Title: Dendritic mRNA Targeting of Jacob and N-Methyl-d-aspartate-induced Nuclear Translocation after Calpain-mediated Proteolysis
Abstract: Jacob is a recently identified plasticity-related protein that couples N-methyl-d-aspartate receptor activity to nuclear gene expression. An expression analysis by Northern blot and in situ hybridization shows that Jacob is almost exclusively present in brain, in particular in the cortex and the limbic system. Alternative splicing gives rise to multiple mRNA variants, all of which exhibit a prominent dendritic localization in the hippocampus. Functional analysis in primary hippocampal neurons revealed that a predominant cis-acting dendritic targeting element in the 3′-untranslated region of Jacob mRNAs is responsible for dendritic mRNA localization. In the mouse brain, Jacob transcripts are associated with both the fragile X mental retardation protein, a well described trans-acting factor regulating dendritic mRNA targeting and translation, and the kinesin family member 5C motor complex, which is known to mediate dendritic mRNA transport. Jacob is susceptible to rapid protein degradation in a Ca2+- and Calpain-dependent manner, and Calpain-mediated clipping of the myristoylated N terminus of Jacob is required for its nuclear translocation after N-methyl-d-aspartate receptor activation. Our data suggest that local synthesis in dendrites may be necessary to replenish dendritic Jacob pools after truncation of the N-terminal membrane anchor and concomitant translocation of Jacob to the nucleus. Jacob is a recently identified plasticity-related protein that couples N-methyl-d-aspartate receptor activity to nuclear gene expression. An expression analysis by Northern blot and in situ hybridization shows that Jacob is almost exclusively present in brain, in particular in the cortex and the limbic system. Alternative splicing gives rise to multiple mRNA variants, all of which exhibit a prominent dendritic localization in the hippocampus. Functional analysis in primary hippocampal neurons revealed that a predominant cis-acting dendritic targeting element in the 3′-untranslated region of Jacob mRNAs is responsible for dendritic mRNA localization. In the mouse brain, Jacob transcripts are associated with both the fragile X mental retardation protein, a well described trans-acting factor regulating dendritic mRNA targeting and translation, and the kinesin family member 5C motor complex, which is known to mediate dendritic mRNA transport. Jacob is susceptible to rapid protein degradation in a Ca2+- and Calpain-dependent manner, and Calpain-mediated clipping of the myristoylated N terminus of Jacob is required for its nuclear translocation after N-methyl-d-aspartate receptor activation. Our data suggest that local synthesis in dendrites may be necessary to replenish dendritic Jacob pools after truncation of the N-terminal membrane anchor and concomitant translocation of Jacob to the nucleus. The link between excitatory neurotransmission and transcriptional and translational regulation has attracted much interest for many years because multiple processes ranging from metabolic homeostasis to learning and memory require activity-driven gene expression in neurons (1Deisseroth K. Mermelstein P.G. Xia H. Tsien R.W. Curr. Opin. Neurobiol. 2003; 13: 354-365Crossref PubMed Scopus (294) Google Scholar, 2Flavell S.W. Greenberg M.E. Annu. Rev. Neurosci. 2008; 31: 563-590Crossref PubMed Scopus (613) Google Scholar). Particularly signaling from N-methyl-d-aspartate (NMDA) 3The abbreviations used are:NMDAN-methyl-d-aspartateDIVday(s) in vitroDTEdendritic targeting elementFMRPfragile X mental retardation proteinGAPDHglyceraldehyde-3-phosphate dehydrogenaseEGFPenhanced green fluorescent proteinGSTglutathione S-transferaseKIF5ckinesin family member 5CMAP2microtubuli-associated protein 2MBPmaltose-binding proteinPABPpoly(A)-binding proteinUTRuntranslated regionRTreverse transcriptaseWTwild typeKOknock-outF-IPimmunoprecipitation with FMRP-specific antibodyP-IPimmunoprecipitation with PABP-specific antibodyIgG-IPimmunoprecipitation with unrelated rabbit IgGsRNPribonucleoproteinαCaMKIICa2+/calmodulin-dependent protein kinase II. type glutamate receptors to the nucleus has been implicated in synaptic plasticity, and some molecules have been identified that can translocate from synaptic and extrasynaptic sites to neuronal nuclei after NMDA receptor activation (3Lai K.O. Zhao Y. Ch'ng T.H. Martin K.C. Proc. Natl. Acad. Sci. U.S.A. 2008; 105: 17175-17180Crossref PubMed Scopus (88) Google Scholar, 4Proepper C. Johannsen S. Liebau S. Dahl J. Vaida B. Bockmann J. Kreutz M.R. Gundelfinger E.D. Boeckers T.M. EMBO J. 2007; 26: 1397-1409Crossref PubMed Scopus (96) Google Scholar, 5Jordan B.A. Fernholz B.D. Khatri L. Ziff E.B. Nat. Neurosci. 2007; 10: 427-435Crossref PubMed Scopus (96) Google Scholar, 6Dieterich D.C. Karpova A. Mikhaylova M. Zdobnova I. König I. Landwehr M. Kreutz M. Smalla K.H. Richter K. Landgraf P. Reissner C. Boeckers T.M. Zuschratter W. Spilker C. Seidenbecher C.I. Garner C.C. Gundelfinger E.D. Kreutz M.R. PLoS Biol. 2008; 6: e34Crossref PubMed Scopus (137) Google Scholar, 7Jordan B.A. Kreutz M.R. Trends in Neurosci. 2009; 32: 392-401Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar). In a recent study, we have identified Jacob, a protein that triggers long lasting changes in the cytoarchitecture of dendrites and the number of spine synapses in pyramidal neurons via coupling of NMDA receptor signaling to nuclear gene expression (6Dieterich D.C. Karpova A. Mikhaylova M. Zdobnova I. König I. Landwehr M. Kreutz M. Smalla K.H. Richter K. Landgraf P. Reissner C. Boeckers T.M. Zuschratter W. Spilker C. Seidenbecher C.I. Garner C.C. Gundelfinger E.D. Kreutz M.R. PLoS Biol. 2008; 6: e34Crossref PubMed Scopus (137) Google Scholar). Following activation of synaptic and extrasynaptic NMDA receptors, Jacob is recruited to neuronal nuclei, and this in turn results in a rapid stripping of synaptic contacts and in a drastically altered morphology of the dendritic tree (6Dieterich D.C. Karpova A. Mikhaylova M. Zdobnova I. König I. Landwehr M. Kreutz M. Smalla K.H. Richter K. Landgraf P. Reissner C. Boeckers T.M. Zuschratter W. Spilker C. Seidenbecher C.I. Garner C.C. Gundelfinger E.D. Kreutz M.R. PLoS Biol. 2008; 6: e34Crossref PubMed Scopus (137) Google Scholar). Nuclear import of Jacob utilizes the classical importin pathway (6Dieterich D.C. Karpova A. Mikhaylova M. Zdobnova I. König I. Landwehr M. Kreutz M. Smalla K.H. Richter K. Landgraf P. Reissner C. Boeckers T.M. Zuschratter W. Spilker C. Seidenbecher C.I. Garner C.C. Gundelfinger E.D. Kreutz M.R. PLoS Biol. 2008; 6: e34Crossref PubMed Scopus (137) Google Scholar, 7Jordan B.A. Kreutz M.R. Trends in Neurosci. 2009; 32: 392-401Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar), and the synaptic Ca2+-binding protein Caldendrin regulates the extra-nuclear localization of Jacob by competing with the binding of importin-α to a bipartite nuclear localization signal in Jacob (6Dieterich D.C. Karpova A. Mikhaylova M. Zdobnova I. König I. Landwehr M. Kreutz M. Smalla K.H. Richter K. Landgraf P. Reissner C. Boeckers T.M. Zuschratter W. Spilker C. Seidenbecher C.I. Garner C.C. Gundelfinger E.D. Kreutz M.R. PLoS Biol. 2008; 6: e34Crossref PubMed Scopus (137) Google Scholar). In the nucleus, Jacob can trigger dephosphorylation of the transcription factor CREB and gene expression that destabilizes synaptic contacts. These events may be part of homoeostatic plasticity, i.e. the constant optimization of synaptic input of a given neuron (6Dieterich D.C. Karpova A. Mikhaylova M. Zdobnova I. König I. Landwehr M. Kreutz M. Smalla K.H. Richter K. Landgraf P. Reissner C. Boeckers T.M. Zuschratter W. Spilker C. Seidenbecher C.I. Garner C.C. Gundelfinger E.D. Kreutz M.R. PLoS Biol. 2008; 6: e34Crossref PubMed Scopus (137) Google Scholar, 8Turrigiano G.G. Cell. 2008; 135: 422-435Abstract Full Text Full Text PDF PubMed Scopus (1020) Google Scholar). Thus, it seems likely that the balance between the nuclear and extra-nuclear pool of Jacob is highly dynamic and tightly controlled. N-methyl-d-aspartate day(s) in vitro dendritic targeting element fragile X mental retardation protein glyceraldehyde-3-phosphate dehydrogenase enhanced green fluorescent protein glutathione S-transferase kinesin family member 5C microtubuli-associated protein 2 maltose-binding protein poly(A)-binding protein untranslated region reverse transcriptase wild type knock-out immunoprecipitation with FMRP-specific antibody immunoprecipitation with PABP-specific antibody immunoprecipitation with unrelated rabbit IgGs ribonucleoprotein Ca2+/calmodulin-dependent protein kinase II. In the present study, we have investigated the local expression and turnover of Jacob in more detail. We identified a number of alternative splice products, all of which are largely restricted to the limbic system and cortex. In the hippocampus, all Jacob mRNAs exhibit a prominent dendritic localization, which is mediated by a dendritic targeting element (DTE) residing in the 3′-untranslated region (3′-UTR). Jacob transcripts are associated both with KIF5c, a motor implicated in cytoplasmic transport of several dendritic mRNAs, and the fragile X mental retardation protein (FMRP), which is thought to play a central role in both dendritic trafficking (9Kanai Y. Dohmae N. Hirokawa N. Neuron. 2004; 43: 513-525Abstract Full Text Full Text PDF PubMed Scopus (853) Google Scholar, 10Dictenberg J.B. Swanger S.A. Antar L.N. Singer R.H. Bassell G.J. Dev. Cell. 2008; 14: 926-939Abstract Full Text Full Text PDF PubMed Scopus (375) Google Scholar) and local translation of mRNAs at postsynaptic sites (11Grossman A.W. Aldridge G.M. Weiler I.J. Greenough W.T. J Neurosci. 2006; 26: 7151-7155Crossref PubMed Scopus (140) Google Scholar, 12Garber K.B. Visootsak J. Warren S.T. Eur. J. Hum. Genet. 2008; 16: 666-672Crossref PubMed Scopus (293) Google Scholar, 13Bardoni B. Davidovic L. Bensaid M. Khandjian E.W. Expert Rev. Mol. Med. 2006; 8: 1-16Crossref PubMed Scopus (77) Google Scholar, 14Ronesi J.A. Huber K.M. Sci. Signal. 2008; 1: pe6Crossref PubMed Scopus (87) Google Scholar). Interestingly, Jacob is a very unstable protein that is rapidly degraded in a Ca2+- and Calpain-sensitive manner. In conjunction with its dynamic translocation from dendrites to the nucleus, this might impose the necessity of replenishing the cytoplasmic protein pool by the local translation of dendritic mRNAs. Northern blot analysis was performed with a multiple tissue Northern blot (Clontech) according to published procedures (15Seidenbecher C.I. Langnaese K. Sanmartí-Vila L. Boeckers T.M. Smalla K.H. Sabel B.A. Garner C.C. Gundelfinger E.D. Kreutz M.R. J. Biol. Chem. 1998; 273: 21324-21331Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar) using a 32P-labeled cDNA sample encompassing bp 104–2827 of the Jacob cDNA. In situ hybridization of rat brain cryostat sections was performed as described (16Laube G. Seidenbecher C.I. Richter K. Dieterich D.C. Hoffmann B. Landwehr M. Smalla K.H. Winter C. Böckers T.M. Wolf G. Gundelfinger E.D. Kreutz M.R. Mol. Cell Neurosci. 2002; 19: 459-475Crossref PubMed Scopus (62) Google Scholar). Control experiments included hybridization with a sense probe and prior RNase treatment. Antisense oligonucleotides were deduced from the Jacob cDNA (accession number AJ293697): 5′-CAG GGC TGG CTC TCT AGA GAT GGT GTA CAC ACG GGG CTG G-3′ (Jacob-pan); 5′-TCT CCC GTT TCC GAC GCT TCC TCT CCG CGT AGC C-3′ (Jacob+exon6); 5′-GGA GTC GTG GGA GGT GTC GGC TTT CAT AGG GGT G-3′ (Jacob+exon8); 5′-AGG TGT TTG CGG AAG TTC GAT ATG GCT TGC ATA G-3′ (Jacob-exon6); 5′-GCT CTG TAG GTC ACT GCT CTG GGC CTT CAC CCG C-3′ (Jacob-exon8); and 5′-CCC ACG ACT CCC GAG ACA CTA ATC TCC TCC AAG GTG-3′ (Jacob-exon 9). The oligonucleotide sequence for tubulin-α probe was 5′-GGA CCA GAA TAA ACA TCC CTG TGA AAG CAG CAC CTT GTG AC-3′. Corresponding sense oligonucleotides were used as controls. Plasmids pNEu2432–3071 and pNEtub have been previously described (11Grossman A.W. Aldridge G.M. Weiler I.J. Greenough W.T. J Neurosci. 2006; 26: 7151-7155Crossref PubMed Scopus (140) Google Scholar). Vectors pNEL134–1603 and pNEL1625–2827 are derivatives of the plasmid pNE (11Grossman A.W. Aldridge G.M. Weiler I.J. Greenough W.T. J Neurosci. 2006; 26: 7151-7155Crossref PubMed Scopus (140) Google Scholar) and contain nucleotides 134–1603 and 1625–2827 of the Jacob cDNA (accession number AJ293697) downstream of the EGFP coding region. In vivo time lapse imaging was done with a plasmid where EGFP is downstream of the coding region of Jacob (6Dieterich D.C. Karpova A. Mikhaylova M. Zdobnova I. König I. Landwehr M. Kreutz M. Smalla K.H. Richter K. Landgraf P. Reissner C. Boeckers T.M. Zuschratter W. Spilker C. Seidenbecher C.I. Garner C.C. Gundelfinger E.D. Kreutz M.R. PLoS Biol. 2008; 6: e34Crossref PubMed Scopus (137) Google Scholar). For bacterial protein production the first 230 amino acids (nucleotides 134–824) of Jacob were cloned into pMALC2X. Wistar rats, Fmr1−/− mice (knock-out; B6.129P2-Fmr1tm1Cgr strain; Jackson Laboratory) and congenic C57BL/6J wild type mice were raised in the animal facility of the University Medical Center Hamburg-Eppendorf or the Leibniz Institute for Neurobiology. Primary neurons were essentially prepared and transfected as described (17Blichenberg A. Schwanke B. Rehbein M. Garner C.C. Richter D. Kindler S. J. Neurosci. 1999; 19: 8818-8829Crossref PubMed Google Scholar). However, the neurons were grown in neurobasal medium (Invitrogen) without glial feeder layer. Rat hippocampal neurons were transfected 8 days after plating (days in vitro (DIV) and fixed 6 days after transfection. Immunocytochemical analysis, synthesis of digoxigenin-labeled RNA probes, in situ hybridization analysis, and scoring of recombinant dendritic transcripts was performed as described (17Blichenberg A. Schwanke B. Rehbein M. Garner C.C. Richter D. Kindler S. J. Neurosci. 1999; 19: 8818-8829Crossref PubMed Google Scholar). Combined fluorescent in situ hybridization/immunocytochemical approaches were performed with Cy3-coupled mouse monoclonal anti-digoxigenin (Roche Applied Science) and rabbit polyclonal anti-MAP2 antibodies (generous gift from Craig C. Garner, Stanford University) and rabbit polyclonal Jacob antibodies (6Dieterich D.C. Karpova A. Mikhaylova M. Zdobnova I. König I. Landwehr M. Kreutz M. Smalla K.H. Richter K. Landgraf P. Reissner C. Boeckers T.M. Zuschratter W. Spilker C. Seidenbecher C.I. Garner C.C. Gundelfinger E.D. Kreutz M.R. PLoS Biol. 2008; 6: e34Crossref PubMed Scopus (137) Google Scholar) followed by Cy3-coupled sheep anti-mouse (Sigma) and AlexaFluor488-coupled goat anti-rabbit antibodies (Molecular Probes-Invitrogen). Jacob RNA probes comprise full-length coding and 3′-untranslated regions. Images captured with a Zeiss Axiovert 135 microscope (Carl Zeiss MicroImaging) and Openlab software (Improvision) or a laser-scanning microscope (TCS-SP2 Leica) were mounted using Adobe Photoshop CS (Adobe Systems) and Freehand (Macromedia) software. For stimulation, primary hippocampal neurons (DIV16 for calpeptin and DIV21 for E64d) were incubated with 60 μm calpeptin and 50 μm E64d for 30 min at 37 °C. NMDA (100 μm) was applied for 5 min in stimulation buffer (Neurobasal medium with 7.5 μm anisomycin) either in the absence or presence of inhibitors. After stimulation, the cells were washed twice with stimulation buffer and then incubated for 30 min at 37 °C. The cells were fixed with 4% paraformaldehyde in 1× phosphate-buffered saline (150 mm NaCl, 20 mm sodium phosphate, pH 7.4) for 10 min at 37 °C and then incubated in permeabilization solution (0.25% Triton X-100 in 1× phosphate-buffered saline) for 10 min at room temperature. Subsequently, they were incubated in blocking solution (2% bovine serum albumin, 2% glycine, 0.2% gelatin, 50 mm NH4Cl) for 1.5 h at room temperature. The primary antibody (Jacob JB150) (6Dieterich D.C. Karpova A. Mikhaylova M. Zdobnova I. König I. Landwehr M. Kreutz M. Smalla K.H. Richter K. Landgraf P. Reissner C. Boeckers T.M. Zuschratter W. Spilker C. Seidenbecher C.I. Garner C.C. Gundelfinger E.D. Kreutz M.R. PLoS Biol. 2008; 6: e34Crossref PubMed Scopus (137) Google Scholar) was diluted 1:100 in blocking solution and incubated overnight at 4 °C. After washing (3 × 10 min in phosphate-buffered saline with 0.3% Triton X-100) the cells were incubated with AlexaFluor-coupled secondary goat anti-rabbit IgG antibody (1:1000 in blocking solution; Molecular Probes-Invitrogen) for 2 h at room temperature, washed as described above, and embedded in Mowiol (Calbiochem). Utilizing 4′,6-diamidino-2-phenylindole staining, the nuclei were identified, and nuclear Jacob levels were determined by calculating the mean pixel intensity from two or three nuclear planes. The differences between groups are described as relative deviations from the control. The nuclear membrane was excluded from the analysis. Images were taken using a Leica DMRXE microscope equipped with a Krypton-Argon-Ion laser (488/568/647 nm) and an acousto-optic-tunable filter for selection and intensity adaptation of laser lines. The images were analyzed with ImageJ software. In vivo time lapse imaging and transfection of hippocampal primary neurons with Jacob-EGFP was performed as described (6Dieterich D.C. Karpova A. Mikhaylova M. Zdobnova I. König I. Landwehr M. Kreutz M. Smalla K.H. Richter K. Landgraf P. Reissner C. Boeckers T.M. Zuschratter W. Spilker C. Seidenbecher C.I. Garner C.C. Gundelfinger E.D. Kreutz M.R. PLoS Biol. 2008; 6: e34Crossref PubMed Scopus (137) Google Scholar). Stimulation was done with 20 μm NMDA either in the presence or absence of 60 μm calpeptin. Western blotting was done as described previously (15Seidenbecher C.I. Langnaese K. Sanmartí-Vila L. Boeckers T.M. Smalla K.H. Sabel B.A. Garner C.C. Gundelfinger E.D. Kreutz M.R. J. Biol. Chem. 1998; 273: 21324-21331Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar). GST and the fusion protein GST-KIF5c containing the cargo binding domain of mouse KIF5c (amino acid residues 826–920, accession number NP_032475) were expressed in Escherichia coli and purified using GSH-Sepharose (GE Biotech). GST-KIF5C-coated Sepharose beads were used to affinity purify KIF5C-associated cargos from mouse brain homogenates as described (10Dictenberg J.B. Swanger S.A. Antar L.N. Singer R.H. Bassell G.J. Dev. Cell. 2008; 14: 926-939Abstract Full Text Full Text PDF PubMed Scopus (375) Google Scholar). Immunoprecipitation experiments were performed with antibodies directed against fragile X mental retardation protein (rabbit polyclonal antibody H-120; Santa Cruz Biotechnology, Heidelberg, Germany) and poly(A)-binding protein (PABP) (18Brendel C. Rehbein M. Kreienkamp H.J. Buck F. Richter D. Kindler S. Biochem. J. 2004; 384: 239-246Crossref PubMed Scopus (62) Google Scholar) as well as irrelevant rabbit IgGs and brain extracts derived from adult male Fmr1−/− mice and congenic C57BL/6J wild type (WT) mice as described (19Iacoangeli A. Rozhdestvensky T.S. Dolzhanskaya N. Tournier B. Schütt J. Brosius J. Denman R.B. Khandjian E.W. Kindler S. Tiedge H. Proc. Natl. Acad. Sci. U.S.A. 2008; 105: 734-739Crossref PubMed Scopus (70) Google Scholar). From brain homogenates and immuno- and GST-precipitates, RNA was extracted using the RNeasy mini kit (Qiagen) and analyzed by RT-PCR and quantitative real time RT-PCR with a Rotor-Gene 3000 (Corbett, Wasserburg, Germany) using the QuantiTect SYBR Green RT-PCR kit (Qiagen) as described (19Iacoangeli A. Rozhdestvensky T.S. Dolzhanskaya N. Tournier B. Schütt J. Brosius J. Denman R.B. Khandjian E.W. Kindler S. Tiedge H. Proc. Natl. Acad. Sci. U.S.A. 2008; 105: 734-739Crossref PubMed Scopus (70) Google Scholar). The PCR conditions were according to the standard protocol of the manufacturer with an annealing temperature of 58 °C (40 cycles) and the following gene-specific primers: BC1, BC1-Fw (5′-GTT GGG GAT TTA GCT CAG TGG-3′) and BC1-Rev (5′-AGG TTG TGT GTG CCA GTT ACC-3′); glyceraldehyde-3-phosphate dehydrogenase (GAPDH), GAPDH-Fw (5′-TGG CAA AGT GGA GAT TGT TGC C-3′) and GAPDH-Rev (5′-AAG ATG GTG ATG GGC TTC CCG-3′); and SAPAP3, SAPAP3-Fw (5′-ACT ATT TGC AGG TGC CGC AAG-3′) and SAPAP3-Rev (5′-GGG CTA CCA TCT GAG TCT CC-3′). For Jacob and Arc/Arg3.1 mRNAs the appropriate QuantiTect Primer Assay was obtained from Qiagen (catalog numbers QT01077832 and QT00250684, respectively). RT-PCR products were analyzed on 2% agarose gels. MBP fusion protein production was performed as described previously (20Landgraf P. Wahle P. Pape H.C. Gundelfinger E.D. Kreutz M.R. J. Biol. Chem. 2008; 283: 25036-25045Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar). Samples containing 5 nm MBP-Jacob or 5 nm MBP alone, 2 mm Ca+2 or 2 mm EGTA, and μ-Calpain (specific activity, 0.195 enzyme units; Calbiochem) were incubated in Calpain reaction buffer (1× Tris-buffered saline, pH 7.6). The reactions were brought to 20 μl and incubated at 30 °C for 10 s, 30 s, and 1 min. The reactions were terminated by the addition of 4× SDS sample buffer. MBP-Jacob protein degradation was visualized on immunoblots. On a rat multiple tissue Northern blot two transcripts, a major one at 2.9 kb and a less intense one at 3.6 kb, can be clearly detected in the brain when hybridized with a Jacob cDNA probe. Very faint signals were detectable in the testis, and even fainter or no signals were identified in other investigated tissues (Fig. 1A). The two bands may reflect alternative splicing of the Jacob pre-mRNA. Subsequent in situ hybridization experiments revealed a strikingly restricted localization of Jacob transcripts to the limbic system and cortical areas of the rat brain. The highest levels of Jacob mRNAs are present in the cerebral cortex, hippocampus, olfactory bulb, thalamus, and amygdala, whereas the transcript concentration in the striatum and cerebellum is very low (Fig. 2A). Interestingly, gene expression appears to be developmentally regulated with highest mRNA levels between the second and third postnatal week, the postnatal period during which synapto-dendritic wiring is most dynamic. The sections treated with RNase H (Fig. 2B) or hybridized with sense controls (data not shown) showed no signal above background. A closer inspection of the hybridization pattern revealed that Jacob mRNAs are abundantly present in the molecular layers of the hippocampus (Fig. 2, A and C), indicating dendritic localization of the transcripts. In contrast, α-tubulin transcripts are restricted to regions of neuronal somata (Fig. 2C).FIGURE 2.Jacob is prominently expressed in cortex and the limbic brain and exhibits a dendritic mRNA in the hippocampus. A, in situ hybridization of horizontal brain sections from postnatal (P) 0, P4, P11, P14, P21, and adult rats. X-ray film images of slices labeled with probe Jacob-pan (see "Experimental Procedures") are depicted. Magnifications of the hippocampal formation are shown in the lower panels. Bo, olfactory bulb; sCtx, somatosensory cortex; Th, thalamus; St, striatum; eCtx, entorhinal cortex; Cb, cerebellum; Hi, hippocampus; CA1, Ammon's horn subfield 1; CA2, Ammon's horn subfield 2; CA3, Ammon's horn subfield 3; DG, dentate gyrus. B, hybridization on sections pretreated with RNase H resulted in no signal above background. C, comparison of Jacob and tubulin-α hybridization signals in the hippocampus.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Alternative splicing of exons 5, 6, 8, and 9 generates multiple Jacob mRNAs (Fig. 3, A and B). Thus, we sought to determine whether distinct transcripts differ in their brain distribution. Surprisingly, in the adult rat brain all tested splice variants exhibit a similar distribution pattern with slightly different levels of abundance (Fig. 3C). Importantly, all investigated alternatively spliced transcripts are present in the molecular layers of the hippocampus (Fig. 3C) and thus should be transported to dendrites. Furthermore these data suggest that the dendritic mRNA localization is due to a sequence element common to all Jacob mRNA variants. Therefore we decided to analyze the dendritic targeting of Jacob mRNAs in more detail. Extrasomatic trafficking of mRNAs in mammalian neurons involves cis-acting DTEs within the localized transcripts (20Landgraf P. Wahle P. Pape H.C. Gundelfinger E.D. Kreutz M.R. J. Biol. Chem. 2008; 283: 25036-25045Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar, 21Bramham C.R. Wells D.G. Nat. Rev. Neurosci. 2007; 8: 776-789Crossref PubMed Scopus (468) Google Scholar, 22Kindler S. Wang H. Richter D. Tiedge H. Annu. Rev. Cell Dev. Biol. 2005; 21: 223-245Crossref PubMed Scopus (168) Google Scholar). To determine the presence and position of DTEs within Jacob transcripts, we transfected DIV8 primary rat hippocampal neurons with eukaryotic expression vectors encoding chimeric reporter mRNAs, all of which contain the EGFP coding region as a common part, a sequence that is incapable to mediate dendritic mRNA targeting (17Blichenberg A. Schwanke B. Rehbein M. Garner C.C. Richter D. Kindler S. J. Neurosci. 1999; 19: 8818-8829Crossref PubMed Google Scholar). Six days after transfection, the subcellular localization of recombinant transcripts was determined by nonradioactive in situ hybridization with an EGFP antisense probe (17Blichenberg A. Schwanke B. Rehbein M. Garner C.C. Richter D. Kindler S. J. Neurosci. 1999; 19: 8818-8829Crossref PubMed Google Scholar). Although reporter transcripts containing the DTE of MAP2 mRNAs displayed dendritic localization in almost 90% of the transfected cells (n = 81; Fig. 4, A and E), chimeric EGFP-α-tubulin transcripts were mainly restricted to somata (9,7% dendritic targeting, n = 154; Fig. 4, B and E). In comparison, reporter transcripts containing either the coding region (pNELjaccr) or the complete 3′-UTR of Jacob mRNAs (pNELjacu) localized to dendrites in about 40% (n = 154) and slightly more than 60% (n = 250) of the neurons, respectively (Fig. 4, C–E). Thus, whereas the coding region of Jacob transcripts appears to contain sequences promoting dendritic mRNA targeting, the major DTE mediating efficient extrasomatic trafficking resides in the 3′-UTR, which is common to all splice variants, in agreement with our previous observations that all Jacob transcripts are present in dendritic fields in the hippocampus. In dendrites, pNELjacu derived reporter mRNAs assembled into distinct granules, which appear to resemble individual RNA transport packages (Fig. 4D). FMRP is an RNA-binding protein that has been implicated in both dendritic mRNA targeting and the local control of translation at postsynaptic sites (10Dictenberg J.B. Swanger S.A. Antar L.N. Singer R.H. Bassell G.J. Dev. Cell. 2008; 14: 926-939Abstract Full Text Full Text PDF PubMed Scopus (375) Google Scholar, 13Bardoni B. Davidovic L. Bensaid M. Khandjian E.W. Expert Rev. Mol. Med. 2006; 8: 1-16Crossref PubMed Scopus (77) Google Scholar, 23Darnell J.C. Mostovetsky O. Darnell R.B. Genes Brain Behav. 2005; 4: 341-349Crossref PubMed Scopus (94) Google Scholar). To determine whether FMRP may be involved in cytoplasmic processing of Jacob transcripts, we probed for an in vivo association of the two components. Utilizing brain homogenates derived from adult male Fmr1−/− knock-out mice (KO) and congenic C57BL/6J WT mice as input material, we performed immunoprecipitations with FMRP-specific (F-IP) and PABP-specific (P-IP) antibodies as well as with unrelated rabbit IgGs (IgG-IP) as a control. P-IP was chosen as a positive control because PABP generally interacts with polyadenylated mRNAs. KO-F-IP and WT-IgG-IP were carried out to check for the possibility of unspecific RNA precipitation. RNA extracted from immunoprecipitates was probed for Jacob, SAPAP3, and BC1 RNAs by RT-PCR and real time RT-PCR. SAPAP3- and BC1-specific amplifications served as positive and negative controls for FMRP-associated RNAs, respectively (19Iacoangeli A. Rozhdestvensky T.S. Dolzhanskaya N. Tournier B. Schütt J. Brosius J. Denman R.B. Khandjian E.W. Kindler S. Tiedge H. Proc. Natl. Acad. Sci. U.S.A. 2008; 105: 734-739Crossref PubMed Scopus (70) Google Scholar, 24Narayanan U. Nalavadi V. Nakamoto M. Thomas G. Ceman S. Bassell G.J. Warren S.T. J. Biol. Chem. 2008; 283: 18478-18482Abstract Full Text Full Text PDF PubMed Scopus (172) Google Scholar, 25Brown V. Jin P. Ceman S. Darnell J.C. O'Donnell W.T. Tenenbaum S.A. Jin X. Feng Y. Wilkinson K.D. Keene J.D. Darnell R.B. Warren S.T. Cell. 2001; 107: 477-487Abstract Full Text Full Text PDF PubMed Scopus (912) Google Scholar). RT-PCR data showed that Jacob and SAPAP3 mRNAs were present in both the WT-F-IP and WT-P-IP but were basically absent or only weakly detected in the KO-F-IP and WT-IgG-IP (Fig. 5A). Moreover, as previously shown (19Iacoangeli A. Rozhdestvensky T.S. Dolzhanskaya N. Tournier B. Schütt J. Brosius J. Denman R.B. Khandjian E.W. Kindler S. Tiedge H. Proc. Natl. Acad. Sci. U.S.A. 2008; 105: 734-739Crossref PubMed Scopus (70) Google Scholar) BC1 RNA was only found in the WT-P-IP, but not in the WT-F-IP, KO-F-IP, and WT-IgG-IP (Fig. 5A), thus underscoring the specificity of the assay. These findings were confirmed by real time RT-PCR analysis (Fig. 5B). Similar to SAPAP3 mRNAs, known FMRP targets (24Narayanan U. Nalavadi V. Nakamoto M. Thomas G. Ceman S. Bassell G.J. Warren S.T. J. Biol. Chem. 2008; 283: 18478-18482Abstract Full Text Full Text PDF PubMed Scopus (172) Google Scholar, 25Brown V. Jin P. Ceman S. Darnell J.C. O'Donnell W.T. Tenenbaum S.A. Jin X. Feng Y. Wilkinson K.D. Keene J.D. Darnell R.B. Warren S.T. Cell. 2001; 107: 477-487Abstract Full Text Full Text PDF PubMed Scopus (912) Google Scholar), Jacob transcripts w