SEARCH

SEARCH BY CITATION

Keywords:

  • dendrite;
  • nuclear localization;
  • ribonucleoprotein;
  • RNA granule;
  • RNA transport

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

In mammalian neurons, transport and translation of mRNA to individual potentiated synapses is believed to occur via a heterogeneous population of RNA granules. To identify components of Staufen2-containing granules, we used the yeast two-hybrid system. A mouse fetal cDNA library was screened with the N-terminal fragment of Staufen2 as bait. ZFR, a three zinc finger protein, was identified as an interacting protein. Confocal microscopy showed that ZFR, although mainly nuclear, was also found in the somatodendritic compartment of primary hippocampal neurons where it localized as granule-like structures. Co-localization with Staufen2 was observed in several granules. Biochemical analyses (immunoprecipitation, cell fractionation) further confirmed the ZFR/Staufen2 association. ZFR was shown to interact with at least the Staufen262 isoform, but not with Staufen1. ZFR also co-fractionated with ribosomes and Staufen259 and Staufen252 in a sucrose gradient. Interestingly, knockdown expression of ZFR through RNA interference in neurons relocated specifically the Staufen262, but not the Staufen259, isoform to the nucleus. Our results demonstrate that ZFR is a native component of Staufen2-containing granules and likely plays its role during early steps of RNA transport and localization. They also suggest that one of these roles may be linked to Staufen262-containing RNA granule formation in the nucleus and/or to their nucleo-cytoplasmic shuttling.

Abbreviations used
C

cytoplasmic

DIV

days in vitro

dsRBD

double-stranded RNA-binding domain

esiRNA

endoribonuclease-prepared siRNA

GFP

green fluorescent protein

MOI

multiplicity of infection

N

nuclear

NLS

nuclear localization signals

P0

ribosomal phosphoprotein

RNAi

RNA interference

RNP

ribonucleoprotein

shRNA

short hairpin RNA

siRNA

short interfering RNA

Stau2

Staufen2

SDS

sodium dodecyl sulfate

YFP

yellow fluorescent protein

In neurons, although most of the mRNAs are restricted to cell bodies, some are found in the somatodendritic compartment. As synthesis of proteins occurs in dendrites, dendritic localization of a specific subset of mRNAs and their local translation are thought to allow neurite growth and plasticity at sites distant from the cell body and the differential plasticity of each individual dendrite in response to neighbouring cells (Kiebler and DesGroseillers 2000; Steward and Worley 2001; Bassell and Kelic 2004; Huang and Richter 2004; Sutton and Schuman 2005). In dendrites, RNA and/or RNA-binding proteins have a punctuated distribution and their movement can be observed in living neurons using the RNA-binding dye SYTO14 (Knowles et al. 1996). These observations lead to a model in which differential delivery of new mRNAs occurs in motile structures called RNA granules. RNA granules are large clusters of ribosomes to which RNAs, translation factors and other proteins are associated (Krichevsky and Kosik 2001). Evidences suggest that they are translationally incompetent and represent reservoirs of silent RNA, which upon cell stimulation release RNAs for local translation on polyribosomes (Krichevsky and Kosik 2001).

Staufen, a double-stranded RNA binding protein plays central role(s) in mRNA transport and localization in Drosophila (St Johnston et al. 1991). It is also required for the de-repression of translation of oskar mRNA in oocytes (Kim-Ha et al. 1995; Micklem et al. 2000). In mammals, two genes coding for Staufen paralogues have been described (Kiebler et al. 1999; Marion et al. 1999; Wickham et al. 1999; Tang et al. 2001; Duchaine et al. 2002). In neurons, Staufen1 and Staufen2 (Stau1 and Stau2) are found in the somatodendritic compartment in independent punctuated structures that do not co-localize (Duchaine et al. 2002). They are both associated with a heterogeneous population of RNA granules and smaller ribosome-free RNA particles (Duchaine et al. 2002; Mallardo et al. 2003) and are believed to play important role(s) in the delivery of RNA to neurites (Kohrmann et al. 1999; Tang et al. 2001). Several isoforms of Stau2 in multiple ribonucleoprotein (RNP) complexes have been described but their respective physiological roles are still unclear (Duchaine et al. 2002; Mallardo et al. 2003; Thomas et al. 2005).

Proteomic approaches have recently been used to identify protein components of RNA granules (Ohashi et al. 2002; Bannai et al. 2004; Brendel et al. 2004; Kanai et al. 2004; Villace et al. 2004) but the role of identified proteins and their interaction within the granules is still unclear. Therefore, the heterogeneous composition of RNA granules and particles, their site of formation, the mechanism of recognition of specific RNAs and the mechanism of their transport/localization are still to be understood. Identification and characterization of mammalian Staufen binding protein(s) should provide more information on the mechanisms of RNA transport in neurons. Using the yeast two-hybrid assay, a first Stau2-binding protein was identified: the zinc finger protein ZFR. Biochemical characterization demonstrated that ZFR is associated with the Stau262-containing RNP complexes, but not with Stau1-containing granules. Down-regulation of ZFR by RNA interference caused the retention of the Stau262 isoform in the nucleus, suggesting that ZFR may play an early role in Stau2-granule formation.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Yeast two-hybrid screen

A cDNA fragment encoding the mouse Stau2 double-stranded RNA-binding domain (dsRBD)-1 and dsRBD2 (amino acids 1–179; GeneBank Accession No. AF459099) was subcloned into the pLexA plasmid (Clontech, Palo Alto, CA, USA), the resulting N-terminal fragment was used as bait to screen 2 × 107 clones from a foetal mouse brain cDNA library (Clontech). Positive clones were selected for both their growth on Leu medium and lacZ expression. All measurements were carried out according to Clontech protocols. Plasmids were rescued from 37 positive clones, and individual plasmids checked by back transformation with the bait. Sequence of the cDNA clones was determined with a LI-COR 4000L automated DNA sequence analyzer using end-labelled primers and a cycle-sequencing protocol (LI-COR, Lincoln, NE, USA). Thirteen unique sequences were identified, from which one encoded for the ZFR400−591 insert.

cDNA constructs

A cDNA encoding the pLexAD-Stau262-N fusion protein was generated by subcloning the PCR amplification product using the pcDNA3-RSV-Stau262 cDNA as template (Duchaine et al. 2002) and the Vent DNA polymerase (New England BioLabs, Mississauga, ON, Canada) into the EcoR1 and Xho1 sites of the pLexA plasmid. Sense and antisense primers were: 5′-TATAGAATTCAAAATGGCAAACCCCAAAGAG (underlined: EcoRI restriction site) and 5′-TATACTCGAGTGGTTTCTGAACTGGTTTGGGAAGCG (underlined: XhoI restriction site), respectively.

To generate pAD-Stau262-HA and Stau259-HA, the KpnI/XbaI fragments of pcDNA3-RSV/Stau262-HA and of pCDNA3-RSV/Stau259-HA, respectively (Duchaine et al. 2002) were isolated and their extremities blunted with T4 DNA polymerase. Similarly, pcDNA3-RSV/Stau1-HA was digested with HindIII and XbaI and treated with T4 DNA polymerase. Resulting fragments were subcloned at the EcoRV site of the pAdPS-CR5-IRES-green fluorescent protein (GFP) vector in which the gene of interest is co-expressed from the Cumate-Regulated promoter (CR5), along with GFP, through an internal ribosome entry site. The activity of CR5 is controlled by the hybrid transactivator cTA made of the DNA binding domain of the bacterial repressor, CymR, fused to VP16 activation domain. High-level expression was achieved by co-transfection of pAdPS-CR5-IRES-GFP encoding Stau262-HA, Stau259-HA or Stau1-HA with pAdCMV-cTA (Massie, B., Mullick, A., Lau P. and Konishi, Y. A system for regulated expression in eukaryotic cells. Patent pending).

For lentivirus production, pCSII-Stau262-HA, pCSII-Stau259-HA and pCSII-Stau155-HA vectors were generated by PCR amplification of the corresponding pcDNA3-RSV-Staufen cDNAs with the Vent DNA polymerase and by subcloning the resulting fragments into the AgeI and XhoI sites of the pCSII plasmid (gift from Dr I. Verma, the Salk Institute for Biological Sciences, La Jolla, CA, USA). Sense primers in Staufen sequences were: Stau1, 5′-TATAACCGGTACCATGGATAAGCCCGTGGACCCTCAC; Stau259, 5′-TATAACCGGTACCATGGATGCTTCAGATAAATCAGATG; Stau262, 5′-TATAACCGGTACCATGGCAAACCCCAAAGAGAAAACTCC (underlined: AgeI restriction site) and a common antisense primer in pcDNA3-RSV, 5′-GGTGACACTATAGAATAGGGCCCTCTAGATG. An XhoI restriction site is present within the common amplified vector sequence.

For antibody production, the ZFR400−596 encoding insert was subcloned in pGEX-4T2 plasmid, using EcoR1 and XhoI restriction sites, and the resulting fusion protein was induced and purified as indicated by the manufacturer (Amersham Pharmacia Biotech, Piscataway, NJ, USA) and injected in mice as described (Wickham et al. 1999).

Cell culture and transfection

Primary hippocampal and cortical neurons were collected from E17-E19 Sprague Dawley rat embryos (Charles River Laboratories, Ottawa, ON, Canada) and cultured at low or high density as described (Duchaine et al. 2002). Cells were plated on poly-d-lysine (Sigma, Oakville, ON, Canada) -coated glass coverslips (18 CIR-1D, Fisher, Nepean, ON, Canada) at 800 cells/mm2 according to standard protocols (Banker and Goslin 1988). Culture media consisted of neurobasal medium supplemented with 1 × GlutaMax, 25 μm glutamate and 2% B27 supplements (Invitrogen Life Technologies, Burlington, ON, Canada). On day 3–4 after plating, one-half of the culture media was removed and replaced with fresh media without glutamate. On day 5 after plating, neurons were transfected using 2 μg/mL Lipofectamine 2000 and 1 μg/mL plasmid DNA in fresh medium during 2 h as described by the manufacturer (Invitrogen Life Technologies). Following transfection, neurons were put back into conditioned media for 48 h. Less than 0.1% of initially plated cells were transfected. Neurons were fixed in 4% PFA/4% sucrose in phosphate-buffered saline (PBS) for 15 min, quenched three times for 5 min in PBS/glycine 0.1 m and finally washed in PBS. All plasmid DNA used for transfection were CsCl double-banded purified.

Neuron induction with lentiviral particles

Viral particles were obtained as described (Follenzi and Naldini 2002). Briefly, 293SF packaging cells (Broussau, S., Jabbour, N., Mullick, A., Durocher, Y., Gilbert, R., and Massie, B. Development of stable packaging cell lines for the production of cSIN lentiviral vectors. Manuscript in preparation) were transfected with a vector encoding the protein of interest. Thirty-six and 60 h post-transfection, viral particles were collected and concentrated by ultracentrifugation. Concentration of infectious virus was estimated by real-time PCR [LightCycler FastStart DNA Master SYBR Green I (Roche Molecular Biochemicals, Indianapolis, IN, USA)] measuring the amount of integrated pro-virus. Primary rat cortical neurons cultured at high density (10 days in vitro; DIV) were infected with the virus (multiplicity of infection (MOI) = 0.25) in conditioned media; the medium was changed with conditioned media 24 h after the addition of the virus, and the cells were allowed to express the protein of interest for at least 96 h before experiments.

RNA interference (RNAi)

RNAi experiments were conducted with either random DICER-endoribonuclease-prepared siRNA (esiRNA), short hairpin RNA (shRNA)-encoding PCR products or short interfering RNA (siRNA). Random esiRNA were generated from in vitro digestion of ZFR400−596 dsRNA with a recombinant human DICER protein according to the manufacturer instructions (DICER siRNA generation kit, Gene Therapy Systems, San Diego, CA, USA). ZFR400−596 dsRNA was obtained by in vitro transcription using the T7 RNA polymerase (Fermentas Life Sciences, Burlington, ON, Canada) and the Qiaquick (Qiagen, Valencia, CA, USA) purified T7-ZFR400−596-T7 PCR product as template. The T7-ZFR400−596-T7 PCR product was obtained by PCR amplification (20 cycles, 30 s at 95°C, 30 s at 55°C and 1 min at 72°C) using the Vent DNA polymerase, the B42AD-ZFR400−596 cDNA as template and the sense and antisense primers: 5′-GCGTAATACGACTCACTATAGGGAGACTCTCCTTCAAGTATTGGC and 5′-GCGTAATACGACTCACTATAGGGAGAGAAGTCTATGTCTTCGTCC (underlined: T7 promoter sequence), respectively. Neurons were transfected with 2 μg/mL Lipofectamine 2000 using 0.5 μm esiRNA.

Short hairpin RNAs (shRNA) directed against the rat and mouse ZFR mRNA were produced in cells following transfection of an shRNA-encoding PCR product under the control of the U6 snRNA promoter (Paddison et al. 2004). Three mouse ZFR shRNA PCR products were generated with the Taq DNA polymerase (Applied Biosystems Canada, Streetsville, ON, Canada) using the human U6 locus-pGEM1 plasmid (Promega, Madison, WI, USA) as template, a common sense primer 5′GGTGTTTCGTCCTTTCCACAA-3′ and three antisense primers, respectively. shRNA I: 5′-AAAAAAGGCCGCTGCCGCCGAACTCTGAACCCTCACAAGCTTCTGAGAGTTCAGAGTTCAGCAGCAGCAGCCGGTGTTTCGTCCTTTCCACAA-3′; shRNA II: 5′AAAAAAGTGCATCTGACGGAACGCAAGAAGCCTCACAAGCTTCTGAGACTCCTTGCATTCCGTCAGATACACGGTGTTTCGTCCTTTCCACAA-3′; and shRNA III: 5′AAAAAAGATAGTACCACAGTAACTGCAACTACCACCAAGCTTCGTGGCAGCTGCGGCTACTGTGGTACCATCGGTGTTTCGTCCTTTCCACAA-3′ (underlined: U6 complementary sequence). Neuroblastoma N2A cell lines were transfected with 2 μg/mL Lipofecteamine 2000 and 200 ng/mL of shRNA PCR product (gel purified and extracted with Qiaex II (Qiagen).

siRNA was designed against the rat and mouse ZFR mRNA. Sense 5′-CUCCUUGCAUUCCGUCAGAUACA-3′ and antisense 5′UGUAUCUGACGGAAUGCAAGGAG-3′. RNAs were annealed as indicated by the manufacturer (Integrated DNA Technologies, Coralville, IA, USA). siRNA (10 nm) was transfected in N2A or rat hippocampal neurons (5 DIV) using Lipofectamine 2000 as in Krichevsky and Kosik (2002).

Cell cytoplasmic extract

Cell cytoplasmic (C) and nuclear (N) extracts were prepared as described (Duchaine et al. 2002; Robb et al. 2005) with the following modifications. Cultures (12–15 DIV) of E17-E19 rat embryo cortical neurons were washed with ice-cold PBS and directly scraped off the dishes with ice-cold low-salt buffer (10 mm Hepes-KOH, pH 7.5, 10 mm KOAc, 2 mm MgOAc, 1 mm DTT, 12.5% sucrose) supplemented with EDTA-free Complete protease inhibitor cocktail (Roche Applied Science, Laval, QC, Canada). Cells were broken on ice by two sets of 15 strokes in a 23-gauge syringe followed by centrifugation at 1500 g for 10 min generating pellet P1 and supernatant S1. Pellet P1 was re-suspended in nucleus buffer (20 mm Hepes-KOH, pH 7.5, 10 mm NaCl, 1 mm MgCl2, 1 mm DTT, 0.3% Triton X-100), nuclei were further purified with two centrifugation through a 30% sucrose cushion and nuclear extracts (N) were obtained as described (Robb et al. 2005). Supernatants S1 were adjusted to 100 mm KCl and allowed to stand on ice for 30 min followed by a centrifugation at 10 000 g for 15 min at 4°C generating post-mitochondrial cytoplasmic extracts (C). Protein content was determined by the Bio-Rad protein assay according to manufacturer's instructions (Bio-Rad, Mississauga, ON, Canada).

Sedimentation studies

Sucrose sedimentation studies were performed as described previously (Krichevsky and Kosik 2001; Duchaine et al. 2002). Briefly, cytoplasmic extracts were loaded on a discontinuous 20–60% sucrose gradient and ultracentrifuged at 175 000 g in a SW41ti rotor at 4°C for 3 h. Resulting fractions (0.8 mL) were hand collected. Proteins were acetone precipitated and loaded onto a sodium dodecyl sulfate (SDS) – polyacrylamide gel and assayed by western blotting.

Western blotting and immunoprecipitation

Western blotting was carried out as previously described (Wickham et al. 1999). Antibodies used were mouse polyclonal anti-ZFR, affinity immunopurified rabbit polyclonal anti-Stau1 and anti-Stau2 (Duchaine et al. 2002), anti-calnexin (1 : 1000, StressGen, Vancouver, BC, Canada), mouse polyclonal anti-L7/SPA (1 : 10000, Novus Biologicals, Littleton, CO, USA), mouse anti-H1/AE4 (1 : 1000, Abcam, Cambridge, MA, USA), human autoimmune anti-ribosomal protein P (generous gift from Dr M. Reichlin, Oklahoma Medical Research Foundation, USA), rabbit polyclonal and mouse monoclonal anti-Stau2 (Duchaine et al. 2002) and mouse monoclonal anti-Stau1 antibodies (Dugre-Brisson et al. 2005).

For immunoprecipitation, Stau2-GST and Stau1-GST fusion proteins were first cross-linked to sepharose resin with DMP (Pierce Biotechnology–MJS Biolynx, Brockville, ON, Canada) (Bar-Peled and Raikhel 1996) and then used to immunopurify rabbit polyclonal anti-Stau2 and anti-Stau1 sera, respectively. Cytoplasmic extracts (1 mg of total proteins) from 12 to 15 DIV rat embryonic E17-19 cortex cell cultures were diluted in 0.5 mL lysis buffer supplemented with 0.1% Nonidet-P40 (immunoprecipitation buffer). Rnase treatment was carried with 300 U/mL Miccrococal Rnase (Fermentas, Burlington, ON, Canada) and 2 mm CaCl2 for 30 min on ice and stopped with 5 mm EGTA. Primary incubation with 10 μg immunopurified antibodies was carried out at 4°C for 4 h. Then, 50 μL of a 50% sepharose-A slurry was added and the solution was further incubated at 4°C for 1 h. The resin was washed 4 × 15 min with immunoprecipitation buffer at 4°C and the precipitate eluted with one volume of 2 × SDS loading buffer. Eluted complexes were loaded onto a 9% SDS–polyacrylamide gel and analyzed by western blotting.

Immunofluorescence

Immunocytochemistry was carried out essentially as described previously (Duchaine et al. 2002) using immunopurified rabbit (a kind gift of Dr R. E. Braun, Washington, USA) or mouse polyclonal anti-ZFR, mouse monoclonal anti-HA (12CA5), mouse monoclonal anti-Stau2 and rabbit immunopurified polyclonal anti-Stau2 antibodies (0.5 μg/mL). Secondary antibodies were anti-mouse and anti-rabbit IgG Alexa 594 and Alexa 647, respectively (Molecular Probes, Burlington, ON, Canada).

Image acquisition and analysis

For confocal images, cells were observed under a laser scanning confocal microscope (Zeiss LSM 510, Carl Zeiss, Jena, Germany) using a Zeiss Plan-Apo 63 × 1.40 NA oil immersion objective lens. Yellow fluorescent protein (YFP) was excited with a 488-nm argon laser and detected with a 505–535 band pass filter. Alexa 594 was excited with a 543-nm helium/neon laser and detected with a 560-nm band pass filter. Alexa 647 was excited with a 633-nm helium/neon and detected with a 650 long pass filter. Fifteen to twenty 0.2-μm optical sections of individual cells were taken from the bottom to the top and projected onto one plane. Then, 3-D reconstruction images were obtained from all optical sections using Zeiss LSM Image Browser software 3.2 (Carl Zeiss). For the co-localization analysis, five neurons were analyzed, adding up approximately 500 granules of each kind per subcellular region. One 0.2-μm optical section from each soma was analysed, whereas normal fluorescence was used for dendrites. Staining with two different anti-Staufen2 antibodies (mouse monoclonal and immunopurified rabbit polyclonal antibodies) and two different anti-ZFR antibodies (rabbit and mouse sera) yielded consistent results.

For epifluorescence images, cells were observed under a Nikon TE2000U using a Plan-Apo 60 × 1.40 NA oil immersion objective with a 1.5 × lens and a CoolSnap HQ (CCD 12-bit) camera (Roper Scientific, Trenton, NJ, USA). To clear up background, images were submitted to a nearest neighbour deconvolution protocol using the MetaMorph software (Universal Imaging Corporation, Downingtown, PA, USA).

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Identification of ZFR as a Stau2-binding protein

As a means to identify and characterize putative protein partners of Stau2 that may associate in Stau2-containing mRNP in vivo, we performed a yeast two-hybrid screen. To increase the probability of identifying proteins involved in the mechanism of RNA transport, we fused the N-terminal dsRBD1 and dsRBD2 of Stau262 (Stau262-N) to the lexA DNA-binding domain (lexAD) (Fig. 1a) and used it as a bait in the yeast assay. We started the screening with the N-terminal fragment of Stau2 to increase the probability of detecting protein partners involved in the cytoskeleton transport of Stau2/mRNA complexes as reported for Drosophila Staufen (Micklem et al. 2000). In addition, we removed the RNA-binding domains (dsRBD3 and -4) to prevent selection of clones through RNA bridging. Screening of 2 × 107 clones of a foetal mouse brain cDNA library fused to the B42 transcription activation domain with this Stau2 N-terminal bait, 37 positive clones were identified and sequenced. These clones contained 13 different open reading frames of mRNAs, of which one contained a 588-nucleotide long insert corresponding to 196 amino acids located in the central region of the zinc finger protein ZFR (Fig. 1b). ZFR is a protein with three spaced C2H2 zinc fingers and a C-terminal 316-amino-acid domain, DZF, which is conserved among a small class of double stranded RNA-binding proteins (Meagher et al. 1999). The Stau2-interacting fragment (ZFR400−596) contains the third zinc finger. The Zfr gene is conserved in Drosophila melanogaster, Caenorhabditis elegans and Homo sapiens (Meagher et al. 1999; Kleines et al. 2001).

image

Figure 1. ZFR400−596 interacts in yeast with the N-terminal fragment of Stau2. (a) Schematic representations of Stau2 and of its N-terminal fragment. Stau2 (GeneBank Accession No. AF459099) contains four dsRBDs (boxes) and a putative tubulin binding domain (oval). dsRBD3 is the major dsRNA-binding domain (black box). The N-terminal fragment of Stau2 (Stau262-N), fused to the DNA binding domain of LexA and used as bait in a yeast two-hybrid assay, starts at the ATG codon and contains Stau262 dsRBD1 and dsRBD2. (b) Schematic representations of ZFR and of its Stau2-binding domain. ZFR (GenBank accession no. NM_011767.1) contains three zinc fingers (ovals), a DZF domain conserved among several dsRNA binding proteins (white box), putative NLS (dark stars) and nuclear export signal (white star). ZFR400−596 was isolated in the yeast assay for its ability to interact with Stau262-N, therefore inducing yeast survival and β-Gal activity. ZFR400−596 encodes the central portion of ZFR that encompasses amino acids 400–596 and contains the third zinc finger. (c) Results of the yeast survival assay on growth restrictive plates (leucine) using different combinations of baits and preys.

Download figure to PowerPoint

ZFR400−596 was tested for specificity. We showed that lexAD-Stau262-N and B42AD-ZFR400−596 did not interact with the B42AD domain alone or the lexAD domain alone, respectively, in the yeast assay (Fig. 1c). Similarly, we showed that B42AD-ZFR400−596 did not interact with other unrelated proteins such as the Drosophila bicoid (LexAD-Bicoid).

ZFR co-localizes with cytoplasmic Stau2 granules

ZFR was described as a nuclear protein in testis and ovary (Meagher et al. 1999). Stau2 was shown to be predominantly present in large granules in the cytoplasm of the cell body and dendrites of mature hippocampal neurons (Tang et al. 2001; Duchaine et al. 2002). Recently, Stau2 was also shown to shuttle in the nucleus (Macchi et al. 2004; Miki and Yoneda 2004; Monshausen et al. 2004). To determine the subcellular distribution of ZFR in neurons and get a first clue on potential site(s) of interaction with Stau2, we double-stained embryonic rat hippocampal neurons with anti-ZFR and anti-Stau2 antibodies and analyzed their subcellular distribution by confocal microscopy. In these cells, as observed before in other tissues, ZFR distribution was mainly nuclear (Fig. 2a). However, particulate staining was also seen in the cell body (Fig. 2b), showing that a subpopulation of ZFR is cytosolic. In this compartment, Stau2 was also found as granule-like structures (Fig. 2b). Stau2/ZFR co-localization was evident in the cytoplasm (Fig. 2b) and in dendrites (Fig. 2c), although not all Stau2-containing granules were labelled with ZFR and not all ZFR-containing granules were labelled with Stau2 (see Table 1).

image

Figure 2. ZFR and Stau2 form somato-dendritic granule-like structures. Rat embryonic hippocampal neurons were grown in culture for 15 days (15 DIV), fixed and labelled with immunopurified rabbit anti-Stau2 (left panels; red) or mouse anti-ZFR (middle panels; green) antibodies and observed under a confocal microscope. The superposition of both red and green signals is demonstrated by a yellow coloration (right panels). (a) ZFR is mainly present in the nucleus but is also visible as granule-like structures in the cytoplasm. Scale bar, 5 μm. (b) Zoomed images of (a) panels showing ZFR cytoplasmic granules that partially co-localize with Stau2 granules (arrowheads). Scale bar, 10 μm. (c) Dendrites also show some ZFR and Stau2 granule co-localization (arrowheads). Scale bar, 5 μm.

Download figure to PowerPoint

Table 1.  Co-localization of ZFR and Stau2 in the somatodendritic compartment of hippocampal neurons in culture
 ZFR/Stau2Stau2/ZFR
  1. Co-localization is expressed as the percentage of double-stained granules relative to the total number of granules positive for a single marker. Data included quantification in soma, proximal dendrites (10–50 μm from cell body) and distal dendrites (60–100 μm from cell body).

Soma34 ± 629 ± 5
Proximal21 ± 927 ± 12
Distal14 ± 1214 ± 11

ZFR cofractionates with Stau2 isoforms

To determine whether ZFR also associates with large complexes, we first prepared cytoplasmic and nuclear fractions from high-density primary cultures of embryonic rat cortical neurons and demonstrated by western blot analysis that, although mostly nuclear, a significant amount of ZFR can be detected in the cytoplasmic fraction, as observed for Stau2 (Fig. 3a). Then, cytoplasmic extracts from high-density primary cultures of rat cortical neurons were fractionated in a sucrose gradient to determine which isoform might be a putative partner of ZFR. We previously described three Stau2 isoforms that segregated in different complexes on a sucrose density gradient: Stau262 fractionated with ribosome-free light RNPs whereas Stau259 and Stau252 isoforms co-fractionated with both ribosomes and complexes of higher density (Duchaine et al. 2002). Fractions from a 20–60% sucrose density gradient were recovered and analyzed by western blotting using anti-ZFR, anti-Stau2, anti-calnexin and anti-ribosomal protein ribosomal phosphoprotein (P0) antibodies. As seen in Fig. 3(b), ZFR fractionated in both the ribosome-free fractions, as did Stau262, and in fractions containing ribosomes as observed for Stau259/Stau252 (Duchaine et al. 2002).

image

Figure 3. ZFR is associated with distinct cytoplasmic Stau2 mRNP complexes. Rat cortical neurons 12–15 DIV were lysed and homogenized. (a) Cytoplasmic (C) and nuclear (N) extracts were prepared and equal amounts of protein extracts were probed by western blotting with mouse monoclonal (panels 1 and 4) or rabbit immunopurified polyclonal anti-Stau2 (panel 2), mouse anti-ZFR (panels 3 and 4), anti-ribosomal protein P0 (panel 4), anti-calnexin (CNX, panel 4) and anti-histone H1 (panel 4) antibodies, as indicated. (b) Cytoplasmic extract was prepared and loaded on a discontinuous 20–60% sucrose gradient. Fractions were collected and probed with mouse anti-ZFR, monoclonal anti-Stau2, anti-calnexin or anti-P0 antibodies, as indicated. I, mRNP; II, cytoplasmic ribosomes; III, endoplasmic reticulum.

Download figure to PowerPoint

ZFR co-immunopurifies with Stau262-containing RNP

Cytoplasmic extracts from high-density primary cultures of rat cortical neurons were immunoprecipitated with different immunopurified sera. Resulting immunoprecipitated products were analyzed by western blotting. When anti-Stau2 antibodies were used for immunoprecipitation, a specific band corresponding to ZFR was clearly visible on the western blot, suggesting that both proteins are present in common cytoplasmic complexes (Fig. 4a). In contrast, pre-immune serum was unable to immunoprecipitate ZFR (Fig. 4a). The Stau2/ZFR interaction was resistant to Rnase treatments before immunoprecipitation, suggesting that RNA bridging may not be necessary to maintain ZFR/Stau2 mRNP complexes.

image

Figure 4. Co-immunoprecipitation of ZFR with Stau262 and ribosomes. For immunoprecipitation, cytoplasmic extracts prepared from 12 to 15 DIV rat cortical neuron were incubated with (+) or without (–) Rnase. (a) Immunopurified rabbit anti-Stau2 or anti-Stau1 antibodies and control rabbit pre-immune serum (PI) were used for immunoprecipitation. Co-immunopurified proteins were analyzed by western blotting with mouse anti-ZFR, mouse monoclonal anti-Stau2 and mouse monoclonal anti-Stau1 antibodies. Asterisks point to IgG. (b) Normal human (NHS) or autoimmune anti-ribosomal P0 sera were used for immunoprecipitation. Pellets were analyzed by western blotting with immunopurified rabbit anti-ZFR and mouse anti-L7 antibodies. Asterisks point to IgG. (c) HA-tagged Stau1, Stau259 and Stau262 expressing-lentiviruses were tranduced in 10 DIV rat cortical neuron. Fourteen days in vitro cytoplasmic extracts were used for HA-immunoprecipitation with agarose-linked rat anti-HA antibodies. Co-immunopurified proteins were analyzed by western blotting with mouse anti-ZFR and mouse monoclonal anti-HA. Asterisks point to non-specific staining.

Download figure to PowerPoint

Immunoprecipitations were repeated with a human autoimmune anti-ribosomal protein P serum or, as control, a normal human serum. ZFR was clearly present in the immunoprecipitate obtained with the anti-ribosomal antibodies, and this association was Rnase resistant (Fig. 4b). ZFR was not detected in precipitate obtained with the normal serum. These results confirm the association between at least a fraction of ZFR and ribosomes and show that this interaction is not likely to be dependent on an RNA ligand.

To further investigate whether ZFR selectively associates with a Stau2 isoform, we used a lentivirus-based system to express HA-tagged Stau2 isoforms in primary cultures of rat cortical neurons (Fig. 4c). Cytoplasmic extracts of transduced neurons were immunoprecipitated with an anti-HA antibody covalently linked to an agarose resin and resulting immunoprecipitated products were analyzed by western blotting. ZFR was detected in association with Stau262-HA complexes but not with Stau259-HA (Fig. 4c).

ZFR does not associate with Stau1

Immunopurified polyclonal anti-Stau1 antibodies were used to determine whether ZFR and endogenous Stau1 are present in the same granules. ZFR was never observed in immunoprecipitates obtained with anti-Stau1 antibodies (Fig. 4a). Similarly, when Stau155-HA was overexpressed in neurons following infection with lentivirus, ZFR was not detected in the Stau155-HA complexes either (Fig. 4c). Taken together, these results show that ZFR and Stau1 are not components of the same complexes.

ZFR is required for Stau262 nuclear export

To uncover a functional meaning for the Stau262/ZFR interaction, RNA interference (RNAi) approaches were developed. Three techniques aimed to induce specific Zfr expression knockdown were used (Fig. 5a). First, recombinant human DICER protein was incubated with a dsRNA substrate corresponding to ZFR400−596 to generate a heterogeneous population of small interfering RNAs (esiRNAs). Second, a PCR-based approach was used to generate a short hairpin interfering RNAs (shRNA) under the control of the U6 promoter. Third, a specific siRNA complementary to the ZFR mRNA was synthesized. The neuronal cell line N2A was used to assess RNAi efficiency using western blot against endogenous ZFR to bypass poor neuron transfection efficiency. When introduced in N2A esiRNAs (Fig. 5b), shRNA III (Fig. 5c) and siRNA (Fig. 5d) significantly reduced the level of ZFR protein. In contrast, shRNA I and shRNA II had no effect (Fig. 5c).

image

Figure 5. Down-regulation of ZFR expression by RNAi. (a) Schematic representation of ZFR and of silencing RNAs. RNAi-inducing molecules are positioned below ZFR mRNA. Position in nucleotides: shRNA I, 23–51; shRNA II, 1919–47; shRNA III, 767–795; esiRNA, 1200–1788; siRNA, 3116–3138. Symbols are as described in the legend to Fig. 1. (b, c, d) The neuronal N2A cell line was transfected with RNAi-inducing molecules, as indicated. Forty-eight hours post-transfection, cells were lysed and ZFR expression was analyzed by western blotting with mouse anti-ZFR and normalized to the expression of endogenous calnexin (CNX). Cells were transfected with increasing concentrations of the esiRNA cocktail generated by in vitro digestion of the ZFR dsRNA with DICER (b), 200 ng/mL of PCR product encoding shRNA I, shRNA II or shRNA III (c) or 10 nm of siRNA (d). Expression of ZFR, after normalization with calnexin expression, is indicated below the blots (%).

Download figure to PowerPoint

Rat embryonic hippocampal neurons were transfected with either a PCR-amplified DNA that expressed shRNA III, the esiRNA cocktail or the siRNA (Fig. 6). Co-transfection of an eYFP-encoding plasmid allowed us to identify transfected neurons. Two days post-transfection, cells were fixed and stained with anti-Stau2 antibodies. As seen in Fig. 6, transfection of DNA that induced the expression of shRNA III partially re-localized endogenous Stau2 to the nucleus (for 19 out of 21 eYFP-expressing neurons). Similar results were obtained following transfection of the DICER-generated esiRNA cocktail or the siRNA. As control, expression of eYFP alone (Fig. 6, upper panels) or co-transfection of the shRNA I or II (not shown) did not affect the subcellular distribution of Stau2. Therefore, nuclear shuttling and/or exit of Stau2 from the nucleus was impaired in neurons that under-expressed ZFR.

image

Figure 6. Nuclear accumulation of endogenous Stau2 following down-regulation of ZFR by RNAi. Embryonic rat hippocampal neurons were co-transfected at 5 DIV with a cDNA encoding YFP as a marker for transfected cells and either the DNA coding for shRNA, the esiRNA cocktail or the siRNA. Neurons were fixed 48 h post-transfection, observed for eYFP autofluorescence (left panels) and stained with the monoclonal anti-Stau2 antibodies to detect endogenous Stau2 (right panels) and observed under a confocal microscope. Expression of eYFP alone did not affect the distribution of endogenous Stau2. The shRNA III molecules, the DICER-generated esiRNA cocktail and the siRNA induced nuclear accumulation of endogenous Stau2. shRNA I and shRNA II had no effect (data not shown). Scale bar, 10 μm.

Download figure to PowerPoint

To determine which Stau2 isoform(s) accumulated in the nucleus following down-regulation of ZFR, Stau262-HA- and Stau259-HA-expressing cDNAs were transfected in neurons either alone or in the presence of shRNA III. Their subcellular distribution was followed 2 days later by immunofluorescence using anti-HA antibodies. When expressed alone, both Stau262-HA and Stau259-HA showed typical cytoplasmic staining with occasional nuclear granule-like labelling (Figs 7b and c, and Table 2). Acquisition parameters were set so as to clearly visualize the subcellular distribution of Stau2-HA in the nucleus and cytoplasm. These parameters do not allow observation of dendritic staining of transfected neurons. When cDNA coding for Stau262-HA was co-transfected with shRNA III, there was a significant increase in the percentage of cells showing nuclear labelling (Fig. 7f and Table 2), suggesting a critical role for ZFR in a putative nucleo-cytoplasmic shuttling of Stau262. In contrast, down-regulation of ZFR did not redistribute Stau259-HA that was mostly expressed in the cytoplasm and excluded from the nucleus (Fig. 7e), demonstrating that the role of ZFR on Stau2 nuclear export is isoform specific. As further control, shRNA III and Stau2-HA-co-transfected neurons were further labelled with anti-Stau2 antibodies to demonstrate that, even in cells where Stau259-HA was restricted to the cytoplasm, some endogenous Stau2 could be detected in the nucleus following RNAi-induced ZFR down-regulation (data not shown), confirming that Stau259 does not contribute to the signal observed in the nucleus.

image

Figure 7. Specific nuclear accumulation of Stau262-HA following down-regulation of ZFR by RNAi. Embryonic rat hippocampal neurons were transfected at 5 DIV and fixed 48 h later. (a–c) Co-transfection of eYFP cDNA as a marker for transfection and of cDNA coding for Stau1-HA (a), Stau259-HA (b) or Stau262-HA (c). Somatodendritic distribution of the proteins was observed. (d–f) Co-transfection of shRNA III and either Stau1-HA (d), Stau259-HA (e) or Stau262-HA (f). In contrast to Stau1-HA and Stau259-HA, a fraction of Stau262-HA was observed in the nucleus. Left panels, expression of YFP; middle panels, Staufen-HA detected with anti-HA antibodies; right panels, nuclear staining with DAPI. Scale bar, 10 μm.

Download figure to PowerPoint

Table 2.  Nuclear localization of HA-tagged Staufen proteins following down-regulation of ZFR by RNAi a
 CTLsiRNAesiRNAshRNA
  • a

    Nuclear localization is expressed as the percentage of neurons showing Staufen nuclear staining relative to the total number of transfected cells. A nuclear distribution pattern was scored positive when at least a fraction of the expressed protein was found in the nucleus/nucleolus in addition to its primarily cytoplasmic distribution.

  • b

    Each value is derived from n = 3 independent transfections.

  • c

    c Total number of transfected cells is shown in parentheses.

  • *

    p < 0,05 (unpaired t-test of RNAi result vs. CTL).

  • NS

    NS Statistically non-significant.

Stau1-HA44 ± 5b (115)c42 ± 4NS (52)38 ± 6NS (33)38 ± 4NS (31)
Stau262-HA26 ± 8 (64)64 ± 3* (35)70 ± 7* (38)77 ± 3* (74)
Stau259-HA18 ± 2 (48)11 ± 7NS (24)31 ± 6NS (29)30 ± 3NS (31)

Finally, using the same approach, we tested whether knockdown expression of ZFR modifies the subcellular distribution of Stau1. As ZFR had not been detected in Stau1-containing mRNP (Figs 4a and c), we expected that depletion of ZFR should not affect the subcellular localization of Stau1-HA. Transfected Stau1-HA showed mostly somato-dendritic distribution, with occasional nucleolar-like signal. As expected, co-transfection of DNA coding for Stau1-HA and shRNA against ZFR did not change the distribution of Stau1-HA (Fig. 7d and Table 2). As control, further staining with anti-Stau2 antibody demonstrated that some endogenous Stau2 was partly delocalized to the nucleus (data not shown). These results further support a model in which ZFR is a bona fide protein partner of Stau2.

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

ZFR is associated with the Stau262 isoform

Understanding the mechanisms of mRNA transport, localization and local translation is impaired by our partial knowledge of the RNA and protein constituents of RNA granules/particles. Identification and characterization of protein partners are crucial to decipher each step of these processes. In this study, we report the identification and characterization of a Stau2-associated protein in neurons, the zinc finger protein ZFR. In hippocampal neurons, several Stau2 isoforms have been observed: the Stau262 isoform is associated with ribosome-free complexes, whereas Stau259 and Stau252 isoforms are found in ribosome-associated granules (Duchaine et al. 2002; Mallardo et al. 2003). Confocal microscopy, immunoprecipitation and cell fractionation are all consistent with an association between ZFR and the Stau262 isoform. This association is specific as ZFR is not found with the paralogue protein Stau1. This observation is consistent with previous reports that demonstrated that, in hippocampal neurons and in oligodendrocytes, Stau1 and Stau2 are components of different granule-like structures (Duchaine et al. 2002; Mallardo et al. 2003; Thomas et al. 2005). Therefore, ZFR is likely to play a specific role with Stau262-containing complexes and is not a universal partner for all types of RNA granules.

Stau259 does not co-immunoprecipitate ZFR from neuron extracts and is not re-localized within the nucleus following down-regulation of ZFR by RNAi, suggesting that the two proteins do not interact. This was originally unexpected as both ZFR and Stau259 co-fractionate with cytoplasmic ribosomes in a sucrose gradient (Fig. 3) and both are co-immunoprecipitated by anti-P0 antibodies (Fig. 4) (Duchaine et al. 2002). Technical reasons, such as a putative unavailability of the Stau259-HA epitopes within the Stau2/ZFR complexes or a low efficacy of immunoprecipitation of ribosomes using anti-Stau2 antibodies, might explain the absence of ZFR in the immunoprecipitate. Alternatively, ZFR might not interact with Stau259 but rather with Stau252, another isoform that co-fractionates with these two proteins. It is also possible that a subpopulation of ZFR associates with ribosomes independently of Stau2 and forms other cytoplasmic RNP complexes of yet unknown function. Confocal microscopic analyses are consistent with this hypothesis of a heterogeneous population of granules containing either ZFR, ZFR/Stau2 or Stau2.

ZFR is an RNA/DNA-binding protein

Analyses of ZFR knockout mice indicate that ZFR is essential for at least some developmental pathways as embryonic death occurs by 8–9 days of gestation (Meagher and Braun 2001). In homozygotes, genetic ablation of ZFR causes increased embryonic cell death and/or decreased cell proliferation rates (Meagher and Braun 2001). In adults, ZFR is highly expressed in brain, ovary and testis, but weaker expression is detected in other tissues as well (Meagher et al. 1999; Kleines et al. 2001). ZFR contains three widely spaced zinc finger domains. Zinc finger proteins with a similar pattern of zinc finger motifs are known to bind RNA, DNA and DNA/RNA hybrids (Finerty and Bass 1997). Although ZFR endogenous substrates in the cells are still unknown, in vitro binding assays demonstrate that it has the capacity to bind both RNA and DNA (Meagher et al. 1999). In the cells, ZFR is found in the nucleus in association with chromosomes (Meagher et al. 1999). It was thus suggested that ZFR could be involved in DNA repair and chromosome organization and/or alternatively, that it could interact with nascent RNA at the site of transcription and be part of a complex regulating RNA processing (Meagher et al. 1999). Its association with Stau2 in the nucleus is consistent with the latter hypothesis.

ZFR is essential for the nucleo-cytoplasmic shuttling of Stau262

In neurons, ZFR is found, not only in the nucleus, but also in the cytoplasm, where it forms granule-like structures. The presence of ZFR in both the nuclear and cytoplasmic compartments suggests that the protein may shuttle between these two compartments. Interestingly, ZFR contains a putative nuclear export signal (Shim et al. 2002) and two putative nuclear localization signals (NLS) (Meagher et al. 1999), which are found inside and flanking the DZF domain, respectively. These signals could be responsible for the observed nucleo-cytoplasmic distribution of ZFR protein in hippocampal neurons. Similarly, Stau2 contains a NLS and, although mainly cytoplasmic, it may transit through the nucleus (this paper and Macchi et al. 2004 and Miki and Yoneda 2004). Interestingly, down-regulation of ZFR favours a nuclear redistribution of the Stau262 isoform. This nuclear retention is isoform specific as it does not affect the shorter Stau259 isoform, and gene specific as it does not affect Stau1. This Stau262 phenotype following ZFR down-regulation was obtained with three different RNAi approaches (shRNA, esiRNA and siRNA) that targeted three non-overlapping sites of the ZFR mRNA. As such, we diminished the probability of an off-target gene silencing effect. This result suggests that ZFR/Stau262 association may first occur in the nucleus and that ZFR is required for the nuclear exit of the complex. This would be consistent with recent data with other RNA-binding proteins that indicate that initial steps of RNA recognition and/or packaging of mRNP complexes to be transported in the cytoplasm occur in the nucleus (Hoek et al. 1998; Long et al. 2001; Larocque et al. 2002; Oleynikov and Singer 2003).

We hypothesize that ZFR first interacts with Stau262 in the nucleus. The affinity of ZFR for DNA, RNA and RNA/DNA hybrids may play a significant role in the selection of specific nascent mRNAs at the site of transcription in the nucleus and in their presentation to Stau262. The ZFR/Stau262/mRNA complexes then exit the nucleus, likely via the exportin-5 pathway (Macchi et al. 2004). Therefore, down-regulation of ZFR should affect RNA transport of specific Stau2-bound RNAs. Identification of a specific endogenous Stau262 RNA ligand will be necessary to test this hypothesis. It will be interesting to determine whether this reduced distribution of Stau262 and Stau262-bound RNAs in the cytoplasm contributes to the lethal phenotype of ZFR knockout mice.

Molecular characterization of Staufen in different species has demonstrated its capacity to establish protein interactions during localization. Oskar, Miranda and Inscuteable have been shown to interact with Staufen in Drosophila (Li et al. 1997; Schuldt et al. 1998; Micklem et al. 2000), and protein phosphatase-I and Barentz with Stau1 in mammals (Monshausen et al. 2002; Macchi et al. 2003). We now show that ZFR interacts with Stau2. The presence of Zfr and of Staufen genes in humans, nematode and flies suggests that the ZFR–Staufen association might be evolutionarily conserved. Whether ZFR associates with Staufen in other species is still unknown. As the null mutation of Zfr in mouse is lethal (Meagher and Braun 2001), it is surprising that mutation of the ZFR homologs in Drosophila or C. elegans have not been reported in the studies of development. It is likely that the different Staufen isoforms in mammals have acquired specialized functions during evolution, which are not necessary for fly development. Analysis of Zfr or Zfr/Staufen mutations in the invertebrate model organisms may contribute to our understanding of this ZFR/Stau2 association.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We thank Thomas Duchaîne for his help with the gradients, Louise Cournoyer for technical assistance with cell cultures, Francis Goyette and Sophie Brousseau for the construction of viral vectors and Dominique Nouel for the acquisition of confocal images. We also thank Dr M. Reichlin (Oklahoma Medical Research Foundation, USA) and Dr R. E. Braun (Washington, USA) for the human anti-ribosomal P antibodies and the rabbit polyclonal anti-ZFR antibodies, respectively, Dr G. Ferbeyre for the RNAi PCR SHAG strategy and Dr M. A. Kiebler for critical reading of the manuscript. This work was supported by a Canadian Institutes for Health Research (CIHR) grant to LDG.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • Banker G. and Goslin K. (1988) Developments in neuronal cell culture. Nature 336, 185186.
  • Bannai H., Fukatsu K., Mizutani A., Natsume T., Iemura S. I., Ikegami T., Inoue T. and Mikoshiba K. (2004) An RNA-interacting protein, SYNCRIP (heterogeneous nuclear ribonuclear protein Q1/NSAP1) is a component of mRNA granule transported with inositol 1,4,5-trisphosphate receptor type 1 mRN7A in neuronal dendrites. J. Biol. Chem. 279, 5342753434.
  • Bar-Peled M. and Raikhel N. V. (1996) A method for isolation and purification of specific antibodies to a protein fused to the GST. Anal. Biochem. 241, 140142.
  • Bassell G. J. and Kelic S. (2004) Binding proteins for mRNA localization and local translation, and their dysfunction in genetic neurological disease. Curr. Opin. Neurobiol. 14, 574581.
  • Brendel C., Rehbein M., Kreienkamp H. J., Buck F., Richter D. and Kindler S. (2004) Characterization of Staufen 1 ribonucleoprotein complexes. Biochem. J. 384, 239246.
  • Duchaine T. F., Hemraj I., Furic L., Deitinghoff A., Kiebler M. A. and DesGroseillers L. (2002) Staufen2 isoforms localize to the somatodendritic domain of neurons and interact with different organelles. J. Cell Sci. 115, 32853295.
  • Dugre-Brisson S., Elvira G., Boulay K., Chatel-Chaix L., Mouland A. J. and DesGroseillers L. (2005) Interaction of Staufen1 with the 5′ end of mRNA facilitates translation of these RNAs. Nucleic Acids Res. 33, 47974812.
  • Finerty P. J. Jr and Bass B. L. (1997) A Xenopus zinc finger protein that specifically binds dsRNA and RNA–DNA hybrids. J. Mol. Biol. 271, 195208.
  • Follenzi A. and Naldini L. (2002) HIV-based vectors. Preparation and use. Meth. Mol. Med. 69, 259274.
  • Hoek K. S., Kidd G. J., Carson J. H. and Smith R. (1998) hnRNP A2 selectively binds the cytoplasmic transport sequence of myelin basic protein mRNA. Biochemistry 37, 70217029.
  • Huang Y. S. and Richter J. D. (2004) Regulation of local mRNA translation. Curr. Opin. Cell Biol. 16, 308313.
  • Kanai Y., Dohmae N. and Hirokawa N. (2004) Kinesin transports RNA: isolation and characterization of an RNA-transporting granule. Neuron 43, 513525.
  • Kiebler M. A. and DesGroseillers L. (2000) Molecular insights into mRNA transport and local translation in the mammalian nervous system. Neuron 25, 1928.
  • Kiebler M. A., Hemraj I., Verkade P., Kohrmann M., Fortes P., Marion R. M., Ortin J. and Dotti C. G. (1999) The mammalian staufen protein localizes to the somatodendritic domain of cultured hippocampal neurons: implications for its involvement in mRNA transport. J. Neurosci. 19, 288297.
  • Kim-Ha J., Kerr K. and Macdonald P. M. (1995) Translational regulation of oskar mRNA by bruno, an ovarian RNA-binding protein, is essential. Cell 81, 403412.
  • Kleines M., Gartner A., Ritter K. and Schaade L. (2001) Cloning and expression of the human single copy homologue of the mouse zinc finger protein zfr. Gene 275, 157162.
  • Knowles R. B., Sabry J. H., Martone M. E., Deerinck T. J., Ellisman M. H., Bassell G. J. and Kosik K. S. (1996) Translocation of RNA granules in living neurons. J. Neurosci. 16, 78127820.
  • Kohrmann M., Luo M., Kaether C., DesGroseillers L., Dotti C. G. and Kiebler M. A. (1999) Microtubule-dependent recruitment of Staufen-green fluorescent protein into large RNA-containing granules and subsequent dendritic transport in living hippocampal neurons. Mol. Biol. Cell 10, 29452953.
  • Krichevsky A. M. and Kosik K. S. (2001) Neuronal RNA granules: a link between RNA localization and stimulation-dependent translation. Neuron 32, 683696.
  • Krichevsky A. M. and Kosik K. S. (2002) RNAi functions in cultured mammalian neurons. Proc. Natl Acad. Sci. USA 99, 1192611929.
  • Larocque D., Pilotte J., Chen T., Cloutier F., Massie B., Pedraza L., Couture R., Lasko P., Almazan G. and Richard S. (2002) Nuclear retention of MBP mRNAs in the quaking viable mice. Neuron 36, 815829.
  • Li P., Yang X., Wasser M., Cai Y. and Chia W. (1997) Inscuteable and Staufen mediate asymmetric localization and segregation of prospero RNA during Drosophila neuroblast cell divisions. Cell 90, 437447.
  • Long R. M., Gu W., Meng X., Gonsalvez G., Singer R. H. and Chartrand P. (2001) An exclusively nuclear RNA-binding protein affects asymmetric localization of ASH1 mRNA and Ash1p in yeast. J. Cell Biol. 153, 307318.
  • Macchi P., Kroening S., Palacios I. M., Baldassa S., Grunewald B., Ambrosino C., Goetze B., Lupas A., St Johnston D. and Kiebler M. (2003) Barentsz, a new component of the Staufen-containing ribonucleoprotein particles in mammalian cells, interacts with Staufen in an RNA-dependent manner. J. Neurosci. 23, 57785788.
  • Macchi P., Brownawell A. M., Grunewald B., DesGroseillers L., Macara I. G. and Kiebler M. A. (2004) The brain-specific double-stranded RNA-binding protein Staufen2: nucleolar accumulation and isoform-specific exportin-5-dependent export. J. Biol. Chem. 279, 3144031444.
  • Mallardo M., Deitinghoff A., Muller J., Goetze B., Macchi P., Peters C. and Kiebler M. A. (2003) Isolation and characterization of Staufen-containing ribonucleoprotein particles from rat brain. Proc. Natl Acad. Sci. USA 100, 21002105.
  • Marion R. M., Fortes P., Beloso A., Dotti C. and Ortin J. (1999) A human sequence homologue of Staufen is an RNA-binding protein that is associated with polysomes and localizes to the rough endoplasmic reticulum. Mol. Cell Biol. 19, 22122219.
  • Meagher M. J. and Braun R. E. (2001) Requirement for the murine zinc finger protein ZFR in perigastrulation growth and survival. Mol. Cell Biol. 21, 28802890.
  • Meagher M. J., Schumacher J. M., Lee K., Holdcraft R. W., Edelhoff S., Disteche C. and Braun R. E. (1999) Identification of ZFR, an ancient and highly conserved murine chromosome-associated zinc finger protein. Gene 228, 197211.
  • Micklem D. R., Adams J., Grunert S. and St Johnston D. (2000) Distinct roles of two conserved Staufen domains in oskar mRNA localization and translation. EMBO J. 19, 13661377.
  • Miki T. and Yoneda Y. (2004) Alternative splicing of Staufen2 creates the nuclear export signal for CRM1 (Exportin 1). J. Biol. Chem. 279, 4747347479.
  • Monshausen M., Rehbein M., Richter D. and Kindler S. (2002) The RNA-binding protein Staufen from rat brain interacts with protein phosphatase-1. J. Neurochem. 81, 557564.
  • Monshausen M., Gehring N. H. and Kosik K. S. (2004) The mammalian RNA-binding protein Staufen2 links nuclear and cytoplasmic RNA processing pathways in neurons. Neuromol. Med. 6, 127144.
  • Ohashi S., Koike K., Omori A., Ichinose S., Ohara S., Kobayashi S., Sato T. A. and Anzai K. (2002) Identification of mRNA/protein (mRNP) complexes containing Purα, mStaufen, fragile X protein, and myosin Va and their association with rough endoplasmic reticulum equipped with a kinesin motor. J. Biol. Chem. 277, 3780437810.
  • Oleynikov Y. and Singer R. H. (2003) Real-time visualization of ZBP1 association with β-actin mRNA during transcription and localization. Curr. Biol. 13, 199207.
  • Paddison P. J., Caudy A. A., Sachidanandam R. and Hannon G. J. (2004) Short hairpin activated gene silencing in mammalian cells. Meth. Mol. Biol. 265, 85100.
  • Robb G. B., Brown K. M., Khurana J. and Rana T. M. (2005) Specific and potent RNAi in the nucleus of human cells. Nat. Struct. Mol. Biol. 12, 133137.
  • Schuldt A. J., Adams J. H., Davidson C. M., Micklem D. R., Haseloff J., St Johnston D. and Brand A. H. (1998) Miranda mediates asymmetric protein and RNA localization in the developing nervous system. Genes Dev. 12, 18471857.
  • Shim J., Lim H., Yates R. and Karin M. (2002) Nuclear export of NF90 is required for interleukin-2 mRNA stabilization. Mol. Cell 10, 13311344.
  • St Johnston D., Beuchle D. and Nusslein-Volhard C. (1991) Staufen, a gene required to localize maternal RNAs in the Drosophila egg. Cell 66, 5163.
  • Steward O. and Worley P. F. (2001) A cellular mechanism for targeting newly synthesized mRNAs to synaptic sites on dendrites. Proc. Natl Acad. Sci. USA 98, 70627068.
  • Sutton M. A. and Schuman E. M. (2005) Local translational control in dendrites and its role in long-term synaptic plasticity. J. Neurobiol. 64, 116131.
  • Tang S. J., Meulemans D., Vazquez L., Colaco N. and Schuman E. (2001) A role for a rat homolog of staufen in the transport of RNA to neuronal dendrites. Neuron 32, 463475.
  • Thomas M. G., Tosar L. J., Loschi M., Pasquini J. M., Correale J., Kindler S. and Boccaccio G. L. (2005) Staufen recruitment into stress granules does not affect early mRNA transport in oligodendrocytes. Mol. Biol. Cell 16, 405420.
  • Villace P., Marion R. M. and Ortin J. (2004) The composition of Staufen-containing RNA granules from human cells indicates their role in the regulated transport and translation of messenger RNAs. Nucleic Acids Res. 32, 24112420.
  • Wickham L., Duchaine T., Luo M., Nabi I. R. and DesGroseillers L. (1999) Mammalian staufen is a double-stranded-RNA- and tubulin-binding protein which localizes to the rough endoplasmic reticulum. Mol. Cell Biol. 19, 22202230.