effect on granules traveling from 0 to 299 µm away from the soma; however, it tended to impede transport more distally, probably due to interference with other motors required for this longer-range transport. The fact that this dominant-negative does not abrogate totally the distal transport of GFP-FMRP granules, suggests the involvement of additional different motors in the long-range transport of FMRP granules. This is in line with our immunofluorescence data showing that not all granules containing FMRP harbor KIF3C. Neuronal mRNA granules being highly heterogenous multimolecular complexes (13,15–17), it can be proposed that different motor proteins could transport specific subpopulations of FMRP-containing granules. Indeed, selective association of RNA granules with specific kinesin motors might provide a mechanism to target mRNAs selectively to axons or dendrites (49). The nature of these motors remains elusive, even though myosin Va, dynein (25), as well as KIF5 (17) were reported to co-immunoprecipitate with FMRP, Staufen and Pur{alpha}.

A large-scale analysis of the mRNA associated with FMRP in dendritic complexes (43) revealed that KIF3C mRNA is a potential target of FMRP. This raises the possibility that FMRP could address and modulate the local translation of KIF3C mRNA in dendrites and at the synapse, thereby controlling the movement of mRNA granules locally. Similar to Fmr1, the expression of KIF3C is strongly upregulated during embryonic development and in differentiating neuroblastomal cells (53), suggesting an important role for this motor protein at the onset of differentiation and throughout all the process of neurite elongation, perhaps in concert with FMRP which is essential for neuronal spines development and maturation (20,23). The KIF3C null mouse is apparently viable, reproduces and develops normally, without noticeable abnormalities (54), probably owing to the compensation of KIF3A and KIF3B. However, no complete anatomical and behavioral study has been performed yet on this mouse strain, and it would be worth searching whether KIF3C null mice display immature-looking abnormal dendritic spines such as Fmr1 null mice (7).

Our study provides evidence that, in certain sub-populations of RNA granules, FMRP interacts with the motor protein KIF3C that controls their targeting towards dendrites. These results provide new insights on FMRP’s functions and raise the possibility that the absence of FMRP in Fragile X patients might therefore not only impede the translation of neurospecific mRNA targets, but also alter the targeting and transport of dendritic mRNA. In the absence of FMRP, classes of RNA granules might remain in the soma, ‘orphan’ of the right adapter to link them to the motor KIF3C required for their dendritic transport. The discovery of an RNA-binding protein as a molecular adaptor between RNA granules and a member of the kinesin family opens new perspectives. This would provide a mechanism by which dendritic mRNA are sorted via specific interaction with RNA-binding proteins and the choice of a motor that targets the granules to specific distal locations. A great challenge resides in the characterization of RNA granules transported in dendrites, both at the RNA and protein level, which will help understanding part of the mechanisms underlying the regulation of dendritic RNA granules transport and local synaptic translation.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
FUNDING
 REFERENCES
 
Yeast two-hybrid screen and GST-pulldowns
Yeast-two-hybrid screening was performed as described previously (33). From ~2.8x106 clones screened, 78 positive colonies showed adenine and histidine prototrophy and ß-galactosidase activity. Among these, one colony carried a 2.3 kb insert. This insert corresponded to the C-terminal portion (amino acids 403–792) as well as a portion of the 3'-UTR of the cDNA of human neurospecific kinesin KIF3C.

Purification of recombinant fusion proteins and GST-pulldown assays
The pBSK(+) construct containing mouse KIF3C full-length cDNA and pGex-KG construct encoding GST-KIF3C 71 C-ter were provided by Zhaohuai Yang and Lawrence S. B. Goldstein (Howard Hughes Medical Institute, University of California, San Diego, CA, USA). The pGex-4T-1/KIF3C N-ter plasmid encoding GST-KIF3C amino acids 1–383 was generated by ligation of the PCR product amplified from the pBSK(+)/mouse KIF3C full-length cDNA (primers: 5'-GCC GGA TCC GCC AGT AAG ACC AAG GCC AG-3', 5'-TTC CTG GAA TTC CCT CAG CAG-3') into the BamHI/EcoRI site of pGex-4T-1 (Amersham, GE Healthcare). Expression and purification of recombinant proteins and GST-pulldown assays were performed as described (27,33).

Immunofluorescence on rat cerebellar sections and primary cultured neurons
Adult rat brain sections and 14 days in vitro (DIV) primary cultured rat hippocampal neurons were prepared and fixed as described (33,55). Affinity-purified anti-KIF3C antibody (27) was used at 1:100 dilution (overnight at 4°C). Immunodetection of FMRP with mAb1C3 (1:500, Chemicon) and ß-tubulin with mAbE7 (1:500, Developmental Studies Hybridoma Bank, University of Iowa) was performed as described previously (33,37). Samples were analyzed in epifluorescence or confocal microscopy with a Nikon Eclipse E800 microscope equipped with a Hammamatsu CCD camera.

Mouse brain microtubule preparation and subcellular fractionation
Microtubules were prepared from the 2-week-old mouse brain through two cycles of temperature-dependant cycles of microtubules polymerization/depolymerization in the presence of GTP, following a protocol adapted from Valle (56), previously used to study the association of FXR1P to testis microtubules (34). For subcellular fractionation studies, mice brains were homogenized in standard PEM buffer and clarified at 12 000 g to lead P1. The resulting supernatant was then incubated with or without 1 mM GTP (Sigma), at 4°C or at 37°C for 30 min, to induce polymerization of the microtubules. Polymerized microtubules were collected by centrifugation (30 min, 4°C) at 25 000 g to lead P2. The supernatant containing unpolymerized microtubules and soluble proteins was further centrifuged at 105 000 g for 2 h, to lead P3 and the final supernatant (Sup). All pellet fractions were re-suspended in a final volume of denaturing SDS–PAGE buffer equivalent to the starting volume used to homogenize the brain. Equal volumes of each fraction were loaded on a 11% SDS–PAGE and analyzed by western-blotting using mAb1C3 against FMRP, mAbE7 against ß-tubulin, anti-L7 antibody as described (34), SYP(D-4) against the synaptic vesicle marker synaptophysin (1:2500; Santa Cruz Biotechnology) and the ER marker GRP78/BiP (1:10000; Stressgen).

Immunoprecipitations using mouse brain microtubule-enriched preparations
Immunoprecipitations were performed using freshly polymerized brain microtubules re-suspended in cold NE-PEM buffer (PEM, supplemented with 0.5% NP40, 10 mM EDTA) and homogenized using a Teflon-glass potter. Homogenates were centrifuged at 15 000 g for 10 min to sediment insoluble material and the resulting supernatant was used for immunoprecipitation. The concentration of salts was then adjusted to 0.15, 0.3 or 0.4 M NaCl. Antibodies [5 µg of mAb1C3 and mAb7G-1 (35; obtained from the Developmental Studies Hybridoma Bank, University of Iowa) or 50 µl of affinity-purified anti-KIF3C antibody] were pre-incubated (overnight at 4°C) with 60 µl protein A/G beads (Calbiochem) in the presence of 0.1 mg/ml of each BSA, yeast tRNA, glycogen (Sigma) in NE-PEM buffer. Beads with bound antibodies were then washed twice with NE-PEM. Proteins were eluted from the beads with 60 µl SDS-buffer and heat denatured for 5 min. Approximately one third of the eluate was loaded on a 7.5% SDS–PAGE and proteins transferred onto nitrocellulose membrane were revealed using the appropriate antibodies.

Time-lapse video microscopy
For direct visualization of KIF3C in primary cultured neurons, mouse KIF3C was fused to the red fluorescent protein (RFP) by cloning by PCR (primers: 5'-GCC CTC GAG CGC CAG TAA GAC CAA GGC CAG-3', 5'-CGG GGA TCC GTC ATG GTC TAC CAC TGT TGC AG-3') its cDNA in the XhoI-BamHI sites of the CMV-promotor driven plasmid peRFP (BD Biosciences). A dominant negative form of KIF3C tagged with RFP (RFP-KIF3C{Delta}N) lacking its motor domain (amino acids 1–380) was also generated. For this purpose, the cDNA corresponding to its C-terminal domain (amino acids 383–796) was excised from peRFP/KIF3C using KIF3C cDNA EcoRI site and peRFP BamHI site, and subcloned into peRFP. To construct peGFP/FMRP, FMRP iso7 cDNA was subcloned from pGBKT7/FMRP iso7 (33) and transferred to the EcoRI/PstI sites of peGFP (BD Biosciences). Primary cultures of mouse telencephalon were carried out as described (57). At DIV5, neurons were transfected with the peGFP-FMRP, peRFP-KIF3C or peRFP-KIF3C{Delta}N vectors using Lipofectamine reagent according to the manufacturer’s instructions (Invitrogen). After 5 h post-transfection, the medium was changed and neurons were allowed to recover overnight before proceeding for analyses. To minimize accumulation of GFP-FMRP, we chose early time after transfection (16–20 h). This allowed detection of individual particles in dendrites and to study of their subsequent intracellular transport.

For time-lapse videomicroscopy, cells were maintained at 37°C in a 5% CO2 live-cell incubation chamber (Saur, Reutlingen, Germany). Images were taken with a cooled CCD camera (Cool Snap HQ, Roper Scientific) mounted on the Leica DM IRE2 microscope and a 100x oil immersion objective lens (N.A. 1.4, Zeiss). The cells were illuminated with a DG4 shutter. The intensities of fluorescence along the processes of each neuron were measured with a user-defined threshold with the MetaMorph software (Universal Imaging). Fluorescent images were captured every 5 s for 15 min. The mean movements of granules were measured on a 15 min scale using the SpotTracker plugging of the ImageJ software (Wayne Rasband, NIH, Bethesda, NY, USA). For analysis of the maximal transport distances of granules, neurons were fixed (see above) and a minimum of n = 100 transfected cells for each condition were observed under a Leica TCS SP2 AOBS confocal microscope. The maximal distance was calculated from the root of the dendrite to the farthest traveling granule of each cell, using the ImageJ software.

All animal procedures conformed to the guidelines of the Canadian or French Council for the care and use of laboratory animals, and all efforts were made to minimize animal suffering.


   FUNDING
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 FUNDING
REFERENCES
 
This work was supported by grants from the Canadian Institutes of Health Research (to E.W.K.), from CNRS-ATIP (to B.B.) and from ANR (to B.B. and M.S.).


   ACKNOWLEDGEMENTS
 
We are indebted to Lawrence S.B. Goldstein and Zhaohuai Yang (Howard Hughes Medical Institute, University of California San Diego, CA, USA) who provided us with KIF3C antiserum, as well as various constructs encoding KIF3C and KIF3B. We thank Francine Nault and Paul De Koninck for the neuronal primary cultures and Josune Olave for critical reading of the manuscript. M.S. and A.-M.L.-B. would like to thank Manuel Rojo, INSERM U582, for advices and Christophe Chamot and Tristan Piolot from the Imaging Center, Institut Jacques Monod for their assistance in videomicroscopy analysis. L.D. was a recipient of a postdoctoral fellowship from the FRAXA Research Foundation (USA) and presently of a Marie Curie Intra-European Fellowship.

Conflict of Interest statement. None declared.


   FOOTNOTES
 
{dagger} Present address: INSERM U567-CNRS UMR8104, Institut Cochin, Université René Descartes-Paris V, 75014 Paris, France Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 FUNDING
 REFERENCES
 

  1. Hagerman R.J., Hagerman P.J. Fragile X Syndrome: Diagnosis, Treatment and Research (2002) Baltimore, MA, USA: The Johns Hopkins University Press.

  2. Feng Y., Gutekunst C.A., Eberhart D.E., Yi H., Warren S.T., Hersch S.M. Fragile X mental retardation protein: nucleocytoplasmic shuttling and association with somatodendritic ribosomes. J. Neurosci. (1997) 17:1539–1547.[Abstract/Free Full Text]

  3. Khandjian E.W., Huot M.E., Tremblay S., Davidovic L., Mazroui R., Bardoni B. Biochemical evidence for the association of fragile X mental retardation protein with brain polyribosomal ribonucleoparticles. Proc. Natl Acad. Sci. USA (2004) 101:13357–13362.[Abstract/Free Full Text]

  4. Stefani G., Fraser C.E., Darnell J.C., Darnell R.B. Fragile X mental retardation protein is associated with translating polyribosomes in neuronal cells. J. Neurosci. (2004) 24:7272–7276.[Abstract/Free Full Text]

  5. Weiler I.J., Irwin S.A., Klintsova A.Y., Spencer C.M., Brazelton A.D., Miyashiro K., Comery T.A., Patel B., Eberwine J., Greenough W.T. Fragile X mental retardation protein is translated near synapses in response to neurotransmitter activation. Proc. Natl Acad. Sci. USA (1997) 94:5395–5400.[Abstract/Free Full Text]

  6. Ule J., Darnell R.B. RNA binding proteins and the regulation of neuronal synaptic plasticity. Curr. Opin. Neurobiol. (2006) 16:102–110.[CrossRef][Web of Science][Medline]

  7. Comery T.A., Harris J.B., Willems P.J., Oostra B.A., Irwin S.A., Weiler I.J., Greenough W.T. Abnormal dendritic spines in fragile X knockout mice: maturation and pruning deficits. Proc. Natl Acad. Sci. USA (1997) 94:5401–5404.[Abstract/Free Full Text]

  8. Hinton V.J., Brown W.T., Wisniewski K., Rudelli R.D. Analysis of neocortex in three males with the fragile X syndrome. Am. J. Med. Genet. (1991) 41:289–294.[CrossRef][Web of Science][Medline]

  9. Wilson B.M., Cox C.L. Absence of metabotropic glutamate receptor-mediated plasticity in the neocortex of fragile X mice. Proc. Natl Acad. Sci. USA (2007) 104:2454–2459.[Abstract/Free Full Text]

  10. Koekkoek S.K., Yamaguchi K., Milojkovic B.A., Dortland B.R., Ruigrok T.J., Maex R., De Graaf W., Smit A.E., VanderWerf F., Bakker C.E., et al. Deletion of FMR1 in Purkinje cells enhances parallel fiber LTD, enlarges spines, and attenuates cerebellar eyelid conditioning in Fragile X syndrome. Neuron (2005) 47:339–352.[CrossRef][Web of Science][Medline]

  11. Antar L.N., Afroz R., Dictenberg J.B., Carroll R.C., Bassell G.J. Metabotropic glutamate receptor activation regulates fragile X mental retardation protein and FMR1 mRNA localization differentially in dendrites and at synapses. J. Neurosci. (2004) 24:2648–2655.[Abstract/Free Full Text]

  12. Antar L.N., Dictenberg J.B., Plociniak M., Afroz R., Bassell G.J. Localization of FMRP-associated mRNA granules and requirement of microtubules for activity-dependent trafficking in hippocampal neurons. Genes Brain Behav. (2005) 4:350–359.[CrossRef][Web of Science][Medline]

  13. Kohrmann M., Luo M., Kaether C., DesGroseillers L., Dotti C.G., Kiebler M.A. Microtubule-dependent recruitment of Staufen-green fluorescent protein into large RNA-containing granules and subsequent dendritic transport in living hippocampal neurons. Mol. Biol. Cell. (1999) 10:2945–2953.[Abstract/Free Full Text]

  14. Kim K.C., Kim H.K. Role of Staufen in dendritic mRNA transport and its modulation. Neurosci. Lett. (2006) 397:48–52.[CrossRef][Web of Science][Medline]

  15. Krichevsky A.M., Kosik K.S. Neuronal RNA granules: a link between RNA localization and stimulation-dependent translation. Neuron (2001) 32:683–696.[CrossRef][Web of Science][Medline]

  16. Knowles R.B., Sabry J.H., Martone M.E., Deerinck T.J., Ellisman M.H., Bassell G.J., Kosik K.S. Translocation of RNA granules in living neurons. J. Neurosci. (1996) 16:7812–7820.[Abstract/Free Full Text]

  17. Kanai Y., Dohmae N., Hirokawa N. Kinesin transports RNA: isolation and characterization of an RNA-transporting granule. Neuron (2004) 43:513–525.[CrossRef][Web of Science][Medline]

  18. Aschrafi A., Cunningham B.A., Edelman G.M., Vanderklish P.W. The fragile X mental retardation protein and group I metabotropic glutamate receptors regulate levels of mRNA granules in brain. Proc. Natl Acad. Sci. USA (2005) 102:2180–2185.[Abstract/Free Full Text]

  19. Weiler I.J., Spangler C.C., Klintsova A.Y., Grossman A.W., Kim S.H., Bertaina-Anglade V., Khaliq H., de Vries F.E., Lambers F.A., Hatia F., et al. Fragile X mental retardation protein is necessary for neurotransmitter-activated protein translation at synapses. Proc. Natl Acad. Sci. USA (2004) 101:17504–17509.[Abstract/Free Full Text]

  20. Khandjian E.W., Bechara E., Davidovic L., Bardoni B. Fragile X mental retardation protein: many partners and multiple targets for a promiscuous function. Curr. Genomics (2005) 6:800–808.

  21. Zhang Y., O’Connor J.P., Siomi M.C., Srinivasan S., Dutra A., Nussbaum R.L., Dreyfuss G. The fragile X mental retardation syndrome protein interacts with novel homologs FXR1 and FXR2. EMBO J. (1995) 14:5358–5366.[Web of Science][Medline]

  22. Darnell J.C., Mostovetsky O., Darnell R.B. FMRP RNA targets: identification and validation. Genes Brain Behav. (2005) 4:341–349.[CrossRef][Web of Science][Medline]

  23. Bardoni B., Davidovic L., Bensaid M., Khandjian E.W. The fragile X syndrome: exploring its molecular basis and seeking a treatment. Expert Rev. Mol. Med. (2006) 8:1–16.[Medline]

  24. Ling S.C., Fahrner P.S., Greenough W.T., Gelfand V.I. Transport of Drosophila fragile X mental retardation protein-containing ribonucleoprotein granules by kinesin-1 and cytoplasmic dynein. Proc. Natl Acad. Sci. USA (2004) 101:17428–17433.[Abstract/Free Full Text]

  25. Ohashi S., Koike K., Omori A., Ichinose S., Ohara S., Kobayashi S., Sato T.A., Anzai K. Identification of mRNA/protein (mRNP) complexes containing Puralpha, mStaufen, fragile X protein, and myosin Va and their association with rough endoplasmic reticulum equipped with a kinesin motor. J. Biol. Chem. (2002) 277:37804–37810.[Abstract/Free Full Text]

  26. Sardella M., Navone F., Rocchi M., Rubartelli A., Viggiano L., Vignali G., Consalez G.G., Sitia R., Cabibbo A. KIF3C, a novel member of the kinesin superfamily: sequence, expression, and mapping to human chromosome 2 at 2p23. Genomics (1998) 47:405–408.[CrossRef][Web of Science][Medline]

  27. Yang Z., Goldstein L.S. Characterization of the KIF3C neural kinesin-like motor from mouse. Mol. Biol. Cell. (1998) 9:249–261.[Abstract/Free Full Text]

  28. Muresan V., Abramson T., Lyass A., Winter D., Porro E., Hong F., Chamberlin N.L., Schnapp B.J. KIF3C and KIF3A form a novel neuronal heteromeric kinesin that associates with membrane vesicles. Mol. Biol. Cell. (1998) 9:637–652.[Abstract/Free Full Text]

  29. Takeda S., Yamazaki H., Seog D.H., Kanai Y., Terada S., Hirokawa N. Kinesin superfamily protein 3 (KIF3) motor transports fodrin-associating vesicles important for neurite building. J. Cell. Biol. (2000) 148:1255–1265.[Abstract/Free Full Text]

  30. Bardoni B., Willemsen R., Weiler I.J., Schenck A., Severijnen L.A., Hindelang C., Lalli E., Mandel J.L. NUFIP1 (nuclear FMRP interacting protein 1) is a nucleocytoplasmic shuttling protein associated with active synaptoneurosomes. Exp. Cell. Res. (2003) 289:95–107.[CrossRef][Web of Science][Medline]

  31. Devys D., Lutz Y., Rouyer N., Bellocq J.P., Mandel J.L. The FMR-1 protein is cytoplasmic, most abundant in neurons and appears normal in carriers of a fragile X premutation. Nat. Genet. (1993) 4:335–340.[CrossRef][Web of Science][Medline]

  32. Kanai Y., Okada Y., Tanaka Y., Harada A., Terada S., Hirokawa N. KIF5C, a novel neuronal kinesin enriched in motor neurons. J. Neurosci. (2000) 20:6374–6384.[Abstract/Free Full Text]

  33. Davidovic L., Bechara E., Gravel M., Jaglin X.H., Tremblay S., Sik A., Bardoni B., Khandjian E.W. The nuclear microspherule protein 58 is a novel RNA-binding protein that interacts with fragile X mental retardation protein in polyribosomal mRNPs from neurons. Hum. Mol. Genet. (2006) 15:1525–1538.[Abstract/Free Full Text]

  34. Huot M.E., Mazroui R., Leclerc P., Khandjian E.W. Developmental expression of the fragile X-related 1 proteins in mouse testis: association with microtubule elements. Hum. Mol. Genet. (2001) 10:2803–2811.[Abstract/Free Full Text]

  35. Ceman S., Zhang F., Johnson T., Warren S.T. Development and characterization of antibodies that immunoprecipitate the FMR1 protein. Methods Mol. Biol. (2003) 217:345–354.[Medline]

  36. De Diego Otero Y., Severijnen L.A., van Cappellen G., Schrier M., Oostra B., Willemsen R. Transport of fragile X mental retardation protein via granules in neurites of PC12 cells. Mol. Cell. Biol. (2002) 22:8332–8341.[Abstract/Free Full Text]

  37. Mazroui R., Huot M.E., Tremblay S., Filion C., Labelle Y., Khandjian E.W. Trapping of messenger RNA by Fragile X Mental Retardation protein into cytoplasmic granules induces translation repression. Hum. Mol. Genet. (2002) 11:3007–3017.[Abstract/Free Full Text]

  38. Davidovic L., Tremblay S., Gravel M., De Koninck P., Khandjian E. Fragile X mental retardation protein: to be or not to be a translational repressor. In: The Molecular Basis of Fragile X Syndrome—Sung Y., Denman R., eds. (2005) Kerala, India: Research Signpost. 129–143.

  39. Huang Y.S., Carson J.H., Barbarese E., Richter J.D. Facilitation of dendritic mRNA transport by CPEB. Genes Dev. (2003) 17:638–653.[Abstract/Free Full Text]

  40. Tiruchinapalli D.M., Oleynikov Y., Kelic S., Shenoy S.M., Hartley A., Stanton P.K., Singer R.H., Bassell G.J. Activity-dependent trafficking and dynamic localization of zipcode binding protein 1 and beta-actin mRNA in dendrites and spines of hippocampal neurons. J. Neurosci. (2003) 23:3251–3261.[Abstract/Free Full Text]

  41. Laggerbauer B., Ostareck D., Keidel E.M., Ostareck-Lederer A., Fischer U. Evidence that fragile X mental retardation protein is a negative regulator of translation. Hum. Mol. Genet. (2001) 10:329–338.[Abstract/Free Full Text]

  42. Li Z., Zhang Y., Ku L., Wilkinson K.D., Warren S.T., Feng Y. The fragile X mental retardation protein inhibits translation via interacting with mRNA. Nucleic Acids Res. (2001) 29:2276–2283.[Abstract/Free Full Text]

  43. Miyashiro K.Y., Beckel-Mitchener A., Purk T.P., Becker K.G., Barret T., Liu L., Carbonetto S., Weiler I.J., Greenough W.T., Eberwine J. RNA cargoes associating with FMRP reveal deficits in cellular functioning in Fmr1 null mice. Neuron (2003) 37:417–431.[CrossRef][Web of Science][Medline]

  44. Bechara E., Davidovic L., Melko M., Bensaid M., Tremblay S., Grosgeorge J., Khandjian E.W., Lalli E., Bardoni B. Fragile X related protein 1 isoforms differentially modulate the affinity of fragile X mental retardation protein for G-quartet RNA structure. Nucleic Acids Res. (2007) 35:299–306.[Abstract/Free Full Text]

  45. Scholey J.M. Kinesin-II, a membrane traffic motor in axons, axonemes, and spindles. J. Cell. Biol. (1996) 133:1–4.[Free Full Text]

  46. Yamazaki H., Nakata T., Okada Y., Hirokawa N. Cloning and characterization of KAP3: a novel kinesin superfamily-associated protein of KIF3A/3B. Proc. Natl Acad. Sci. USA (1996) 93:8443–8448.[Abstract/Free Full Text]

  47. Gu C., Zhou W., Puthenveedu M.A., Xu M., Jan Y.N., Jan L.Y. The microtubule plus-end tracking protein EB1 is required for Kv1 voltage-gated K+ channel axonal targeting. Neuron (2006) 52:803–816.[CrossRef][Web of Science][Medline]

  48. Aronov S., Aranda G., Behar L., Ginzburg I. Visualization of translated tau protein in the axons of neuronal P19 cells and characterization of tau RNP granules. J. Cell Sci. (2002) 115:3817–3827.[Abstract/Free Full Text]

  49. Hirokawa N., Takemura R. Molecular motors and mechanisms of directional transport in neurons. Nat. Rev. Neurosci. (2005) 6:201–214.[CrossRef][Web of Science][Medline]

  50. Muslimov I.A., Santi E., Homel P., Perini S., Higgins D., Tiedge H. RNA transport in dendrites: a cis-acting targeting element is contained within neuronal BC1 RNA. J. Neurosci. (1997) 17:4722–4733.[Abstract/Free Full Text]

  51. Wallace C.S., Lyford G.L., Worley P.F., Steward O. Differential intracellular sorting of immediate early gene mRNAs depends on signals in the mRNA sequence. J. Neurosci. (1998) 18:26–35.[Abstract/Free Full Text]

  52. Rook M.S., Lu M., Kosik K.S. CaMKIIalpha 3' untranslated region-directed mRNA translocation in living neurons: visualization by GFP linkage. J. Neurosci. (2000) 20:6385–6393.[Abstract/Free Full Text]

  53. Navone F., Consalez G.G., Sardella M., Caspani E., Pozzoli O., Frassoni C., Morlacchi E., Sitia R., Sprocati T., Cabibbo A. Expression of KIF3C kinesin during neural development and in vitro neuronal differentiation. J. Neurochem. (2001) 77:741–753.[CrossRef][Web of Science][Medline]

  54. Yang Z., Roberts E.A., Goldstein L.S. Functional analysis of mouse kinesin motor Kif3C. Mol. Cell. Biol. (2001) 21:5306–5311.[Abstract/Free Full Text]

  55. Gonzalez-Billault C., Avila J., Caceres A. Evidence for the role of MAP1B in axon formation. Mol. Biol. Cell. (2001) 12:2087–2098.[Abstract/Free Full Text]

  56. Valle R.B. Reversible assembly purification of microtubules without assembly-promoting agents and further purification of tubulin, microtubule-associated proteins and MAP. Methods Enzymol. (1986) 134:89–104.[Web of Science][Medline]

  57. Khelfaoui M., Guimiot F., Simonneau M. Early neuronal and glial determination from mouse E10.5 telencephalon embryonic stem cells: an in vitro study. Neuroreport (2002) 13:1209–1214.[CrossRef][Web of Science][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 NeuroscientistHome page
B. E. Pfeiffer and K. M. Huber
The State of Synapses in Fragile X Syndrome
Neuroscientist, October 1, 2009; 15(5): 549 - 567.
[Abstract] [PDF]

Home page
 Cereb CortexHome page
L.-F. Qiu, T.-J. Lu, X.-L. Hu, Y.-H. Yi, W.-P. Liao, and Z.-Q. Xiong
Limbic Epileptogenesis in a Mouse Model of Fragile X Syndrome
Cereb Cortex, July 1, 2009; 19(7): 1504 - 1514.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Cell. Biol.Home page
M. Kim, M. Bellini, and S. Ceman
Fragile X Mental Retardation Protein FMRP Binds mRNAs in the Nucleus
Mol. Cell. Biol., January 1, 2009; 29(1): 214 - 228.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
16/24/3047    most recent
ddm263v2
ddm263v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (15)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Davidovic, L.
Right arrow Articles by Khandjian, E. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Davidovic, L.
Right arrow Articles by Khandjian, E. W.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?