Barentsz, a new component of the Staufen-containing ribonucleoprotein particles in mammalian cells, interacts with Staufen in an RNA-dependent manner

Abstract

Staufen1, the mammalian homolog of Drosophila Staufen, assembles into ribonucleoprotein particles (RNPs), which are thought to transport and localize RNA into dendrites of mature hippocampal neurons. We therefore investigated whether additional components of the RNA localization complex besides Staufen are conserved. One candidate is the mammalian homolog of Drosophila Barentsz (Btz), which is essential for the localization of oskar mRNA to the posterior pole of the Drosophila oocyte and is a component of the oskar RNA localization complex along with Staufen. In this study, we report the characterization of mammalian Btz, which behaves like a nucleocytoplasmic shuttling protein. When expressed in the Drosophila egg chamber, mammalian Btz is still able to interact with Drosophila Staufen and reach the posterior pole in the wild-type oocyte, but does not rescue the btz mutant phenotype. Most interestingly, we show by immunoprecipitation assays that Btz interacts with mammalian Staufen in an RNA-dependent manner through a conserved domain, which encompasses the region of homology to the Drosophila Btz protein and contains a novel conserved motif. One candidate for an RNA that mediates this interaction is the dendritically localized brain cytoplasmic 1 transcript. In addition, Btz and Staufen1 colocalize within particles in the cell body and, to a more variable extent, in dendrites of mature hippocampal neurons. Together, our data suggest that the mRNA transport machinery is conserved during evolution, and that mammalian Btz is an additional component of the dendritic RNPs in hippocampal neurons.

Figure 1.

Domain structure of mammalian Barentsz. Schematic representation of rat Btz protein. The different domains are indicated. Two putative NLSs and an NES have been predicted. The black bar indicates the region of homology to Drosophila Btz. A comparison of the NES sequences of mBtz, drosulfakinin (Dsk1), Abelson murine leukemia viral oncogene (c-ABL), and protein kinase inhibitor (PKI) is shown on the top of the figure. Hydrophobic residues important for NES activity are indicated by open boxes.

Figure 2.

Expression of Btz mRNA in different tissues. A, Northern blot analysis of polyA mRNA from the following different mouse tissues: heart (He), brain (Br), spleen (Sp), lung (Lu), liver (Li), skeletal muscle (Mu), kidney (Ki), and testis (Te). Bottom panel,β-Actin control. B, Btz mRNA is expressed in the hippocampus. In situ hybridization on P14 mouse brain sections showing the expression of Btz mRNA. Coronal section of the hippocampus (1), including a high magnification insert (2). 3, In situ hybridization performed with the sense probe in the hippocampus.

Figure 3.

Leptomycin B-sensitive export of mBtz protein. A, LMB treatment of BHK cells expressing mBtz–GFP fusion protein (mBtz–GFP). The CRM1-dependent axin1 served as control (F. Fagotto, unpublished observations). Three hour treatment with LMB causes the retention of Btz protein within the nucleus. Arrows indicate the perinuclear and cytoplasmic pattern of Btz-GFP. The experiments were performed either in the absence (-LMB) or in the presence (+LMB) of LMB. Scale bar, 10 μm. B, Mapping of the nuclear localization and nuclear export signal of mammalian Btz. BHK cells transfected with Btz–GFP show cytoplasmic localization of the fusion protein (i). In the subsequent panels, cells were transfected with a vector expressing the N-terminal half of the protein fused with GFP, Btz_1–359–GFP (ii), Btz_1–158–GFP (iii), or Btz_152–359–GFP (iv). The bottom panel (v) shows the nuclear accumulation of Btz_L465A–GFP, in which leucine 465 of the putative NES has been mutated to alanine. Scale bar, 10 μm (for all images in B). C, Expression of Btz–GFP in transiently transfected hippocampal neurons confirms the phenotype observed in transfected BHK cells. GFP fluorescence is shown on the left, whereas both the respective 4′,6′-diamidino-2-phenylindole (DAPI) staining and phase contrast (PC) are shown on the right. Transfected glia are shown in the small panels. Arrowheads indicate the particulate Btz–GFP structures observed in dendrites. The inset represents a high magnification with increased exposure time of the area indicated by the dotted square. Scale bar, 10 μm (for all images in C).

Figure 4.

Mammalian Btz colocalizes with Staufen and oskar mRNA in the Drosophila oocyte. A, Localization of mBtz–GFP in the wild-type (WT) Drosophila germ line. Similar to Drosophila Btz, mBtz accumulates in the stage 5 oocyte (left panel), in a rim around the nuclei of the nurse cells (middle panel), and at the posterior pole of the stage 9 oocyte (right panel, arrow), where both Staufen and oskar mRNA localize. B, Localization of Drosophila Staufen (dStau) (red) and mBtz–GFP (green) in btz1/TM6B and btz1 mutant egg chambers expressing mBtz–GFP (top and middle panels, respectively). The arrowheads indicate the colocalization of Drosophila Staufen and mBtz–GFP proteins to the posterior pole. The bottom panel shows Staufen localization in btz1 mutant egg chambers that do not express mBtz–GFP. In the btz1 mutant oocyte, both Staufen and mBtz–GFP do not accumulate at the posterior pole but instead localize at the anterior pole (middle panel, arrows). Thus, mBtz–GFP is unable to rescue the oskar mRNA localization phenotype in barentsz mutant egg chambers.

Figure 5.

Mammalian Btz is a 115 kDa protein. A, Western blot of extracts from hippocampal neurons probed with either preimmune (PI) (lane 1) or the corresponding immune antisera. Lane 2, A monospecific mouse polyclonal antisera (I); lane 3, affinity-purified (Pur.), rabbit polyclonal anti-Btz. Our different monospecific polyclonal antisera recognize a 115 kDa protein; this is in good accordance with the observed molecular mass of Drosophila Btz, which runs as a 125 kDa protein. Right panel, In vitro translation experiment showing that mBtz protein migrates as a 115 kDa protein. A negative control (-) was performed without DNA. The arrowhead identifies the phosphorylated form of mBtz. B, Western blot using extracts of different tissues. The antibodies recognize the protein overexpressed in BHK (right panel) but also the endogenous protein from BHK or from different neuronal tissues (left panel). αBtz, Anti-Btz antibodies.

Figure 6.

Intracellular localization of mBtz protein in hippocampal neurons. A, Endogenous mBtz and mBtz–GFP show a remarkably similar localization in the cell body of primary hippocampal neurons (10–12 DIV). Neurons were immunostained for mBtz protein (top panel) or transiently transfected with a construct encoding mBtz–GFP, and fluorescent images were taken from a representative cell (bottom panel). Both proteins yielded a typical punctate, perinuclear localization (see also inset). B, Confocal images from fixed mature hippocampal neurons in culture. Like Staufen1, mBtz forms particles that localize to the dendritic compartment. The immunostaining was performed with affinity-purified rabbit polyclonal anti-mBtz antibodies. C, Higher magnifications of confocal microscopical images. Two different regions of dendrites (proximal–top to distal–bottom) are shown. Hippocampal neurons were labeled with immunopurified rabbit polyclonal anti-mBtz (green) and mouse anti-Stau1 (red). D, Quantitative analysis to evaluate the colocalization between mBtz and Stau1. Three different regions (shown in the schematic drawing) were chosen. The three-column set represents quantifications from three distinct neurons. The number on the top of each column indicates the number of areas analyzed per region. E, In vivo colocalization of Staufen1 and mBtz in transfected hippocampal neurons. Mature neurons were transiently transfected with constructs encoding mBtz and Staufen1–HA, and then immunostaining was performed. Fluorescent images were taken from three different cells. Panels marked Merge in B and E are an overlay of the two images to the left. αBtz, Anti-Btz antibodies; αStau1, anti-Staufen1 antibodies.

Figure 7.

mBtz coimmunoprecipitates with Staufen1. A, Endogenous Staufen1 immunoprecipitates Btz. IPs were performed with two different antisera (S1, S2), and mBtz was detected with a specific antibody. As negative control, the IP was performed with the preimmune serum (PI). B, C, In vivo Staufen1–mBtz interaction. BHK cells were cotransfected with different constructs and lysed, and immunoprecipitations were performed as indicated. B, Btz–GFP and Btz_1–359–GFP coimmunoprecipitate with Stau1–HA. The left panel includes the indicated cell extracts (CE) showing the respective molecular masses of mBtz–GFP and Staufen1–HA. IPs (right panel) were performed with anti-GFP antibodies (αGFP), and proteins were visualized by Western blotting (underlined tags indicate the various antibodies). C, The interaction of mBtz is specific for Staufen1. Cotransfection of BHK cells with constructs expressing both mBtz–GFP and axin1–myc did not result in a coimmunoprecipitation of the two recombinant proteins. αStau1, anti-Staufen1 antibodies.

Figure 8.

The interaction of Btz and Staufen1 is RNA dependent. A, RNA-dependent interaction of mBtz with organelles. RNase treatment of S14 supernatants from 3-week-old rat forebrain extracts releases mBtz from the P100 (P) into the soluble fraction (S). DNase treatment, in contrast, does not change the distribution of mBtz protein. The same membrane was redecorated with monoclonal anti-calnexin antibodies [kind gift of Ari Helenius (Eidgenössische Technische Hochschule, Zürich, Switzerland)] to show that the ER marker calnexin is not removed by RNase treatment. B, mBtz interacts with Staufen1 in an RNA-dependent manner. BHK cells were cotransfected and processed as indicated in Figure 7. Extracts were pretreated with RNase where indicated, and IPs were performed with anti-GFP antibodies. In this experiment, Btz–GFP and Stau1–HA only coimmunoprecipitate in the presence of RNA. Treatment with RNase A before the IP abolishes this interaction. As a control, GFP alone does not interact with Staufen1. The underlining indicates the portion of the antigen recognized by the antibodies. C, The conserved central domain of mammalian Btz is responsible for the RNA-dependent interaction with Staufen1. Different constructs coding for various parts of Btz were expressed together with Stau1–HA in BHK cells, and IPs were performed using anti-GFP antibodies in the presence or absence of RNase. The region from amino acid 152–359 in mouse Btz protein mediates the interaction with Stau1–HA. This domain contains the two identified motifs (1 and 2) (Fig. 1) that are conserved between Drosophila and mBtz proteins. CE, Cell extract; αGFP, Anti-GFP antibodies.

Figure 9.

Characterization of RNA in mBtz immunoprecipitates. A, The dendritically localized BC1 transcript is present in the mBtz IPs. RNA isolated from the mBtz IPs was retrotranscribed, and the resulting cDNA was amplified using primers specific for either dendritically localized (BC1, MAP2) or nonlocalized (β-actin, Histone H1) transcripts. RNA was collected from IPs performed either in absence of antibodies (lane 1) or with immunopurified anti-mBtz polyclonal antibodies (lane 2) and then retrotranscribed, and the indicated cDNAs were detected by PCR. As positive controls, PCRs were also performed on cDNA from S100 (lane 3). B, Radioactive labeling of mRNAs presentin the mBtz IPs. Radioactive RT-PCR was performed on RNA extracted from either mBtz immunoprecipitates (lane 4) or the corresponding supernatant (lane 2). As negative control, the anti-mBtz antibodies were omitted for the IPs (lane3) to detect for mRNAs unspecifically bound to protein A-Sepharose beads. Lane 1 represents the corresponding supernatant. Arrowheads label a distinct set of bands that were specifically enriched in the mBtz IPs (lane 4).