SEARCH

SEARCH BY CITATION

Keywords:

  • cell polarity;
  • extrasomatic translation;
  • ribonucleoprotein particle;
  • RNA-binding protein;
  • RNA localization

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Subcellular protein fractions, enrichment and affinity purification of MARTA1
  5. In vitro RNA synthesis and UV cross-linking assays
  6. Peptide sequence analysis
  7. PCR and cDNA sequence analysis
  8. Cell culture
  9. Antibodies, immunoprecipitation, immunoblotting and immunohistochemistry
  10. Results
  11. Purification of rat MARTA1
  12. Determination of rat MARTA1 peptide sequences
  13. Cloning of rat MARTA1 cDNAs
  14. Sequence specific RNA-binding
  15. Rat MARTA1 is present in nuclei and the somatodendritic cytoplasm of neurones
  16. Discussion
  17. Acknowledgements
  18. References

In neurones, the somatodendritic microtubule-associated protein 2 regulates the stability of the dendritic cytoskeleton. Its extrasomatic localization appears to be a multicausal mechanism that involves dendritic mRNA trafficking, a process that depends on a dendritic targeting element in the 3′ untranslated region. Two rat MAP2-RNA trans-acting proteins, MARTA1 and MARTA2, exhibit specific high-affinity binding to the dendritic targeting element. We have now affinity-purified MARTA1 from rat brain. Analysis of proteolytic peptides revealed that rat MARTA1 is the orthologue of the human RNA-binding protein KSRP. Rat MARTA1 is a 74-kDa protein that contains four putative RNA-binding domains and is 98% identical to human KSRP. Both purified rat MARTA1 and human KSRP preferentially bind to the dendritic targeting element, but do not strongly interact with other investigated regions of mRNAs encoding microtubule-associated protein 2 and α-tubulin. In rat brain neurones and cultured neurones derived from superior cervical ganglia, MARTA1 is primarily intranuclear, but is also present in the somatodendritic cytoplasm. Thus, MARTA1 may play a role in nucleocytoplasmic mRNA targeting.

Abbreviations used
DTE

dendritic targeting element

DTT

dithiothreitol

FBP

far upstream element binding protein

HEK

human embryonic kidney

hnRNP

heterogeneous nuclear ribonucleoprotein

KH domain

K homology domain

KSRP

K homology-type splicing regulatory protein

MAP

microtubule-associated protein

MARTA

MAP2-RNA trans-acting protein

PCR

polymerase chain reaction

RSW

ribosomal salt wash

UV

ultraviolet

UTR

untranslated region.

A key feature of eukaryotic cells is the functional specialization of individual subareas based upon locally restricted protein patterns. Asymmetric protein distribution is established by at least two basic mechanisms: selective protein trafficking and mRNA sorting allowing a spatially restricted, regulated translation of individual transcripts in subcellular regions. An ordered interaction between cis-acting RNA elements and trans-acting proteins appears to regulate cytoplasmic mRNA targeting and site-specific translation (Jansen 2001).

In mammalian neurones, dendritic RNA sorting is an energy-dependent mechanism that requires cytoskeletal filaments (Kiebler and DesGroseillers 2000; Wells et al. 2000). Identification of polysomes in dendritic shafts led to the hypothesis that some mRNAs are extrasomatically translated in dendrites (Steward and Schuman 2001). This idea was supported by the detection of protein synthesis in isolated dendrites (Job and Eberwine 2001). In addition, dendritic translation seems to influence synaptic plasticity (Steward and Schuman 2001).

In transcripts encoding rat isoforms of the microtubule-associated protein 2 (MAP2), a 640-nucleotide cis-acting dendritic targeting element (DTE) in the 3′ untranslated region (3′ UTR) is required and sufficient for effective dendritic sorting of chimeric reporter transcripts (Blichenberg et al. 1999). Two major trans-acting factors, MARTA1 and MARTA2, specifically bind to the MAP2-DTE in vitro with nanomolar affinity (Rehbein et al. 2000). In this study, we affinity-purified MARTA1. Sequence analysis of proteolytic peptides revealed that rat MARTA1 is the orthologue of the human K homology-type splicing regulatory protein (KSRP; Min et al. 1997). The 74-kDa rat MARTA1 contains four KH domains and is about 98% identical to human KSRP. Both purified rat MARTA1 and human KSRP specifically bind to the MAP2-DTE. In neurones, MARTA1 is present in nuclei and the somatodendritic cytoplasm. Thus, it may regulate dendritic targeting of MAP2 mRNAs via a specific interaction with the DTE.

Subcellular protein fractions, enrichment and affinity purification of MARTA1

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Subcellular protein fractions, enrichment and affinity purification of MARTA1
  5. In vitro RNA synthesis and UV cross-linking assays
  6. Peptide sequence analysis
  7. PCR and cDNA sequence analysis
  8. Cell culture
  9. Antibodies, immunoprecipitation, immunoblotting and immunohistochemistry
  10. Results
  11. Purification of rat MARTA1
  12. Determination of rat MARTA1 peptide sequences
  13. Cloning of rat MARTA1 cDNAs
  14. Sequence specific RNA-binding
  15. Rat MARTA1 is present in nuclei and the somatodendritic cytoplasm of neurones
  16. Discussion
  17. Acknowledgements
  18. References

Crude lysates, S90 cytosolic and RSW fractions were prepared as described (Rehbein et al. 2000). For purification, MARTA1 was precipitated with 30% (w/v) ammonium sulphate from an S90 cytosolic fraction from 22 g of rat brain. Precipitated proteins were solubilized and desalted in loading buffer [10 mm Tris–HCl pH 7.6, 5% (v/v) glycerol, 0.1 mm ethylenediaminetetraacetic acid, 1 mm dithiothreitol (DTT), 0.1 mm phenylmethylsulfonyl fluoride (PMSF)] via a HiTrap G25 column (Amersham Pharmacia Biotech, Uppsala, Sweden), and 30–40 mg were applied to a 5-mL HiTrap heparin column (Amersham Pharmacia Biotech). Bound proteins were eluted with a step gradient of NaCl (0.1, 0.2, 0.3, 0.4 and 0.5 m) in loading buffer. MARTA1 containing fractions (0.1 and 0.2m NaCl) were equilibrated with 1 mm Tris–HCl pH 7.5, 0.1 mm DTT, and concentrated by lyophilization. Lyophilized proteins (4.4 mg) were incubated with DTE-coated magnetic streptavidin beads (6 mg; Promega, Madison, WI, USA) in 10 mL optimized binding buffer (10 mm HEPES pH 7.6; 0.7% glycerol, 1 mm DTT, 1 mg/mL heparin) containing 150 units of human placental ribonuclease inhibitor (Amersham Pharmacia Biotech) and 0.1% (v/v of stock solution) proteinase inhibitor cocktail Complete (Roche Diagnostics, Heidelberg, Germany) for 40 min at room temperature. Unbound proteins were removed and beads were washed three times for 30 s each in 5 mL optimized binding buffer, and MARTA1 was eluted in 800, 400 and 200 µL 10 mm HEPES pH 7.6, 0.7% (v/v) glycerol, 1 mm DTT, 0.25 m NaCl, 3 mm MgCl2, 0.1% proteinase inhibitor cocktail for 3 min each, respectively.

In vitro RNA synthesis and UV cross-linking assays

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Subcellular protein fractions, enrichment and affinity purification of MARTA1
  5. In vitro RNA synthesis and UV cross-linking assays
  6. Peptide sequence analysis
  7. PCR and cDNA sequence analysis
  8. Cell culture
  9. Antibodies, immunoprecipitation, immunoblotting and immunohistochemistry
  10. Results
  11. Purification of rat MARTA1
  12. Determination of rat MARTA1 peptide sequences
  13. Cloning of rat MARTA1 cDNAs
  14. Sequence specific RNA-binding
  15. Rat MARTA1 is present in nuclei and the somatodendritic cytoplasm of neurones
  16. Discussion
  17. Acknowledgements
  18. References

In vitro synthesis of RNA probes was performed as described (Rehbein et al. 2000). Biotinylated RNA was generated with the RiboMax System (Promega) in the presence of Biotin-16-UTP (Roche Diagnostics) and unmodified UTP at a 1 : 50 molar ratio. UV cross-linking assays were performed in standard buffer containing 10 mm HEPES, pH 7.6, 3 mm MgCl2, 40 mm KCl, 5% (v/v) glycerol, 1 mm DTT, 0.1 mm EDTA, and 10 µg heparin/µL as described (Rehbein et al. 2000).

Peptide sequence analysis

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Subcellular protein fractions, enrichment and affinity purification of MARTA1
  5. In vitro RNA synthesis and UV cross-linking assays
  6. Peptide sequence analysis
  7. PCR and cDNA sequence analysis
  8. Cell culture
  9. Antibodies, immunoprecipitation, immunoblotting and immunohistochemistry
  10. Results
  11. Purification of rat MARTA1
  12. Determination of rat MARTA1 peptide sequences
  13. Cloning of rat MARTA1 cDNAs
  14. Sequence specific RNA-binding
  15. Rat MARTA1 is present in nuclei and the somatodendritic cytoplasm of neurones
  16. Discussion
  17. Acknowledgements
  18. References

Affinity-purified MARTA1 was excised from a Coomassie-stained sodium dodecyl sulphate (SDS)-gel and digested with endoprotease LysC. Proteolytic peptides were separated as described (Heidebrecht et al. 1997). Peptide sequences were determined by automated standard Edman degradation (473 A Protein Sequencer, Applied Biosystems, Foster City, CA, USA). Mass spectrometry was performed as described (Lellek et al. 2000).

PCR and cDNA sequence analysis

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Subcellular protein fractions, enrichment and affinity purification of MARTA1
  5. In vitro RNA synthesis and UV cross-linking assays
  6. Peptide sequence analysis
  7. PCR and cDNA sequence analysis
  8. Cell culture
  9. Antibodies, immunoprecipitation, immunoblotting and immunohistochemistry
  10. Results
  11. Purification of rat MARTA1
  12. Determination of rat MARTA1 peptide sequences
  13. Cloning of rat MARTA1 cDNAs
  14. Sequence specific RNA-binding
  15. Rat MARTA1 is present in nuclei and the somatodendritic cytoplasm of neurones
  16. Discussion
  17. Acknowledgements
  18. References

A rat EST sequence (GenBank accession number EST291720) and a genomic mouse sequence (AF094696) show high similarity to the human KSRP cDNA (U94832). Primers derived from these sequences (5′-CCATGTCGGACTACAGCACG-3′ and 5′-CCAATGATCAAGCCCACCATGC-3′, nucleotides 1–485 in the MARTA1cDNA; 5′-AAGGACGCTTTCGCCGACGCNGTN-3′ and 5′-GAACTTTGCAGCCTGAGTCCT-3′, nucleotides 214–538; 5′-GCATGGTGGGCTTGATCATTGG-3′ and 5′-GGGCAGGTGGTTGCCACTCG-3′, nucleotides 464–1678; 5′-CGAGTGGCAACCACCTGC-3′ and 5′-GGAAAAAATATTTTCACAGATGAAG-3′, nucleotides 1659–2179) were used for PCR with a pGAD10 adult rat brain cDNA library (Clontech Laboratories, Palo Alto, CA) as template. PCR products were subcloned into pGEM (Promega) and sequenced. 5′-Rapid amplification of cDNA ends was performed with the SMART Marathon kit (Clontech Laboratories, Palo Alto,CA, USA), total brain RNA from 13-day-old rats and twonestedprimers (5′-CTCTTCTGTCATTGAGGTCCTG-3′; 5′-GAAATCAGGAGTGTTGTTATTCAC-3′).

Antibodies, immunoprecipitation, immunoblotting and immunohistochemistry

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Subcellular protein fractions, enrichment and affinity purification of MARTA1
  5. In vitro RNA synthesis and UV cross-linking assays
  6. Peptide sequence analysis
  7. PCR and cDNA sequence analysis
  8. Cell culture
  9. Antibodies, immunoprecipitation, immunoblotting and immunohistochemistry
  10. Results
  11. Purification of rat MARTA1
  12. Determination of rat MARTA1 peptide sequences
  13. Cloning of rat MARTA1 cDNAs
  14. Sequence specific RNA-binding
  15. Rat MARTA1 is present in nuclei and the somatodendritic cytoplasm of neurones
  16. Discussion
  17. Acknowledgements
  18. References

A pET 15b vector (Novagen, Madison, WI, USA) for expression of His6-tagged full-length KSRP in Escherichia coli BL21-CodonPlus (DE3-RP; Stratagene GmbH, Heidelberg, Germany) was generously provided by Dr Douglas Black (Howard Hughes Medical Institute, University of California at Los Angeles). From bacterial lysates, the protein was purified with a NiNTA resin (Promega). Antibodies were generated in rabbits and affinity purified by BioGenes (Berlin, Germany). For immunoprecipitation, protein A-agarose beads (20 µL; Santa Cruz Biotechnology, Santa Cruz, CA, USA) were coated with 10 µL anti-KSRP serum, pre-immune serum, or used without coupled IgG. Fifty microliters of a RNase-digested standard UV cross-linking assay containing 20 µg rat brain S90 protein and 25 fmoles of the DTE probe were added to the protein A Sepharose beads. Total volume was adjusted to 250 µL with IP-buffer (10 mm Tris–HCl pH 8.0, 0.15 m NaCl, 0.1% NP-40) and the mixture was incubated at 4°C overnight on a rotator. After six washes with 100 µL IP-buffer each, beads were resuspended in 30 µL Laemmli buffer and incubated at 70°C for 10 min. Precipitated proteins (15 µL samples) and 16 µL from the supernatant fractions of the immunoprecipitation assays were separated on an SDS polyacrylamide gel. For western blotting, all primary antibodies were diluted 1 : 1000. Reactive bands were visualized by enhanced chemiluminescence (ECL-western blotting analysis system, Amersham Pharmacia Biotech). Immunohistochemistry on brain sections (1 : 1000 dilution of KSRP antiserum) and immunocytochemistry on primary neurones (1 : 500 dilution of affinity-purified KSRP antibodies) were performed as described (Blichenberg et al. 1999; Monshausen et al. 2001).

Purification of rat MARTA1

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Subcellular protein fractions, enrichment and affinity purification of MARTA1
  5. In vitro RNA synthesis and UV cross-linking assays
  6. Peptide sequence analysis
  7. PCR and cDNA sequence analysis
  8. Cell culture
  9. Antibodies, immunoprecipitation, immunoblotting and immunohistochemistry
  10. Results
  11. Purification of rat MARTA1
  12. Determination of rat MARTA1 peptide sequences
  13. Cloning of rat MARTA1 cDNAs
  14. Sequence specific RNA-binding
  15. Rat MARTA1 is present in nuclei and the somatodendritic cytoplasm of neurones
  16. Discussion
  17. Acknowledgements
  18. References

Two proteins, MARTA1 and MARTA2, specifically bind to the MAP2-DTE (Rehbein et al. 2000). We now purified MARTA1 from rat brain (Fig. 1a). From a S90 supernatant fraction from 22 g of brain MARTA1 was precipitated with 30% ammonium sulphate (Fig. 1b). The solubilized 30% ammonium sulphate precipitate (Fig. 1b, lane 5) was loaded on a heparin column and eluted with a step gradient of sodium chloride. MARTA1 was enriched in the 0.1–0.2 m salt eluate fraction (Fig. 1b, lane 6). Prior to affinity purification, the conditions for the DTE–MARTA1 interaction were optimized with this fraction in UV cross-linking assays. Depletion of either potassium chloride, magnesium chloride or both salts from the assay buffer significantly increased MARTA1 binding to the DTE (Fig. 1c, lanes 2–4, respectively). The RNA–protein interaction was further enhanced by depleting glycerol (Fig. 1c, lane 5) and reducing the heparin concentration to one tenth in a glycerol-free buffer (Fig. 1c, lane 6). For affinity purification, MARTA1 eluted from the heparin column was incubated with magnetic streptavidin beads coated with biotinylated DTE-RNA in a salt- and glycerol-free optimized binding buffer containing a low heparin concentration. After washing the beads, MARTA1 was eluted in a salt-containing buffer. MARTA1 binding activity was weak in the supernatant fraction of the batch purification and undetectable in wash fractions (Fig. 1d, lanes 2–5), but was strong in the eluates (Fig. 1d, lanes 6–8). In a silver-stained SDS–polyacrylamide gel an 85-kDa band was visible in eluate fractions (Fig. 1e, lanes 6–8), but not in wash fractions (Fig. 1e, lanes 3–5). The relative intensity of the 85-kDa band in different eluate fractions correlated well with the MARTA1 binding activity obtained in corresponding UV cross-linking assays, indicating that this band represents MARTA1.

image

Figure 1. Purification of MARTA1 from adult rat brain. (a) Flow diagram of experimental protocol. (b, c, d) MARTA1 enrichment during purification monitored in UV cross-linking assays with 5 fmoles of DTE probe. MARTA1 positions are marked (arrowheads). (b) Assays were performed in standard buffer with 5 µg protein each of S90 (lane 1), the 30% and 50% ammonium sulphate pellet (lanes 2, 3, respectively), the 50% ammonium sulphate supernatant (lane 4) fractions, and with 1 µg protein each of the 30% ammonium sulphate pellet (lane 5) and the pooled MARTA1 peak fractions of the heparin column (lane 6). (c) Different buffers were used in UV cross-linking assays with the pooled MARTA1 peak fractions of the heparin column; standard buffer (lane1), buffer depleted for either KCl (lane 2), MgCl2 (lane 3), KCl and MgCl2 (lane 4), or both salts and glycerol (lane 5). In lane 6, the heparin concentration of the salt- and glycerol-free binding buffer was reduced by 90% to a final concentration of 1 µg/mL (d) UV cross-linking assays were performed in optimized buffer with 1 µg protein eluted from the heparin column (lane 1) and 2 µL each of different affinity purification steps (lane 2, supernatant fraction of affinity-beads; lanes 3–5, wash fractions 1–3, respectively; lanes 6–8, eluate fractions 1–3; for details see Materials and methods). Rat MARTA1 is found in all eluates, but not in wash fractions. (e) Samples [1 µg protein in lanes 1, 2; 20 µL each of the other fractions used in (d), lanes 3–8] were analysed on a silver-stained SDS–polyacrylamide gel. An 85-kDa band in eluate samples correlates with MARTA1 binding activity shown in (d). For a description of lanes 1–8 see (d). Lane 9, molecular weight marker.

Download figure to PowerPoint

In UV cross-linking assays, the purified 85-kDa protein strongly associated with RNA probes containing the MAP2-DTE (Fig. 2a, lanes 1 and 2), but interacted only very weakly with fragment cE (Fig. 2b) from the MAP2 coding region (Fig. 2a, lane 3) or the complete α-tubulin transcript (Fig. 2a, lane 4). In competition assays, an 83-fold molar excess of the unlabeled DTE or a larger DTE containing fragment (uA; Fig. 2b) strongly interfered with MARTA1 binding to the labeled DTE probe (Fig. 2a; lanes 5 and 6, respectively). However, fragment cE from the MAP2 coding region or the complete α-tubulin mRNA did not compete with the MARTA1–DTE interaction (Fig. 2a; lanes 7 and 8, respectively). Thus, purified MARTA1 exhibits sequence-specific DTE-binding.

image

Figure 2. (a) Sequence-specific RNA-binding of purified rat MARTA1 in UV cross-linking assays. Affinity-purified MARTA1 (1 µL) was incubated with 3 fmoles each of different 32P-labeled transcripts alone (lanes 1–4), or together with 250 fmoles unlabeled competitor RNA (lanes 5–8). Probes: MAP2-DTE (lanes 1, 5–8), fragmentuA(MAP2–3′ UTR, lane 2), fragment cE (MAP2 coding region, lane 3), α-tubulin mRNA (lane 4). Competitor RNAs: MAP2-DTE (lane 5), uA (lane 6), cE (lane 7), α-tubulin mRNA (lane 8). Molecular weight markers are shown on the right. (b) Schematic representation of 9.6-kb MAP2 mRNA. Boxes and gray lines represent coding and non-coding regions, respectively. Black bars: regions used as probes or competitors in UV cross-linking assays. RNA fragments DTE and uA mediate dendritic mRNA trafficking in primary neurones.

Download figure to PowerPoint

Determination of rat MARTA1 peptide sequences

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Subcellular protein fractions, enrichment and affinity purification of MARTA1
  5. In vitro RNA synthesis and UV cross-linking assays
  6. Peptide sequence analysis
  7. PCR and cDNA sequence analysis
  8. Cell culture
  9. Antibodies, immunoprecipitation, immunoblotting and immunohistochemistry
  10. Results
  11. Purification of rat MARTA1
  12. Determination of rat MARTA1 peptide sequences
  13. Cloning of rat MARTA1 cDNAs
  14. Sequence specific RNA-binding
  15. Rat MARTA1 is present in nuclei and the somatodendritic cytoplasm of neurones
  16. Discussion
  17. Acknowledgements
  18. References

Affinity-purified rat MARTA1 was isolated from a Coomassie blue-stained SDS–gel and digested with endoprotease LysC. Peptides were separated by reverse phase high-pressure liquid chromatography and subjected to automated Edman sequencing. Sequences obtained from peptides 1, 2 and 3 comprise 14, 20 and 12 amino acid residues, respectively (Fig. 3a). A computer-based homology search (www.ncbi.nlm.nih.gov/BLAST) revealed that peptides 1, 2 and 3 are 100%, 95% and 92% identical tocorresponding sequences in human KSRP (GenBank accession number U94832; Min et al. 1997). KSRP is thought to be involved in neurone-specific splicing of the c-src premRNA mediated by an intronic splicing enhancer element. Peptides 1 and 3 also show 86% and 83% identity to sequences in the human far upstream element binding protein (FBP; accession number U05409; Duncan et al. 1994). However, peptide 2 is not found in FBP. FBP binds to a single-stranded sequence far upstream of the c-myc gene thus enhancing its transcription. Sequences of 13 additional peptides were identified by mass spectrometry (Table 1). Ten peptides (numbers 4, 6–10, 12–14 and 16) reside in the human KSRP sequence (Fig. 3a). Three further peptides (numbers 5, 11 and 15) each contain a single amino acid residue exchange when compared to corresponding parts in KSRP (Fig. 3a). These findings imply that MARTA1 is the rat orthologue of human KSRP.

image

Figure 3. MARTA1 sequence and domain structure. (a) Alignment of rat MARTA1 (GenBank accession number AF308818), human KSRP (hu KSRP) and human FBP (hu FBP). Identical amino acid residues are shown in boxes. Dashes indicate missing amino acid residues. Peptides 1–3 identified by Edman sequencing are underlined. Numbers on the left indicate amino acid positions. (b) MARTA1 domain structure. The amino-terminal proline/glycine rich domain, four putative RNA-binding KH domains and two carboxy-terminal25-amino acid residue imperfect repeats are shown as gray, black and stippled boxes,respectively. Numbers indicate aminoacidpositions.

Download figure to PowerPoint

Table 1.  MARTA1 peptides identified by mass spectrometry
PeptideAmino acid position in MARTA1
  1. All peptides were analysed by tandem mass spectrometry, aexcept where indicated.

4122–143
5123–143
6144–163
7179–191
8192–204
9207–215
10246–252a
11322–332
12333–341
13386–395
14450–463a
15475–479a
16630–647a

Cloning of rat MARTA1 cDNAs

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Subcellular protein fractions, enrichment and affinity purification of MARTA1
  5. In vitro RNA synthesis and UV cross-linking assays
  6. Peptide sequence analysis
  7. PCR and cDNA sequence analysis
  8. Cell culture
  9. Antibodies, immunoprecipitation, immunoblotting and immunohistochemistry
  10. Results
  11. Purification of rat MARTA1
  12. Determination of rat MARTA1 peptide sequences
  13. Cloning of rat MARTA1 cDNAs
  14. Sequence specific RNA-binding
  15. Rat MARTA1 is present in nuclei and the somatodendritic cytoplasm of neurones
  16. Discussion
  17. Acknowledgements
  18. References

The MARTA1 cDNA sequence was determined from overlapping PCR products derived from reverse transcribed adult rat brain total RNA (AF308818; see footnote 1). It consists of 2179 nucleotides including a 2163-nucleotide open reading frame that encodes a hydrophilic 74.2-kDa protein comprising 721 amino acid residues (Fig. 3a). MARTA1 is 98% and 70% identical to human KSRP (Min et al. 1997) and FBP (Duncan et al. 1994), respectively, and exhibits 36% identity to the P-element somatic inhibitor protein (Siebel et al. 1994; Siebel et al. 1995), a tissue-specific splicing factor from Drosophila. All peptide sequences characterized by Edman sequencing and mass spectrometry are present in the deduced amino acid sequence (Fig. 3a and Table 1). MARTA1 contains four KH domains (Fig. 3b), which were first described in the heterogeneous nuclear ribonucleoprotein (hnRNP) K and found to mediateRNA-binding (Siomi et al. 1993, 1994). An amino-terminal domain (amino acid residues 7–69; Fig. 3) possesses a high content of proline and glycine residues (83%) and two 25-amino acid residue imperfect repeat regions (68%identity) are located carboxy-terminal of the KH domains (Fig. 3b).

The high degree of identity between rat MARTA1 and human KSRP was confirmed by western blotting. A polyclonal KSRP antiserum reacted with a 85-kDa protein found in crude lysate, S90 and ribosomal salt wash (RSW) fractions derived from rat brain as well as with purified MARTA1 (Fig. 4a, lanes 1–4, respectively). The immunoreactive protein was highly enriched in the RSW preparation (lane 3) and the eluate from the affinity-column (lane 4). The relative strength of this band in different cell fractions correlates well with the MARTA1 labeling intensity observed in UV cross-linking assays (Fig. 4b). Furthermore, polyclonal KSRP antiserum immunoprecipitated MARTA1 that had been cross-linked to the radioactively labeled MAP2-DTE probe (Fig. 4c, lane 1). No cross-linked proteins remained in the supernatant fraction of the immunoprecipitation (Fig. 4c, lane 2). In contrast, pre-immune serum or protein A Sepharose alone did not precipitate cross-linked MARTA1 (Fig. 4c, lanes 3–6). These data show that rat MARTA1 is the orthologue of human KSRP and interacts directly with the MAP2-DTE.

image

Figure 4. Subcellular fractionation and immunoprecipitation. (a) In a western blot performed with anti-KSRP rabbit antiserum and 10 µg protein each from different rat brain fractions (crude lysate, S90 supernatant and RSW fractions; lanes 1–3, respectively) or 16 µL affinity-purified MARTA1 (lane 4). (b) UV cross-linking assays performed with the DTE-probe and 2.5 µg protein each from crude lysate, S90 supernatant and RSW fractions (lanes 1–3, respectively) and 2 µL eluate from affinity-beads (lane 4). (c) Immunoprecipitation of cross-linked MARTA1. UV cross-linking assays were performed with 20 µg S100 brain proteins and 25 fmoles DTE probe. RNA digested samples were incubated with protein A Sepharose coated with anti-KSRP (lanes 1, 2) or pre-immune serum (lanes 3, 4), respectively, or with uncoated beads (lanes 5, 6). Pellet (lanes 1, 3, 5) and supernatant fractions (lanes 2, 4, 6) were separated on an SDS–polyacrylamide gel, and cross-linked MARTA1 was detected by autoradiography. Position of MARTA1 is marked (arrowhead).

Download figure to PowerPoint

Sequence specific RNA-binding

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Subcellular protein fractions, enrichment and affinity purification of MARTA1
  5. In vitro RNA synthesis and UV cross-linking assays
  6. Peptide sequence analysis
  7. PCR and cDNA sequence analysis
  8. Cell culture
  9. Antibodies, immunoprecipitation, immunoblotting and immunohistochemistry
  10. Results
  11. Purification of rat MARTA1
  12. Determination of rat MARTA1 peptide sequences
  13. Cloning of rat MARTA1 cDNAs
  14. Sequence specific RNA-binding
  15. Rat MARTA1 is present in nuclei and the somatodendritic cytoplasm of neurones
  16. Discussion
  17. Acknowledgements
  18. References

To analyse whether human KSRP exhibits a similar sequence-specific in vitro RNA-binding capacity as rat MARTA1, crude HEK cell lysates were examined. In a western blot, a single 85-kDa anti-KSRP reactive band was detected (Fig. 5, lane 1). In UV cross-linking assays, an 85-kDa protein strongly bound to the MAP2-DTE (Fig. 5, lanes 2 and 6) and fragment uA containing the DTE (lane 3), but did not interact with fragment cE from the MAP2 coding region (lane 4) or the α-tubulin transcript (lane 5). In competition experiments, unlabeled DTE and fragment uA, but not fragment cE or α-tubulin mRNA, drastically interfered with binding of the DTE probe to KSRP (Fig. 5, lanes 6–10, respectively). Thus, rat MARTA1 and human KSRP exhibit similar DTE-specific in vitro RNA-binding.

image

Figure 5. RNA-binding specificity of human KSRP. Lane 1, Western blot with 15 µg crude HEK cell lysate incubated with anti-KSRP antiserum; lanes 2–10, UV cross-linking assays with crude lysate and different RNA probes: MAP2-DTE (lane 2), MAP2 mRNA fragments uA (lane 3) and cE (lane 4), α-tubulin (lane 5); lanes 6–10, crude lysate cross-linked with MAP2-DTE probe in the absence (lane 6) or presence of a 100-fold molar excess of competitor (MAP2-DTE, lane 7; uA, lane 8; cE, lane 9; α-tubulin, lane 10). Position of KSRP is marked (arrowhead).

Download figure to PowerPoint

Rat MARTA1 is present in nuclei and the somatodendritic cytoplasm of neurones

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Subcellular protein fractions, enrichment and affinity purification of MARTA1
  5. In vitro RNA synthesis and UV cross-linking assays
  6. Peptide sequence analysis
  7. PCR and cDNA sequence analysis
  8. Cell culture
  9. Antibodies, immunoprecipitation, immunoblotting and immunohistochemistry
  10. Results
  11. Purification of rat MARTA1
  12. Determination of rat MARTA1 peptide sequences
  13. Cloning of rat MARTA1 cDNAs
  14. Sequence specific RNA-binding
  15. Rat MARTA1 is present in nuclei and the somatodendritic cytoplasm of neurones
  16. Discussion
  17. Acknowledgements
  18. References

Immunostaining of adult rat brain sections and cultured neurones derived from superior cervical ganglia revealed that MARTA1 is highly concentrated in neuronal nuclei incerebral cortex, hippocampus and primary neurones (arrowheads in Figs 6a–c). A weaker immunostaining signal was detected in the cytoplasm of neuronal cell bodies and dendrites (arrows in Figs 6a and b). In primary neurones, the cytoplasmic staining pattern in somata and proximal dendrites was slightly granular (arrows in Fig. 6c). No staining was observed in parallel control experiments without primary antibody (data not shown). Thus, in rat neurones MARTA1 exhibits a somatodendritic distribution with high intranuclear protein concentrations.

image

Figure 6. Distribution of MARTA1 in neurones. Coronal sections from adult rat brain and primary neurones derived from superior cervical ganglia were immunostained with anti-KSRP antiserum. MARTA1 is highly concentrated in nuclei (arrowheads) of pyramidal neurones in the cerebral cortex (a) and hippocampal CA1 area (b) as well as primary neurones (c and d). In the neuronal cytoplasm, MARTA1 is found in somata and dendrites (arrows). (c) In the inset, dendritic MARTA1 granules are visible. Panels are bright light (a and b), fluorescence (c; Alexa488-coupled goat anti-rabbit secondary antibody) and phase contrast (d) micrographs. Scale bars: 25 µm (a and b) and 50 µm (d).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Subcellular protein fractions, enrichment and affinity purification of MARTA1
  5. In vitro RNA synthesis and UV cross-linking assays
  6. Peptide sequence analysis
  7. PCR and cDNA sequence analysis
  8. Cell culture
  9. Antibodies, immunoprecipitation, immunoblotting and immunohistochemistry
  10. Results
  11. Purification of rat MARTA1
  12. Determination of rat MARTA1 peptide sequences
  13. Cloning of rat MARTA1 cDNAs
  14. Sequence specific RNA-binding
  15. Rat MARTA1 is present in nuclei and the somatodendritic cytoplasm of neurones
  16. Discussion
  17. Acknowledgements
  18. References

We have purified MARTA1, a rat brain protein that exhibits high in vitro RNA-binding specificity for the cis-acting MAP2-DTE (Fig. 2; Rehbein et al. 2000). It is 98% identical to human KSRP that was first identified in a protein complex that regulates neural specific splicing of c-src pre-mRNA (Min et al. 1997). In rat liver, KSRP was found in an mRNA editing complex (Lellek et al. 2000). Thus, KSRP appears to be involved in distinct nuclear RNA processing events. A cytoplasmic function of MARTA1 is suggested by its presence in cytosolic and RSW fractions from brain as determined by UV cross-linking assays and western blots (Fig. 4) and its immunohistochemical detection in the somatodendritic cytoplasm of neurones (Fig. 6). The strong MAP2-DTE-specific binding of purified MARTA1 (Fig. 2) hints toward a function in dendritic mRNA localization. This is consistent with a very recent report showing that the predominantly nuclear zipcode binding protein 2, a chicken homologue of MARTA1/KSRP, is involved in cytoplasmic β-actin mRNA targeting in fibroblasts and differentiating neurones (Gu et al. 2002). Further observations in distinct cell systems support the assumption that nuclear RNA-binding proteins play a role in cytoplasmic mRNA localization (Hoek et al. 1998; Cote et al. 1999; Lall et al. 1999; Munro et al. 1999) and that RNA transcription, capping, splicing, polyadenylation and nuclear export are functionally coupled processes (Barabino and Keller 1999; Hirose and Manley 1998; Bentley 1999). In Drosophila, Y14 and Mago seem to associate with nuclear mRNA during splicing and escort it to its final cytoplasmic destination (Hachet and Ephrussi 2001; Le Hir et al. 2001). Similarly, a variety of RNA-binding proteins shuttle between the nucleus and the cytoplasm, indicating a role in nucleocytoplasmic transcript trafficking (Nakielny and Dreyfuss 1997).

Both rat MARTA1 and human KSRP contain four central KH domains. KH domains were first described in hnRNP K and mediate RNA-binding (Siomi et al. 1993, 1994). KH domains consist of a three-stranded antiparallel β-sheet oriented against three α-helices (Braddock et al. 2002). A potential RNA-binding surface of this βααββα fold is exposed in a loop between helices 1 and 2 (Musco et al. 1996). Amino- and carboxy-termini of an individual KH domain are positioned at opposite sides, so that multiple KH domains can be linked together in a single protein. Each of the four KH domains in MARTA1/KSRP may interact with a distinct sequence element within the DTE. Repetitive motifs potentially involved in protein binding have been identified in the vegetal localization element of the Vg1 mRNA (Deshler et al. 1997; Gautreau et al. 1997; Havin et al. 1998). Interestingly, Vg1-RBP/VERA, a trans-factor binding to the vegetal localization element, also contains four KH domains (Deshler et al. 1997; Havin et al. 1998). Moreover, Vg1-RBP/VERA is highly related to zipcode binding protein 1 that regulates β-actin mRNA targeting in chicken neurones, myo- and fibroblasts (Oleynikov and Singer 1998; Bassell et al. 1999; Zhang et al. 2001). In the future, it will be interesting to learn how a specific steric arrangement of several KH domains mediates sequence-specific RNA-binding.

This work was supported by the Deutsche Forschungsgemeinschaft (Ri192–19–1, Ri192–21–1, FOR 296/2–1). We like to thank Dr Joel Yisraeli, Jerusalem, Israel, and Dr Jacob Nielsen, Copenhagen, Denmark, for helpful advice with the protein purification, DrDouglas L. Black, Los Angeles, USA, for the generous gift of the human KSRP expression vector, Eva Stübe, Christiane Schröder-Birkner and Saskia Siegel for excellent technical assistance. This article is in part based on a doctoral study by K. Wege in the Faculty of Biology, University of Hamburg.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Subcellular protein fractions, enrichment and affinity purification of MARTA1
  5. In vitro RNA synthesis and UV cross-linking assays
  6. Peptide sequence analysis
  7. PCR and cDNA sequence analysis
  8. Cell culture
  9. Antibodies, immunoprecipitation, immunoblotting and immunohistochemistry
  10. Results
  11. Purification of rat MARTA1
  12. Determination of rat MARTA1 peptide sequences
  13. Cloning of rat MARTA1 cDNAs
  14. Sequence specific RNA-binding
  15. Rat MARTA1 is present in nuclei and the somatodendritic cytoplasm of neurones
  16. Discussion
  17. Acknowledgements
  18. References
  • Barabino S. M. and Keller W. (1999) Last but not least: regulated poly(A) tail formation. Cell 99, 911.
  • Bassell G. J., Oleynikov Y. and Singer R. H. (1999) The travels of mRNAs through all cells large and small. FASEB J. 13, 447454.
  • Bentley D. (1999) Coupling RNA polymerase II transcription with pre-mRNA processing. Curr. Opin. Cell Biol. 11, 347351.
  • Blichenberg A., Schwanke B., Rehbein M., Garner C., Richter D. and Kindler S. (1999) Identification of a cis-acting dendritic targeting element in MAP2 mRNAs. J. Neurosci. 19, 88188829.
  • Braddock D. T., Louis J. M., Baber J. L., Levens D. and Clore G. M. (2002) Structure and dynamics of KH domains from FBP bound to single-stranded DNA. Nature 415, 10511056.DOI: 10.1038/4151051a
  • Cote C. A., Gautreau D., Denegre J. M., Kress T. L., Terry N. A. and Mowry K. L. (1999) A Xenopus protein related to hnRNP I has a role in cytoplasmic RNA localization. Mol. Cell 4, 431437.
  • Deshler J. O., Highett M. I. and Schnapp B. J. (1997) Localization of Xenopus Vg1 mRNA by Vera protein and the endoplasmic reticulum. Science 276, 11281131.
  • Duncan R., Bazar L., Michelotti G., Tomonaga T., Krutzsch H., Avigan M. and Levens D. (1994) A sequence-specific, single-strand binding protein activates the far upstream element of c-myc and defines a new DNA-binding motif. Genes Dev. 8, 465480.
  • Gautreau D., Cote C. A. and Mowry K. L. (1997) Two copies of a subelement from the Vg1 RNA localization sequence are sufficient to direct vegetal localization in Xenopus oocytes. Development 124, 50135020.
  • Gu W., Pan F., Zhang H., Bassell G. J. and Singer R. H. (2002) A predominantly nuclear protein affecting cytoplasmic localization of β-actin mRNA in fibroblasts and neurons. J. Cell Biol. 156, 4151.
  • Hachet O. and Ephrussi A. (2001) Drosophila Y14 shuttles to the posterior of the oocyte and is required for oskar mRNA transport. Curr. Biol. 11, 16661674.
  • Havin L., Git A., Elisha Z., Oberman F., Yaniv K., Schwartz S. P., Standart N. and Yisraeli J. K. (1998) RNA-binding protein conserved in both microtubule- and microfilament-based RNA localization. Genes Dev. 12, 15931598.
  • Heidebrecht H. J., Buck F., Steinmann J., Sprenger R., Wacker H. H. and Parwaresch R. (1997) p100: a novel proliferation-associated nuclear protein specifically restricted to cell cycle phases S, G2, and M. Blood 90, 226233.
  • Hirose Y. and Manley J. L. (1998) RNA polymerase II is an essential mRNA polyadenylation factor. Nature 395, 9396.
  • 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.
  • Jansen R. P. (2001) mRNA localization: message on the move. Nat. Rev. Mol. Cell Biol. 2, 247256.DOI: 10.1038/35067016
  • Job C. and Eberwine J. (2001) Localization and translation of mRNA in dendrites and axons. Nat. Rev. Neurosci. 2, 889898.
  • Kiebler M. A. and DesGroseillers L. (2000) Molecular insights into mRNA transport and local translation in the mammalian nervous system. Neuron 25, 1928.
  • Kreienkamp H. J., Sine S. M., Maeda R. K. and Taylor P. (1994) Glycosylation sites selectively interfere with α-toxin binding to thenicotinic acetylcholine receptor. J. Biol. Chem. 269, 81088114.
  • Lall S., Francis-Lang H., Flament A., Norvell A., Schupbach T. and Ish-Horowicz D. (1999) Squid hnRNP protein promotes apical cytoplasmic transport and localization of Drosophila pair-rule transcripts. Cell 98, 171180.
  • Le Hir H., Gatfield D., Braun I. C., Forler D. and Izaurralde E. (2001) The protein Mago provides a link between splicing and mRNA localization. EMBO Rep. 2, 11191124.
  • Lellek H., Kirsten R., Diehl I., Apostel F., Buck F. and Greeve J. (2000) Purification and molecular cloning of a novel essential component of the apolipoprotein B mRNA editing enzyme-complex. J. Biol. Chem. 275, 1984819856.
  • Min H., Turck C. W., Nikolic J. M. and Black D. L. (1997) A new regulatory protein, KSRP, mediates exon inclusion through an intronic splicing enhancer. Genes Dev. 11, 10231036.
  • Monshausen M., Putz U., Rehbein M., Schweizer M., DesGroseillers L., Kuhl D., Richter D. and Kindler S. (2001) Two rat brain Staufen isoforms differentially bind RNA. J. Neurochem. 76, 155165.DOI: 10.1046/j.1471-4159.2001.00061.x
  • Munro T. P., Magee R. J., Kidd G. J., Carson J. H., Barbarese E., Smith L. M. and Smith R. (1999) Mutational analysis of a heterogeneous nuclear ribonucleoprotein A2 response element for RNA trafficking. J. Biol. Chem. 274, 3438934395.
  • Musco G., Stier G., Joseph C., Castiglione Morelli M. A., Nilges M., Gibson T. J. and Pastore A. (1996) Three-dimensional structure and stability of the KH domain: molecular insights into the fragile X syndrome. Cell 85, 237245.
  • Nakielny S. and Dreyfuss G. (1997) Nuclear export of proteins and RNAs. Curr. Opin. Cell Biol. 9, 420429.
  • Oleynikov Y. and Singer R. H. (1998) RNA localization: different zipcodes, same postman? Trends Cell Biol. 8, 381383.
  • Rehbein M., Kindler S., Horke S. and Richter D. (2000) Two trans-acting rat brain proteins, MARTA1 and -2, interact specifically with the dendritic targeting element in MAP2 mRNAs. Mol. Brain Res. 79, 192201.
  • Siebel C. W., Kanaar R. and Rio D. C. (1994) Regulation of tissue-specific P-element pre-mRNA splicing requires the RNA-binding protein PSI. Genes Dev. 8, 17131725.
  • Siebel C. W., Admon A. and Rio D. C. (1995) Soma-specific expression and cloning of PSI, a negative regulator of P-element pre-mRNA splicing. Genes Dev. 9, 269283.
  • Siomi H., Matunis M. J., Michael W. M. and Dreyfuss G. (1993) The pre-mRNA binding K protein contains a novel evolutionarily conserved motif. Nucleic Acids Res. 21, 11931198.
  • Siomi H., Choi M., Siomi M. C., Nussbaum R. L. and Dreyfuss G. (1994) Essential role for KH domains in RNA binding: impaired RNA binding by a mutation in the KH domain of FMR1 that causes fragile X syndrome. Cell 77, 3339.
  • Steward O. and Schuman E. M. (2001) Protein synthesis at synaptic sites on dendrites. Annu. Rev. Neurosci. 24, 299325.
  • Wells D. G., Richter J. D. and Fallon J. R. (2000) Molecular mechanisms for activity-regulated protein synthesis in the synapto-dendritic compartment. Curr. Opin. Neurobiol. 10, 132137.
  • Zhang H. L., Eom T., Oleynikov Y., Shenoy S. M., Liebelt D. A., Dictenberg J. B., Singer R. H. and Bassell G. J. (2001) Neurotrophin-induced transport of a β-actin mRNP complex increases β-actin levels and stimulates growth cone motility. Neuron 31, 261275.