This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Soller, M.
Right arrow Articles by White, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Soller, M.
Right arrow Articles by White, K.

 Previous Article  |  Next Article 

Molecular and Cellular Biology, September 2005, p. 7580-7591, Vol. 25, No. 17
0270-7306/05/$08.00+0     doi:10.1128/MCB.25.17.7580-7591.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

ELAV Multimerizes on Conserved AU4-6 Motifs Important for ewg Splicing Regulation

Matthias Soller* and Kalpana White*

Department of Biology and Volen Center for Complex Systems, MS 008, Brandeis University, Waltham, Massachusetts 02454

Received 23 February 2005/ Returned for modification 21 March 2005/ Accepted 27 May 2005


arrow
ABSTRACT
 
ELAV is a gene-specific regulator of alternative pre-mRNA processing in Drosophila neurons. Since ELAV/Hu proteins preferentially bind to AU-rich regions that are generally abundant in introns and untranslated regions, it has not been clear how gene specificity is achieved. Here we used a combination of in vitro biochemical experiments together with phylogenetic comparisons and in vivo analysis of Drosophila transgenes to study ELAV binding to the last ewg intron and splicing regulation. In vitro binding studies of ELAV show that ELAV multimerizes on the ewg binding site and forms a defined and saturable complex. Further, sizing of the ELAV-RNA complex and a series of titration experiments indicate that ELAV forms a dodecameric complex on 135 nucleotides in the last ewg intron. Analysis of the substrate RNA requirements for ELAV binding and complex formation indicates that a series of AU4-6 motifs spread over the entire binding site are important, but not a strictly defined sequence element. The importance of AU4-6 motifs, but not spacing between them, is further supported by evolutionary conservation in several melanogaster species subgroups. Finally, using transgenes we demonstrate in fly neurons that ELAV-mediated regulation of ewg intron 6 splicing requires several AU4-6 motifs and that introduction of spacer sequence between conserved AU4-6 motifs has a minimal effect on splicing. Collectively, our results suggest that ELAV multimerization and binding to multiple AU4-6 motifs contribute to target RNA recognition and processing in a complex cellular environment.


arrow
INTRODUCTION
 
Expression of eukaryotic genes follows a highly ordered sequence of events from initiation of transcription to translation of a mature mRNA within various subcellular compartments. All along this process a large number of different RNA binding proteins dynamically interact with target RNA to affect all aspects of the biogenesis of a mRNA and its translation by associating into large particles (for reviews, see references 14, 18, 30, 39, and 61). Elucidation of mechanisms that allow RNA binding proteins to specifically recognize the complements of their target RNAs in a complex cellular environment is central to understand the role that RNA binding proteins play in posttranscriptional gene regulation.

RNA binding proteins have a modular composition consisting of one or more RNA binding domains and, in most cases, at least one additional auxiliary domain (18, 30, 39, 61). The RNA recognition motif (RRM) is the most prevalent RNA binding domain and also one of the most frequent protein domains in eukaryotes (3, 10, 17, 18). The ~80-amino-acid RRM contains two highly conserved short motifs, RNP1 and RNP2, and folds into a characteristic structure of a four-stranded antiparallel ß-sheet and two {alpha}-helices that binds 4 to 7 nucleotides (nt) (59). Binding specificities for many RRM-containing proteins have been studied in detail with short sequences in vitro (18, 52, 61). Predicting sequence-specific binding in a larger context, however, is difficult due to sequence redundancy and therefore relies mainly on experimental determination of binding sites (see, e.g., references 9, 20, 35, 57, and 58).

The ELAV/Hu family of RNA binding proteins is characterized by three RRMs and by their predominant expression in neurons (4, 42). ELAV (embryonic lethal abnormal visual system), originally identified in Drosophila (48), has been shown to affect neuronal pre-mRNA processing as a gene-specific regulator (29). In contrast, the neuron-specific members of the human ELAV family, HuB (Hel-N1), HuC, and HuD, similar to the ubiquitously expressed HuA (HuR), have been attributed only cytoplasmic RNA-processing functions (6, 9, 19, 31, 33). Common to ELAV/Hu proteins is their preferential binding to AU-rich sequences that are also frequently clustered, as, e.g., found in the c-myc 3' untranslated region (UTR) (1, 20, 32, 50). Since AU-rich sequences are abundant in intronic and untranslated regions, it has not been clear how ELAV/Hu proteins achieve gene-specific binding resulting in distinct phenotypes in neuronal development and plasticity (6, 36, 43, 45).

ELAV is essential for producing neuroglian (nrg) and erect wing (ewg) neuronal mRNAs, which require splicing of the terminal intron, resulting in distinct 3' ends (27, 34, 55). In ewg splicing regulation, ELAV's effect is attributed entirely to the inhibition of 3'-end processing, as no reduction in splicing efficiency is seen when the ELAV binding site is mutated in the absence of a functional poly(A) site at pA2 (55). ELAV binds to a stretch of AU-rich sequence (divided into elements m1, m2, and m3) (Fig. 1E) extending over 135 nucleotides 3' of the regulated poly(A) site (pA2) and inhibits intronic 3'-end processing to allow neural splicing (Fig. 1D) (55).



View larger version (39K):
[in this window]
[in a new window]
  		FIG. 1. ELAV binds as a multimer to ewg RNA pA2-I. (A) Schematic of ELAV and the N-terminal truncation mutant RBD60. (B) Coomassie blue-stained SDS gel of recombinant ELAV (50 kDa) and RBD60 (40 kDa). Protein was loaded at 1.25, 2.5, 5, and 10 µg. Marker proteins were bovine serum albumin (66 kDa), ovalbumin (45 kDa), and carbonic anhydrase (29 kDa). (C) Binding of recombinant ELAV and RBD60 to poly(U) RNA captured by DEAE-Sepharose beads and detection by Western blotting. sn, supernatant. (D) Schematic of the last ewg intron 6 and 3' UTR with splicing and polyadenylation choices. The two major isoforms are those with intron 6 spliced from H to J in the presence of ELAV (solid line) or with polyadenylation in intron 6 at pA2 (boldface) in the absence of ELAV. (E) Magnification of the ELAV binding region used as substrate RNA pA2-I, depicting polyadenylation site pA2 and part of exon I as well as AU4-6 motifs in sequence elements m1, m2, and m3. Py, polypyrimidine tract. (F) EMSA of ewg RNA pA2-I with ELAV and RBD60 protein. Uniformly 32P-labeled RNA pA2-I (100 pM) was incubated with recombinant ELAV or recombinant RBD60 (1.38 nM, 5.5 nM, 22 nM, 87.5 nM, 0.35 µM, and 1.4 µM) and separated on 4% native polyacrylamide gels. Lane 1 shows input RNA without protein. (G) Graphic representation of EMSA data from ELAV and RBD60 binding to RNA pA2-I. The percentage of bound RNA (input RNA – unbound RNA/input RNA x 100) is plotted against the concentration of recombinant ELAV or RBD60. Results are from three experiments. Error bars indicate standard deviations.

Here we analyze the properties of an ELAV complex and its binding site on ewg RNA in relation to ELAV-controlled splicing regulation of the last ewg intron in neurons. We demonstrate that ELAV assembles on ewg RNA pA2-I to a dodecameric complex. Further, a combination of in vitro and in vivo analyses along with phylogenetic comparisons allows us to conclude that the ewg ELAV complex binding site consists of a number of spaced AU4-6 motifs spread over 135 nt. Furthermore, in vitro binding data support that functionally conserved AU4 motifs in the 3' region of the 135-nt ewg binding site act as initiation sites for ELAV binding and complex formation.


arrow
MATERIALS AND METHODS
 
Fly genetics and recombinant DNA technology. Fly breeding, genetics and transformation, and recombinant DNA technology were according to standard procedures as detailed by Soller and White (55). Additional mutations in the ELAV binding site of ewg were generated by site directed mutagenesis using the following oligonucleotides (antisense): spacer A, CAAAAAATGTATATTCCTGGTTGTTATTAATTTTTATTTTTAATTATGAGACCAACAATAATTAAAAATAAAAATTAATTTGAAG; spacer B, CAAAATATACTTACATATTATATTTTACTCCTGTTCGTTTTTTACATAACTCAATATAAAATGAGGACAAG; and spacer C, CTGGTTAAAAGAAAAGAACATAAAGTATAAAATTATATTCTATTTTACATATTATCGTTTTATATGAAGATAAAATGTATAATAG. pA2-I contains the sequence after the AAUAAA up to 6 nucleotides of exon I in a modified pB SK+, adding the vector sequence GGGCGAATTGGGTACGATCCTCTA-ewg pA2-I-GGCCGCCACCGCGGTGGAG to the ewg substrate when transcribing the Ecl136 II (Fermentas)-linearized plasmid with T7 RNA polymerase. Deletions in pA2-I were made by cloning overlapping oligonucleotides into the parental vector.

Drosophila species were obtained from the Tucson Stock Center, and evolutionary relationships can be viewed at http://flybase.bio.indiana.edu/allied-data/lk/phylogeny/Drosophilidae-Tree/. The ELAV binding site in ewg was amplified by inverse PCR. Briefly, DNA from ~10 flies was extracted as described by Koushika et al. (28), and 1/10 was digested with either AseI or HypCH4 IV. After heat inactivation of restriction enzymes, the DNA was ligated in 400 µl overnight at 4°C. PCR was done on 1/10 of the ligated DNA with primers ewgInv F1 (GTCGCCTGGAATGCACCAGATGATGATC) and ewgInv R1 (CTCCAGCTGRTCGCCATTCTCCAC) for 35 cycles (94°C for 30 s, 58°C for 60 s, and 72°C for 90 s, with an initial denaturation step at 94°C for 30 s and a final extension at 72°C for 5 min).

EMSA, gel filtration, and ELAV protein cross-linking. Production of recombinant proteins and electrophoretic mobility shift assays (EMSAs) were as described by Soller and White (55). To test for protein activity, binding assays were as in EMSAs except that poly(U) RNA (Amersham) was used as the substrate, 0.05% NP-40 was added to buffers, and DEAE-Sepharose beads (Amersham) were blocked by 10% serum.

For gel filtration on Superose 6 (HR 10/30; Amersham), complexes were formed in the same way as for EMSA and run in 150 mM NaCl, 0.5x Tris-borate-EDTA at 4°C.

ELAV protein cross-linking was done in 75 mM HEPES (pH 7.3),120 mM NaCl, 2.5 mM EDTA in a final volume of 20 µl by adding 1 µl 1,8-bis-maleimidotriethyleneglycol (BM[PEO]3) (Pierce) in water (at 1:1, 1:3, and 1:9 ratios of sulfhydryl groups to cross-linker). After 30 min at room temperature, the reaction was quenched for 15 min with 2 µl 1 M glycine, 7.5 mM Na2HPO4, 250 mM dithiothreitol; the reaction mixture was mixed with sample buffer, and an aliquot was run on a sodium dodecyl sulfate (SDS) gel.

Boundary experiment. For 5' end labeling, substrate RNA was in vitro transcribed using a guanosine (Sigma)-to-GTP ratio of 8:1 according to standard procedures. Transcripts were 5' labeled with T4 polynucleotide kinase (NEB) and [{gamma}-32P]ATP (5,000 mCi; NEN) and gel purified. 3'-end labeling was done with [3'-32P]dATP (Cordycepin, 3,000 mCi; NEN) and yeast poly(A) polymerase (Amersham) after removing free nucleotides with a G-50 spin column (Amersham).

Substrate RNA (2 fmol) for boundary experiments was ~50% hydrolyzed (in 50 mM NaHCO3-Na2CO3, pH 9.4, at 96°C for 5 to 10 min); complexes were formed as for EMSA analysis with an ELAV concentration of 2 µM and run on an agarose-polyacrylamide composite gel (40). ELAV complexes were cut out, agarose was digested with agarase, and the RNA was phenol-chloroform extracted and ethanol precipitated.

RNA extraction, RT-PCR, and RPA. RNA extraction and reverse transcription-PCR (RT-PCR) were done as described by Soller and White (55), and RNase protection assays (RPAs) were according to standard procedures. Nuclear extract preparation and UV-cross-linking assays were as described by Soller and White (55).


arrow
RESULTS
 
ELAV forms a dodecameric complex on ewg pre-mRNA. We previously noted that several ELAVs form a complex upon binding to substrate RNA in EMSAs and that the complex formation concentration was dependent on the sequence of the substrate RNA (55). Here we analyzed the stoichiometry of this ELAV complex that binds to ewg substrate RNA pA2-I (from pA2 to exon I [Fig. 1E]) and determined its RNA binding properties. For this analysis we used full-length ELAV and an N-terminally truncated ELAV mutant, RBD60, a hypomorph able to rescue lethal alleles of elav (Fig. 1A) (63). Recombinant ELAV and RBD60 were bacterially produced as glutathione S-transferase fusion proteins and proteolytically cleaved off the glutathione S-transferase moiety (Fig. 1B). Both ELAV and RBD were about 90% active in a poly(U) binding assay (Fig. 1C). In EMSAs, the truncated RBD60 protein formed a faster-migrating complex than the ELAV complex, but otherwise dissociation constants (17 nM and 25 nM for ELAV and RBD60, respectively) were comparable and an initial complex formed at around 350 nM of ELAV or RBD60 (Fig. 1F [compare lanes 6 and 7 and lanes 12 and 13] and G,). Additionally, several intermediate shift products were detectable with both proteins, indicating that smaller ELAV- or RBD60-RNA complexes run at intermediate positions (Fig. 1F, lanes 4, 5, 10, and 11).

Next, to obtain an estimate of the size of the ELAV complex bound to ewg RNA, we performed size exclusion chromatography experiments using 32P-labeled RNA pA2-I complexed with recombinant, N-terminally hemagglutinin (HA)-tagged ELAV (54 kDa) or RBD60 (40 kDa) at saturating concentrations (Fig. 2A and B). The RNA-ELAV complex runs at a size of about 714 kDa as estimated from prerun standards under the same conditions. Most free ELAV protein runs at 229 kDa, with minor peaks at 95 kDa and 52 kDa. ELAV without RNA resulted in the same major peak at the size of 237 kDa, with minor peaks at 104 kDa and 57 kDa (Fig. 2C). The RBD60-RNA complex runs at 498 kDa, and free RBD60 protein runs in three major peaks at sizes of 167 kDa, 67 kDa, and 33 kDa (Fig. 2B). RBD60 without RNA resulted in two major peaks at sizes of 176 kDa and 73 kDa and a minor peak at 40 kDa (Fig. 2C). Free RNA runs above its expected size of 72 kDa, at 198 kDa, due to the elongated nature of this 213-nt RNA (Fig. 2C). Although retention on gel filtration columns is also influenced by the shape of a protein (complex), the presence of a major peak of free ELAV at around 216 kDa together with minor peaks at around 108 kDa and 54 kDa, and accordingly reduced sizes for RBD60, suggests that free ELAV forms a tetramer in solution. The free RBD60 tetramer is less stable than the full-length ELAV tetramer under these conditions, which indicates an involvement of the A/Q-rich N terminus in tetramer stability.



View larger version (35K):
[in this window]
[in a new window]
  		FIG. 2. ELAV tetramerizes and forms a 700-kDa complex with ewg RNA pA2-I. (A and B) Superose 6 gel filtration of ELAV or RBD60 complexes. ELAV (A) or RBD60 (B) RNA-pA2-I complexes were formed with 10 fmol 32P-labeled RNA pA-I and 320 pmol ELAV or RBD60 in 100 µl and loaded onto the gel filtration column. The top panels show an exposure of the RNA elution profile and the middle panels a Western blot of the elution profile for ELAV (A) or RBD60 (B). The bottom panels show a quantification of RNA (dashed line) and ELAV (A) or RBD60 (B) elution profiles (solid line) in conjunction with prerun standards. Peaks of free ELAV or RBD60 are numbered according to their multimerization state (1, monomer; 2, dimer; and 4, tetramer). The protein peak is slightly retarded compared to the RNA peak, which might be due to the length of the run and substoichiometric stability of the complex (see Fig. 3A and B). (C) Superose 6 gel filtration of RNA pA2-I, ELAV, or RBD60 alone. The top panel shows an exposure of the RNA elution profile, and the two panels below show Western blots of the elution profile for ELAV or RBD60. The bottom panel shows a quantification of RNA (dotted line) and ELAV (solid line) or RBD60 (dashed line) elution profiles in conjunction with prerun standards. Peaks of free ELAV or RBD60 are numbered according to their multimerization state (1, monomer; 2, dimer; and 4, tetramer). (D) Determination of multimerization states of ELAV and RBD60 by cross-linking. Recombinant ELAV and RBD60 were cross-linked at two concentrations (3 µM and 0.75 µM) with BM[PEO]3 in a maleimide reaction starting at equimolar concentrations over sulfhydryl groups (6, 18, and 36 µM and 1.5, 4.5 and 13.5 µM, respectively). Proteins were separated on a 7% SDS gel and detected by Western blotting with an anti-ELAV antibody. Lanes 1 and 8 show input protein without cross-linker (same preparation as in Fig. 1B), and the multimerization state is labeled at the right of ELAV and RBD60 protein lanes, respectively. Since about 5 to 10% of a shorter proteolysis fragment is present, the lower of the two dimer bands might include this fragment, which is, however, not observed with tri- and tetramer adducts. The positions of the two cysteines in ELAV are depicted in a schematic of ELAV above the gel.

To verify the tetramer configuration of ELAV in solution, we physically cross-linked recombinant ELAV and RBD60. The two cysteine residues in ELAV (Fig. 2D) were used to cross-link ELAV or RBD60 in a maleimide reaction with BM[PEO]3 starting at equimolar amounts (1x, 3x, and 9x), and cross-linked products were then detected on Western blots. With both ELAV and RBD60 three major band doublets corresponding to dimer, trimer, and tetramer sizes were detected at a concentration of 3 µM (Fig. 2D, lanes 2 to 4 and 9 to 11). At a lower concentration of 0.75 µM (and a fourfold reduced cross-linker concentration), the same bands were detected with ELAV, while RBD60 cross-linked predominantly as dimer, consistent with its reduced tetramer stability seen in gel filtration experiments (Fig. 2D [compare lanes 5 to 7 with 12 to 14]; Fig. 2B and C). Since cross-linked products run in doublets, they likely represent different configurations of cross-linked ELAV and RBD60. Results comparable to those with BM[PEO]3 were obtained with other cross-linkers that preferentially target primary amines in lysine residues (e.g., glutaraldehyde). In these cases, however, the dimer product was more prominent, and increasing cross-linker concentrations resulted in degradation of ELAV or RBD60 due to attacks on peptide bonds (data not shown).

To gain further information about the stoichiometry of the ELAV complex and how ELAV assembles on its ewg target RNA, EMSAs were performed well above the dissociation constant. Under these conditions, formation of the ELAV complex is stoichiometric and the final ELAV-RNA complex is observed around a 1:12 ratio of RNA to ELAV or RBD60 (Fig. 3A and B, lanes 9). As cooperative interactions among ELAV proteins compensate for a substoichiometric composition (Fig. 3A and B), a defined complex forms at around a 1:9 ratio but is further retarded up to a 1:12 ratio (Fig. 3A and B and data not shown). With both ELAV and RBD60, a total of six bands in addition to the input RNA were detected (Fig. 3A and B, lanes 2 to 8), suggesting stepwise addition of dimers with each additional band up to a final complex of around 12 ELAVs. Plotting the amount of bound RNA against the equivalents of the protein-to-RNA ratio further indicates that assembly of the ELAV complex on ewg RNA occurs mostly as dimer units (Fig. 3C). The RBD60 protein associates more readily with ewg pA2-I RNA than ELAV (Fig. 3C), likely as a consequence of the reduced tetramer stability of RBD60 compared to ELAV. Although a semistable intermediate of ELAV at the tetramer size is observed (Fig. 3A, lane 7), recruitment of ELAV into the ELAV RNA complex clearly involves rearrangement of the ELAV tetramer. Intermediate RBD60-RNA complexes appear not as sharp as the final RBD60 RNA complex in EMSAs, likely reflecting a decreased stability of these complexes (Fig. 3B).



View larger version (50K):
[in this window]
[in a new window]
  		FIG. 3. ELAV binds as a dodecamer to one ewg RNA pA2-I. (A and B) Stoichiometry EMSA of ewg RNA pA2-I with ELAV or RBD60 protein. Recombinant ELAV (A) or RBD60 (B) was incubated with trace 32P-labeled RNA pA2-I (0.2 µM) well above the dissociation constant (17 nM) and separated on 4% native polyacrylamide gels. Protein concentrations are shown above the gel. Lane 1 shows input RNA without protein. The arrow in A points towards a semistable intermediate probably representing a tetramer. The final complex forms at an RNA/protein ratio of between 1:8 and 1:9 (compare lanes 8 and 9 in A and B, and data not shown). (C) Graphic representation of EMSA data from ELAV and RBD60 binding to RNA pA2-I above the dissociation constant as shown in A and B. The percentage of bound RNA (input RNA – unbound RNA/input RNA x 100) is plotted against the molar equivalent of protein to RNA. (D) Titration EMSA of ELAV against RBD60 with RNA pA2-I at complex-forming concentrations. Uniformly 32P-labeled RNA pA2-I (100 pM) was incubated with recombinant ELAV, recombinant RBD60, or a combination of both at a final concentration of 3.2 µM. The recombinant ELAV concentrations in successive lanes are 0%, 5%, 10%, 20%, 40%, 60%, 80%, and 100% (the concentrations of RBD60 are accordingly reduced). The first lane shows the complex with RBD60 alone, and the last lane shows the complex with ELAV alone. Complexes were separated on 4% native polyacrylamide gels. (E) EMSA of ELAV with two separable substrate RNAs. Uniformly 32P-labeled RNA pA2-I (100 pM), two-copy RNA pA2-I (100 nM), or a mixture of both (50 pM each) was incubated with recombinant ELAV at a concentration of 3.2 µM (lanes 4 to 6) and separated on 4% native polyacrylamide gels. Input RNAs are shown in lanes 1 to 3.

To exclude cooperative interactions among ELAVs for compensating for a substoichiometric composition of the final ELAV complex (Fig. 3A and B), ELAV was titrated against RBD60 at complex-forming concentrations (Fig. 3D) to determine the number of ELAV molecules present in the ELAV complex formed on ewg RNA pA2-I. EMSA analysis showed that at 100% RBD60, only the fast-migrating complex was formed, but with increasing ELAV concentration, a ladder of complexes was observed, with all the RBD60 complex almost disappearing at 40%ELAV (Fig. 3D). A closer inspection reveals 13 discrete complexes, suggesting that the final ELAV complex on ewg RNA pA2-I contains 12 ELAV molecules (Fig. 3D).

Finally, to show that only one RNA molecule is present in the final ELAV complex, we used a longer RNA containing two copies of the ewg ELAV complex-forming site pA2-I, mixed it with the regular one-copy pA2-I substrate RNA, and complexed this mixture with ELAV (Fig. 3E). In EMSAs, two ELAV complexes separated, as expected for complexes that contain only one RNA (Fig. 3E, lane 6).

The ELAV complex binding site on ewg RNA covers over 135 nucleotides. Parts of the RNA from pA2-I can be aligned (elements m1, m2, and m3) (55), and mutations in each of these elements reduced ELAV binding in nuclear extracts in UV-cross-linking assays (55). Therefore, these elements could potentially represent three individual binding sites, one for each ELAV tetramer. To address this issue, we made a series of end deletions in substrate RNA pA2-I and tested if they form complexes of intermediate size in EMSAs (Fig. 4). The length of substrate RNAs ranges from 67 to 213 nt, including 43 nt of vector sequence (see Materials and Methods for details).



View larger version (59K):
[in this window]
[in a new window]
  		FIG. 4. Multiple sequence elements spread over 135 nucleotides contribute to ELAV complex binding on ewg RNA. (A) Schematic of ewg substrate RNA pA2-I (from polyadenylation site pA2 to exon I). AU4-6 motifs are indicated as ovals in sequence elements m1, m2, and m3. Py, polypyrimidine tract. (B) Deletions in RNA pA2-I used for EMSA analysis. RNAs used were from 67 to 213 nt long, including 43 nt of vector sequence (see Materials and Methods for sequence). Averaged binding affinities from at least three EMSA experiments are shown on the right. (C and D) Representative gels of EMSA analysis using various uniformly 32P labeled substrate RNAs (100 pM) as indicated at the top. ELAV concentrations were 12.5 nM, 50 nM, 0.2 µM, 0.8 µM, and 3.2 µM, except 3.2 µM for pA2-I + and {Delta}134 (lanes 19 and 20 in D). Absence of ELAV is indicated above the first lane of each substrate RNA. Complexes were separated on 4% native polyacrylamide gels. (E) Graphic representation of EMSA data. The percentage of bound RNA (input RNA – unbound RNA/input RNA x 100) is plotted against the concentration of recombinant ELAV. Results are from at least three experiments. Error bars indicate standard deviations.

Deleting different parts of substrate RNA pA2-I results in either ELAV complexes of comparable size, indicating recruitment of a fixed number of ELAV molecules (Fig. 4C, lanes 5, 6, 11, 12, 17, and 18, and D, lanes 5, 6, 11, 12, and 17 to 19), or absence of binding and complex formation (RNAs {Delta}134 and {Delta}5 to {Delta}7) (Fig. 4B and D, lane 20). In contrast, input RNAs separate according to their size (Fig. 4C, lanes 1, 7, and 13, and D, lanes 1, 7, 13, and 20). Further, deletion of AU4-6 motifs reduces binding affinity, which is indicated by an increasing Kd as determined from EMSA (Fig. 4B and E). In addition, an increasing slope of the binding curve with shorter RNAs suggests that cooperativity among ELAV compensates for suboptimal binding sites in these shorter RNAs (Fig. 4E). AU4-6 motifs in the m3 element and the polypyrimidine tract, however, are more important for full-affinity binding of ELAV than AU4-6 motifs in the m1 and m2 element. Since ELAV forms a similar-size complex on substrate RNA Py (67 nt), sequence composition matters and those RNAs which do not bind ELAV are not simply too short (RNAs {Delta}134 and {Delta}5 to {Delta}7, 70 to 74 nt) (Fig. 4B and D and data not shown). In summary, these experiments suggest that RNA from pA2 to exon I consists of a single ELAV complex binding site leading to formation of a dodecameric ELAV complex.

ELAV complex formation requires no unique binding element. Next, we wanted to determine the part of the RNA pA2-I that is covered by the ELAV complex. To map the boundaries, pA2-I RNA was either 5' or 3' end labeled, partially hydrolyzed to a single-nucleotide ladder, and complexed with ELAV. The ELAV complex was then purified from an EMSA gel, and the RNA was extracted and analyzed on a denaturing gel. ELAV forms the same sized complex with partially hydrolyzed RNA as with full-length pA2-I RNA (Fig. 4D, lanes 5, 6, 11, 12, and 17 to 19), and no clearly defined intermediate-sized complexes appear that could be purified (data not shown).

Analysis of the RNA complexed with ELAV revealed that with both 3'- and 5'-labeled substrate, ELAV complex formation starts with relatively short RNAs (Fig. 5A and B). 5'-labeled RNA is present after a length of about 58 nt (Fig. 5A, lane 4), and 3' labeled RNA is present after 43 nt (Fig. 5B, lane 4). Labeling intensity increases continuously with the length of the RNA for both end-labeled substrates. Particularly, for 5'-labeled substrate affinities to bind RNAs dramatically increase after the m1 element (Fig. 5A, lane 4), and for 3'-labeled substrates they increase after the pyrimidine tract (Fig. 5B, lane 4). Comparable results were obtained with end-labeled RNA {Delta}12 (Fig. 4B); namely, 5'-labeled RNA is present after a length of about 47 nt (after the m3l element) and 3' labeled RNA also after 43 nt, as was observed with pA2-I RNA (data not shown). In conclusion, overlapping of 5' and 3' selected RNAs indicates the absence of a unique sequence element for binding (Fig. 5C). Unlike for other RNA binding proteins, which bind to a specific motif in a longer sequence context (e.g., Nova or FMRP) (15, 16), length in combination with uridine repeats seems to be important for ELAV complex formation in this assay.



View larger version (40K):
[in this window]
[in a new window]
  		FIG. 5. ELAV complex boundary experiment with substrate RNA pA2-I. (A and B) Partially hydrolyzed 5' (A) or 3' (B) 32P-labeled substrate RNA pA2-I was extracted from gel-purified ELAV complexes and analyzed on 5% denaturing gels. Markers are partial RNase A (lanes 1) or RNase A and T1 (lanes 2) digests of pA2-I RNA or a single-nucleotide ladder of pA2-I RNA (lanes 3). Positions within the substrate RNA are indicated on the left, and sequence landmarks are shown on the right. Binding by the ELAV complex is indicated by a black bar for 5'-labeled substrate RNA and by a gray bar for 3'-labeled substrate RNA on the right sides of A and B, respectively. (C) Schematic summary of the boundary experiment. RNAs bound by the ELAV complex are indicated as black bars for 5'-labeled substrate RNA above the pA2-I sequence and as gray bars for 3'-labeled substrate RNA below the pA2-I sequence. The starts of the bars indicate the minimal lengths of the RNA required for complex formation. Poly(U) motifs are in boldface. Vector sequence is indicated by dashed lines and is shown in Materials and Methods. Note that 5'- and 3'-selected RNAs largely overlap and thus indicate absence of a unique sequence element for binding.

Phylogenetic comparison of the ewg ELAV complex binding site. To independently identify regulatory elements, we sequenced a series of closely related species from the Drosophila melanogaster group for phylogenetic comparisons. This allows tracking of the evolutionary drift of the ELAV binding site and identification of conserved regulatory elements, as regions unimportant to ewg intron 6 splicing will start to diverge first. In the distantly related species Drosophila virilis, the part of ELAV essential for RNA binding (RBD60) is 100% conserved (64), and splicing of ewg intron 6 and EWG protein expression also are conserved (M. Soller, I. Haussmann, D. Motola, and K. White, unpublished data).

An alignment of the sequences between pA2 and exon I of 13 closely related species from all six subgroups belonging to one of the main branches in the melanogaster species group is shown in Fig. 6A (see Materials and Methods for links to phylogenetic trees in Flybase). Splicing and polyadenylation signals are well conserved in all species (Fig. 6A). In addition, six AU4-6 motifs are conserved, and each one might represent the binding site for one ELAV dimer according to the assembly in Fig. 3. Consistent with our in vitro binding data (Fig. 4), the m3r element and polypyrimidine tract show the highest conservation (Fig. 6A). Although the adenine is strictly present in the D. melanogaster AU4-6 motifs, replacement with guanosine is found in other species, and uracil is sometimes replaced by cytosine. Further conserved sequences are found around several AU4-6 motifs, particularly in the m1r, m2r, m3l, and m3r elements. Surprisingly, m1l, m2l, and m3m elements start either to diverge or are deleted, pointing out that binding platforms for individual parts of the ELAV complex show flexibility in spatial positioning.



View larger version (49K):
[in this window]
[in a new window]
  		FIG. 6. Evolutionary comparison of the ELAV binding site in ewg intron 6. (A) The DNA sequence from the AAUAAA of pA2 up to exon I was obtained from a series of Drosophila species from one of the main lines in the melanogaster species group and aligned. Sequence landmarks of corresponding transcripts are shown above the alignment above a consensus sequence deduced as majority (e.g., ≥7 from 13). Consensus strength is indicated by colored bars (red, 12 or 13 identical from 13; orange, 10 or 11 identical from 13; green, 8 or 9 identical from 13; light blue, 6 or 7 identical from 13; and dark blue, 4 or 5 identical from 13). R, purine; Y, pyrimidine; and W, A or T. Ns or spaces are shown as dots. Consensus nucleotides are in yellow in the alignment, and conserved AU4 motifs are in yellow in the consensus sequence. If the consensus nucleotide is a purine, then the minor portion is in green. Similarly, orange is used for pyrimidines and blue for A/T. Taxonomical relations are shown at the 3' end of the sequence. (B) Tree of molecular relations based on nucleotide substitutions (generated with MegAlign, DNAstar version 5.06).

Multiple AU4-6 motifs contribute to ELAV-dependent regulation of ewg in vivo. Since introducing mutations in each individual AU4-6 element (m1, m2, and m3) reduced ELAV binding in nuclear extracts when UV-cross-linking was used (55), we wanted to test if AU4-6 motifs of individual elements m1, m2, and m3 contribute to ELAV-mediated splicing regulation of ewg intron 6 in neurons and if these contributions correlate with our data on ELAV complex formation in vitro. Therefore, we introduced AU4-6 mutations in individual elements and combinations thereof into the tcgER reporter gene and established transgenic fly lines (Fig. 7A) (55). RPAs were used to examine ewg pre-mRNA processing in dissected third-instar larval brains with probes around the HA and vesicular stomatitis virus (VSV) tags to protect spliced and unspliced RNA (Fig. 7A).



View larger version (40K):
[in this window]
[in a new window]
  		FIG. 7. Inhibition of 3' end formation at pA2 depends on ELAV binding at multiple and spaced AU4-6 motifs in vivo. (A) Schematic of the last ewg intron 6 and 3' UTR with splicing and polyadenylation choices as present with HA and VSV tags in the tcgER rescue reporter construct (top) (55). The two major isoforms are those with intron 6 spliced from H to J in the presence of ELAV (solid line) or with polyadenylation in intron 6 at pA2 in the absence of ELAV. ewg probes used for RPA are shown in the middle. A magnification of the ELAV binding region is shown at the bottom, depicting polyadenylation site pA2 and part of exon I as well as AU4-6 motif and spacer mutations introduced into tcgER reporter transgenes. (B) RPA of RNA prepared from 20 third-instar larval brains from wild-type and AU4-6 motif mutants of ewg rescue reporter construct transgenes in a wild-type elav background. After purification, RNA was run on a 6% denaturing gel. Protected fragments are indicated at the right, and protected fragments from endogenous ewg transcripts are indicated by asterisks. The fragment marked with x derives from exon J of the transgene (judged by its accumulation following the usage of intronic polyadenylation sites). The loading control protects a fragment from Appl. Quantification of protected fragments from intron 6 (unspliced) to spliced transcripts is shown at the bottom of the panel. Additional lines with independent insertions from each construct gave comparable results. (C) Splicing of ewg intron 6 analyzed by quantitative RT-PCR (primers 6F and VSV-R) with RNA extracted from third-instar eye imaginal disks from wild-type and AU4-6 motif mutants of ewg reporter construct transgenes in a wild-type elav background. For total transcripts a 5' fragment of the reporter transcript was amplified with primers eeF and eeR (55). Uniformly 32P-dCTP labeled PCR products were separated on 8% polyacrylamide gels. (D) Splicing of ewg intron 6 analyzed by quantitative RT-PCR with RNA extracted from third-instar eye imaginal disks from wild-type and spacer mutants of ewg reporter construct transgenes in a wild-type elav background. For total transcripts a 5' fragment of the reporter transcript was amplified with primers eeF and eeR. Uniformly 32P-dCTP labeled PCR products were separated on 8% polyacrylamide gels. Quantification from three independent insertions is shown as means at the bottom.

In the mutant m1-3, splicing of intron 6 was about 20-fold reduced compared to that in the wild-type reporter transgene in larval brains (Fig. 7B, lanes 1 and 2). Similar results were obtained with RT-PCR analysis of RNA from developing photoreceptor neurons (Fig. 7C, lanes 1 and 2) and with larval brains (data not shown).

In contrast to the transgene carrying mutation m1-3, mutations in AU4-6 motifs of one element (m1 or m3) revealed a minimal effect on intron 6 splicing (Fig. 7B and C, lanes 3 and 6). When AU4-6 motifs of two elements (m1+2 or m2+3) are mutated, a more pronounced effect is seen (Fig. 7B and C, lanes 4 and 5). Mutations in elements m2+3 are about twice as effective as mutations in elements m1+2 in reducing intron 6 splicing (Fig. 7B and C, compare lanes 4 and 5), but the reduction in intron 6 splicing is far below the effect observed with mutations in all AU4-6 motifs (m1-3) (Fig. 7B and C, lane 2). Levels of intron 6 splicing are comparable in larval brains (Fig. 7B) and in developing photoreceptor neurons (Fig. 7C). We thus conclude that in vivo multiple AU4-6 motifs contribute to ELAV-mediated splicing of intron 6, which supports our in vitro data on ELAV acting as a macromolecular complex spread over 135 nt.

Spacing between conserved AU4-6 elements is not critical for ELAV-dependent regulation of ewg in vivo. Phylogenetic analysis suggests that spacer regions between individual AU4-6 motifs can vary in length, as some species have small deletions in nonconserved areas of the ewg ELAV complex binding site but no obvious compensations for this loss (Fig. 6A). To test if this is indeed the case, we introduced 30-nt spacers (between m1 and m2, between m2 and m3, and between m3m and m3r elements) in regions which harbor deletions and/or are less conserved (Fig. 6A and 7A). As sequence for these spacers, the reversed complementary sequence following the insertion site was chosen, since it is A rich and not U rich but has a composition frequently found in introns.

Splicing of intron 6 was analyzed in transgenic flies carrying spacer insertions in the tcgER rescue reporter gene. Only marginal differences in intron 6 splicing were observed in transgenes with spacer insertions between m1 and m2 and between m2 and m3 elements (Fig. 7D, compare lane 1 with lanes 3 and 4). A spacer insertion between elements m3m and m3r had a slight stimulatory effect on splicing of intron 6 in neurons (Fig. 7D, lane 5). Since additional RNA binding proteins could bind to the spacer elements and act in a compensatory manner together with ELAV, we did UV-cross-linking assays with spacer-containing substrate RNAs in neuronal nuclear extracts. No additional RNA binding proteins were detected (data not shown). These results suggest that spacer regions between individual elements involved in ELAV binding and complex formation can vary in length.


arrow
DISCUSSION
 
The experiments presented in this paper show that ELAV forms a dodecameric complex upon binding to ewg RNA pA2-I and that multiple and phylogenetically conserved AU4-6 motifs spread over 135 nt are important for high-affinity binding in vitro and regulation of ewg intron 6 splicing in vivo. Thus, the ability of a binding site to support multimeric ELAV complex formation may provide the answer to the puzzle of how ELAV achieves target specificity in vivo in light of its general binding preferences for short AU-rich sequences that are abundant in noncoding regions of transcripts. Below we discuss the composition and the assembly of ELAV into a multimeric complex with regard to the organization of the first characterized ELAV complex binding site in the ewg gene. We further point out implications of these findings in target gene regulation and posttranscriptional control of gene networks.

Evidence for a dodecameric ELAV complex on ewg RNA pA2-I. Based on several lines of evidence, we propose a model for a multimeric ELAV complex consisting of 12 ELAV molecules that associate with ewg RNA between pA2 and exon I in intron 6 in vitro. First, ELAV assembles with RNA into a defined and saturable RNA-protein complex when assayed by EMSA. This association occurs in an RNA substrate-specific manner, as some RNAs do not form an ELAV-RNA complex even at a concentration of 3.2 µM ({Delta}134, {Delta}5 to {Delta}7), which thus clearly distinguishes the ELAV-RNA complex from an unspecific aggregation. Second, two substrate RNAs of different size form two separable complexes, demonstrating that only one RNA is present in the final ELAV complex. Third, in size exclusion chromatography experiments under physiological salt conditions, ELAV bound to ewg RNA pA2-I results in a defined complex of about 700 kDa, and the smaller RBD60 protein yields an RNA-protein complex of appropriately reduced size of about 500 kDa, suggesting assembly of a complex in the range of 12 protein molecules. Fourth, in stoichiometry EMSAs the final ELAV complex forms at around a ratio of one RNA per 12 ELAV molecules. Fifth, titration of ELAV against RBD60 at complex-forming concentrations in EMSAs reveals 13 bands, as expected for a dodecameric complex. Sixth, reducing the length of the substrate RNA does not result in ELAV complexes of intermediate size, indicating that binding of ELAV as dodecameric complex is an intrinsic property of ELAV to associate with target RNA. Although the tools to demonstrate an in vivo assembly of a dodecameric ELAV complex with target RNA in fly neurons are currently not available, circumstantial evidence supports the presence of large ELAV-RNA complexes in vivo. In the nucleus, ELAV has been found to sublocalize to sites of higher concentration in discrete dots and webs (62), indicating that complex formation with ewg pA2-I RNA at around 350 nM in vitro could meet in vivo conditions. Further, Hu proteins have also been shown to be present in large particles in cells and neurites (5, 7, 25, 54).

Although ELAV shares the tetramer configuration characteristics with general heterogeneous nuclear ribonucleoprotein particle (hnRNP) proteins (for reviews, see references 18, 30, 39, and 61), binding to RNA induces a rearrangement into dimers. In addition, the ELAV complex forms on 43 to 135 nucleotides of RNA pA2-I, while a tetramer unit of general hnRNPs isolated from 40S particles binds to 200 to 240 nt (23, 37), and the RNA present in the 40S particle is about 500 to 800 nt in length (12, 18, 37). The length differences of the RNAs present in the ELAV complex and general hnRNP complexes likely reflect a different packaging mode. Models for hnRNP C binding to RNA have favored a loose wrapping around the tetramer (51). For an ELAV-RNA complex, however, a different model might apply, as the two RRMs of Sex-lethal (Sxl) and the first two RRMs of HuD, both closely related to ELAV, were shown to cover 11 nt in cocrystallization experiments (22, 60). A linear assembly of 12 ELAVs with RNA is therefore unlikely, particularly with RNAs as short as 43 nt, unless not all RRMs are in contact with RNA in such a complex. Moreover, phylogenetically conserved AU4-6 motifs contain only 5 to 7 nucleotides. A possible alternate model for the assembly of the ELAV complex might therefore include that ELAV surrounds the RNA upon binding, similar to the core of Sm proteins bound to snRNA (24).

Although our data point towards a dodecameric ELAV complex bound to RNA pA2-I, based on the titration experiments in Fig. 3, and to an ELAV tetramer in solution, based on gel filtration and protein cross-linking data, alternative interpretations of these data are also possible. For example, titration of ELAV with RBD60 in EMSAs could result in an overestimation of the stoichiometry, as RNA-protein complexes of identical composition but with different conformations could appear as distinguishable bands in EMSAs. Further, ELAV complex size might also be overestimated in gel filtration experiments due to an elongated shape or interaction with the column matrix, as illustrated by the following example. Initial experiments with polypyrimidine tract binding protein (PTB) indicated dimerization in solution, as deduced from gel filtration experiments that were supported by protein-protein interaction assays (41, 44). More recent experiments using physical methods, however, revealed that PTB is predominantly a monomer in solution and that the larger size of PTB in gel filtration is attributed to its elongated shape and the N-terminal 55-amino-acid auxiliary domain of PTB (2, 53). According to these recent data, earlier results from cross-linking experiments might have shown only a minor portion of PTB that is associated into a dimer, and interaction in yeast two-hybrid assays might have been due to unspecific binding to yeast RNA as PTB dimerizes on c-src RNA (2). Similar to the PTB scenario, we might overestimate the ELAV complex, and further physical experiments are needed to prove our model.

Sequence features of the ewg ELAV complex binding site. The assembly of a dodecameric ELAV complex on RNA pA2-I suggests that an array of repetitive cis elements might mediate complex formation. Results from various approaches show that a series of AU4-6 motifs present between pA2 and exon I in the last ewg intron are important for ELAV complex binding (this study and reference 55). First, using RNA substrates with mutations in AU4-6 motif element m1, m2, or m3 demonstrated that all elements spread over 135 nt contribute to ELAV binding in vitro in UV-cross-linking assays and EMSAs (55). Second, UV-cross-linking assays with segmentally labeled substrate RNAs using RNA pA2-ivs further demonstrate that the ELAV complex binding site extends over about 135 nt (55). Third, phylogenetic analysis of the ELAV binding site reveals evolutionary conservation of six AU4-6 motifs, suggesting that an ELAV dimer might bind per AU4-6 motif. Fourth, functional importance in vivo of AU4-6 motifs is further shown in ELAV-mediated splicing of the last ewg intron, using Drosophila transgenes. Although an array of AU4-6 motifs is important for ELAV complex formation, not all AU4-6 motifs contribute equally. In particular, AU4-6 motifs in the m3 element and the polypyrimidine tract have a much higher impact on high-affinity binding than the m1 and m2 elements, both in vitro and in vivo. A similar situation has been observed in the hnRNP A1 binding site in intron 3 of human immunodeficiency virus tat transcripts, and the following model has been proposed. A few hnRNP A1 molecules bind first to the high-affinity portion of the binding site and then recruit further hnRNP A1 molecules to nucleate to a higher-order complex (13, 65). A similar model might also apply to ELAV complex formation. Here, high-affinity binding of few ELAV molecules to the 3' part of the complex binding site could lead to recruitment of more ELAVs that will enhance complex formation in the presence of additional AU4-6 motifs. Alternatively, clustering of binding motifs might enhance cooperative interactions among ELAVs that then trigger formation of a stable complex upon reaching local concentrations close to the stoichiometry of the final ELAV complex at a specific target site, thereby contributing to gene-specific recognition of target RNAs.

The ewg ELAV complex binding site from the melanogaster species subgroup harbors three tandem AU4-6 motifs (in m1, m2, and m3 elements) that can be aligned (55). Collectively, the results presented here argue against tandem AU4-6 motifs in element m1, m2, or m3 either as individual tetramer binding sites or as overlapping binding sites for two tetramers. First, ELAV assembles as dimers in stoichiometry EMSAs. Second, deleting tandem AU4-6 motifs does not result in ELAV complexes of intermediate size, as no one- or two-tetramer complexes are detected as the main product in EMSAs. Rather, ELAV complex formation and its affinity for a specific substrate RNA depend on length and poly(U) content of the substrate RNA. Third, tandem AU4-6 motifs in the m1 and m2 elements are not sufficient for complex formation in EMSAs. Fourth, only six AU4-6 motifs (m1r, m2r, m3l, m3r, and two in the polypyrimidine tract) are evolutionarily conserved. The additional AU4-6 motifs present in the melanogaster species subgroup might therefore represent redundancy. This is also indicated by the minimal difference in affinity between RNA pA2-I, {Delta}1, and {Delta}12 in EMSAs, as the remaining AU4-6 motifs still suffice for almost optimal positioning of the ELAV complex in this assay.

Implications for regulation of target genes. Besides increasing binding specificity, multimerization of RNA binding proteins might also provide a mechanism to locate distant RNA-processing signals by looping out intronic sequence to bring splice sites into proximity. The organization of the ELAV complex binding site illustrates flexibility in positioning of AU4-6 motifs relative to each other, as they are not strictly conserved between different Drosophila species and introducing spacer sequences only marginally affects processing of the last ewg intron 6. Forming a complex with distant recognition sequences also offers an appealing explanation for ELAV-mediated regulation of pre-mRNA processing in the more complex situation of nrg, where a distal terminal exon is chosen in the presence of ELAV (27). UV-cross-linking studies of the whole nrg-regulated intron revealed four areas of binding in the vicinity of splice sites and poly(A) signals (34). An ELAV complex bound to these extensively spaced binding sites would exclude nonneuronal RNA-processing signals from recognition. A similar model has been proposed for autoregulation of hnRNP A1 alternative splicing. Here, hnRNP A1 binds sequences flanking both sides of the regulated exon, leading to skipping of this exon (8).

Functional importance of multimerization has also been indicated for hnRNP A1-regulated alternative splicing of intron 3 from human immunodeficiency virus tat transcripts. Here, multimerization of hnRNP A1 in the context of RNA secondary structure on intronic and exonic splicing silencer sequences competes with other factors for exon recognition (13, 56, 65). In the case of PTB-regulated splicing of a neural exon in c-src (11), a neuronal form of PTB, nPTB, is postulated to interfere with multimerization of PTB and release the blocked exon for inclusion in neurons (38). nPTB has also been found to interact with Nova-1 and to inhibit Nova-1-stimulated GlyR{alpha}2 E3A alternative splicing (46). In Drosophila, overexpression of the ELAV family member FNE can regulate expression of ELAV, similar to autoregulation by ELAV at endogenous levels (50). As ELAV family members interact in yeast two-hybrid assays, they likely can also form heteromultimeric complexes in vivo (21, 25; L. Toba and K. White, unpublished data), and multimeric binding of HuB at the c-myc 3' UTR has been indicated (20, 32). In addition, many other RNA binding proteins have also been shown to engage in homo- and heterophilic interactions (see, e.g., references 21, 47, and 49).

In conclusion, multimerization of RNA binding proteins into macromolecular complexes likely is an important feature of this abundant class of proteins to localize pre-mRNA processing signals in a complex cellular environment in constitutive splicing and affect use of alternative splice and polyadenylation sites. In addition, multimerization of RNA binding proteins might also provide the combinatorial setup to posttranscriptionally coordinate the expression of functionally related genes in higher eukaryotes, as postulated by Keene and Tennenbaum (26).


arrow
ACKNOWLEDGMENTS
 
We thank E. Kubli for Drosophila suzukii; I. Haussmann for help with cloning and fly transformations; Maydung Nguyen for help with gel filtration; and I. Haussmann, M. Jurica, M. Moore, M. Rosbash, and many colleagues at Brandeis for advice, discussions, and comments on the manuscript.

This work was supported by National Institute of Health grant NS 44232.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Biology Department, MS 008, Brandeis University, Waltham, MA 02454. Phone: (781) 736-3177. Fax: (781) 736-3107. E-mail for Matthias Soller: msoller{at}brandeis.edu. E-mail for Kalpana White: white{at}brandeis.edu. Back


arrow
REFERENCES
 
    1
  1. Abe, R., E. Sakashita, K. Yamamoto, and H. Sakamoto. 1996. Two different RNA binding activities for the AU-rich element and the poly(A) sequence of the mouse neuronal protein mHuC. Nucleic Acids Res. 24:4895-4901.[Abstract/Free Full Text]
    2. 2
  2. Amir-Ahmady, B., P. L. Boutz, V. Markovtsov, M. L. Phillips, and D. L. Black. 2005. Exon repression by polypyrimidine tract binding protein. RNA 11:699-716.[Abstract/Free Full Text]
    3. 3
  3. Anantharaman, V., E. V. Koonin, and L. Aravind. 2002. Comparative genomics and evolution of proteins involved in RNA metabolism. Nucleic Acids Res. 30:1427-1464.[Abstract/Free Full Text]
    4. 4
  4. Antic, D., and J. D. Keene. 1997. Embryonic lethal abnormal visual RNA-binding proteins involved in growth, differentiation, and posttranscriptional gene expression. Am. J. Hum. Genet. 61:273-278.[Medline]
    5. 5
  5. Antic, D., and J. D. Keene. 1998. Messenger ribonucleoprotein complexes containing human ELAV proteins: interactions with cytoskeleton and translational apparatus. J. Cell Sci. 111:183-197.[Abstract]
    6. 6
  6. Antic, D., N. Lu, and J. D. Keene. 1999. ELAV tumor antigen, hel-N1, increases translation of neurofilament M mRNA and induces formation of neurites in human teratocarcinoma cells. Genes Dev. 13:449-461.[Abstract/Free Full Text]
    7. 7
  7. Aronov, S., G. Aranda, L. Behar, and I. Ginzburg. 2002. Visualization of translated tau protein in the axons of neuronal P19 cells and characterization of tau RNP granules. J. Cell Sci. 115:3817-3827.[Abstract/Free Full Text]
    8. 8
  8. Blanchette, M., and B. Chabot. 1999. Modulation of exon skipping by high-affinity hnRNP A1-binding sites and by intron elements that repress splice site utilization. EMBO J. 18:1939-1952.[CrossRef][Medline]
    9. 9
  9. Brennan, C. M., and J. A. Steitz. 2001. HuR and mRNA stability. Cell Mol. Life Sci. 58:266-277.[CrossRef][Medline]
    10. 10
  10. Burd, C. G., and G. Dreyfuss. 1994. Conserved structures and diversity of functions of RNA binding proteins. Science 265:615-621.[Abstract/Free Full Text]
    11. 11
  11. Chou, M. Y., J. G. Underwood, J. Nikolic, M. H. Luu, and D. L. Black. 2000. Multisite RNA binding and release of polypyrimidine tract binding protein during the regulation of c-src neural-specific splicing. Mol. Cell 5:949-957.[CrossRef][Medline]
    12. 12
  12. Conway, G., J. Wooley, T. Bibring, and W. M. LeStourgeon. 1988. Ribonucleoproteins package 700 nucleotides of pre-mRNA into a repeating array of regular particles. Mol. Cell. Biol. 8:2884-2895.[Abstract/Free Full Text]
    13. 13
  13. Damgaard, C. K., T. O. Tange, and J. Kjems. 2002. hnRNP A1 controls HIV-1 mRNA splicing through cooperative binding to intron and exon splicing silencers in the context of a conserved secondary structure. RNA 8:1401-1415.[Abstract]
    14. 14
  14. Daneholt, B. 2001. Assembly and transport of a premessenger RNP particle. Proc. Natl. Acad. Sci. USA 98:7012-7017.[Abstract/Free Full Text]
    15. 15
  15. Darnell, J. C., K. B. Jensen, P. Jin, V. Brown, S. T. Warren, and R. B. Darnell. 2001. Fragile X mental retardation protein targets G quartet mRNAs important for neuronal function. Cell 107:489-499.[CrossRef][Medline]
    16. 16
  16. Dredge, B. K., and R. B. Darnell. 2003. Nova regulates GABA(A) receptor gamma2 alternative splicing via a distal downstream UCAU-rich intronic splicing enhancer. Mol. Cell. Biol. 23:4687-4700.[Abstract/Free Full Text]
    17. 17
  17. Dreyfuss, G., V. N. Kim, and N. Kataoka. 2002. Messenger-RNA-binding proteins and the messages they carry. Nat. Rev. Mol. Cell. Biol. 3:195-205.[CrossRef][Medline]
    18. 18
  18. Dreyfuss, G., M. J. Matunis, S. Pinol-Roma, and C. G. Burd. 1993. hnRNP proteins and the biogenesis of mRNA. Annu. Rev. Biochem. 62:289-321.[CrossRef][Medline]
    19. 19
  19. Ford, L. P., J. Watson, J. D. Keene, and J. Wilusz. 1999. ELAV proteins stabilize deadenylated intermediates in a novel in vitro mRNA deadenylation/degradation system. Genes Dev. 13:188-201.[Abstract/Free Full Text]
    20. 20
  20. Gao, F. B., C. C. Carson, T. Levine, and J. D. Keene. 1994. Selection of a subset of mRNAs from combinatorial 3' untranslated region libraries using neuronal RNA-binding protein Hel-N1. Proc. Natl. Acad. Sci. USA 91:11207-11211.[Abstract/Free Full Text]
    21. 21
  21. Giot, L., J. S. Bader, C. Brouwer, et al. 2003. A protein interaction map of Drosophila melanogaster. Science 302:1727-1736.[Abstract/Free Full Text]
    22. 22
  22. Handa, N., O. Nureki, K. Kurimoto, I. Kim, H. Sakamoto, Y. Shimura, Y. Muto, and S. Yokoyama. 1999. Structural basis for recognition of the tra mRNA precursor by the Sex-lethal protein. Nature 398:579-585.[CrossRef][Medline]
    23. 23
  23. Huang, M., J. E. Rech, S. J. Northington, P. F. Flicker, A. Mayeda, A. R. Krainer, and W. M. LeStourgeon. 1994. The C-protein tetramer binds 230 to 240 nucleotides of pre-mRNA and nucleates the assembly of 40S heterogeneous nuclear ribonucleoprotein particles. Mol. Cell. Biol. 14:518-533.[Abstract/Free Full Text]
    24. 24
  24. Kambach, C., S. Walke, R. Young, J. M. Avis, E. Delafortelle, V. A. Raker, R. Luhrmann, J. Li, and K. Nagai. 1999. Crystal structure of two Sm protein complexes and their implications for the assembly of the splicosomal snRNPs. Cell 96:375-387.[CrossRef][Medline]
    25. 25
  25. Kasashima, K., E. Sakashita, K. Saito, and H. Sakamoto. 2002. Complex formation of the neuron-specific ELAV-like Hu RNA-binding proteins. Nucleic Acids Res. 30:4519-4526.[Abstract/Free Full Text]
    26. 26
  26. Keene, J. D., and S. A. Tenenbaum. 2002. Eukaryotic mRNPs may represent posttranscriptional operons. Mol. Cell 9:1161-1167.[CrossRef][Medline]
    27. 27
  27. Koushika, S. P., M. J. Lisbin, and K. White. 1996. ELAV, a Drosophila neuron-specific protein, mediates the generation of an alternatively spliced neural protein isoform. Curr. Biol. 6:1634-1641.[CrossRef][Medline]
    28. 28
  28. Koushika, S. P., M. Soller, S. M. DeSimone, D. M. Daub, and K. White. 1999. Differential and inefficient splicing of a broadly expressed Drosophila erect wing transcript results in tissue-specific enrichment of the vital EWG protein isoform. Mol. Cell. Biol. 19:3998-4007.[Abstract/Free Full Text]
    29. 29
  29. Koushika, S. P., M. Soller, and K. White. 2000. The neuron-enriched splicing pattern of Drosophila erect wing is dependent on the presence of ELAV protein. Mol. Cell. Biol. 20:1836-1845.[Abstract/Free Full Text]
    30. 30
  30. Krecic, A. M., and M. S. Swanson. 1999. hnRNP complexes: composition, structure, and function. Curr. Opin. Cell Biol. 11:363-371.[CrossRef][Medline]
    31. 31
  31. Kullmann, M., U. Gopfert, B. Siewe, and L. Hengst. 2002. ELAV/Hu proteins inhibit p27 translation via an IRES element in the p27 5'UTR. Genes Dev. 16:3087-3099.[Abstract/Free Full Text]
    32. 32
  32. Levine, T. D., F. Gao, P. H. King, L. G. Andrews, and J. D. Keene. 1993. Hel-N1: an autoimmune RNA-binding protein with specificity for 3' uridylate-rich untranslated regions of growth factor mRNAs. Mol. Cell. Biol. 13:3494-3504.[Abstract/Free Full Text]
    33. 33
  33. Levy, N. S., S. Chung, H. Furneaux, and A. P. Levy. 1998. Hypoxic stabilization of vascular endothelial growth factor mRNA by the RNA-binding protein HuR. J. Biol. Chem. 273:6417-6423.[Abstract/Free Full Text]
    34. 34
  34. Lisbin, M. J., J. Qiu, and K. White. 2001. The neuron-specific RNA-binding protein ELAV regulates neuroglian alternative splicing in neurons and binds directly to its pre-mRNA. Genes Dev. 15:2546-2561.[Abstract/Free Full Text]
    35. 35
  35. Lopez de Silanes, I., M. Zhan, A. Lal, X. Yang, and M. Gorospe. 2004. Identification of a target RNA motif for RNA-binding protein HuR. Proc. Natl. Acad. Sci. USA 101:2987-2992.[Abstract/Free Full Text]
    36. 36
  36. Loria, P. M., A. Duke, J. B. Rand, and O. Hobert. 2003. Two neuronal, nuclear-localized RNA binding proteins involved in synaptic transmission. Curr. Biol. 13:1317-1323.[CrossRef][Medline]
    37. 37
  37. Lothstein, L., H. P. Arenstorf, S. Y. Chung, B. W. Walker, J. C. Wooley, and W. M. LeStourgeon. 1985. General organization of protein in HeLa 40S nuclear ribonucleoprotein particles. J. Cell Biol. 100:1570-1581.[Abstract/Free Full Text]
    38. 38
  38. Markovtsov, V., J. M. Nikolic, J. A. Goldman, C. W. Turck, M. Y. Chou, and D. L. Black. 2000. Cooperative assembly of an hnRNP complex induced by a tissue-specific homolog of polypyrimidine tract binding protein. Mol. Cell. Biol. 20:7463-7479.[Abstract/Free Full Text]
    39. 39
  39. McAfee, J., M. Huang, S. Soltaninassab, J. Rech, S. Iyengar, and W. LeStourgeon. 1997. The packaging of pre-mRNA, p. 68-102. In A. Krainer (ed.), Eukaryatioc mRNA Processing. IRL Press at Oxford University Press, Oxford, United Kingdom.
    40. 40
  40. Melin, L., D. Soldati, R. Mital, A. Streit, and D. Schumperli. 1992. Biochemical demonstration of complex formation of histone pre-mRNA with U7 small nuclear ribonucleoprotein and hairpin binding factors. EMBO J. 11:691-697.[Medline]
    41. 41
  41. Oh, Y. L., B. Hahm, Y. K. Kim, H. K. Lee, J. W. Lee, O. Song, K. Tsukiyama-Kohara, M. Kohara, A. Nomoto, and S. K. Jang. 1998. Determination of functional domains in polypyrimidine-tract-binding protein. Biochem. J. 331:169-175.
    42. 42
  42. Okano, H. J., and R. B. Darnell. 1997. A hierarchy of Hu RNA binding proteins in developing and adult neurons. J. Neurosci. 17:3024-3037.[Abstract/Free Full Text]
    43. 43
  43. Pascale, A., P. A. Gusev, M. Amadio, T. Dottorini, S. Govoni, D. L. Alkon, and A. Quattrone. 2004. Increase of the RNA-binding protein HuD and posttranscriptional up-regulation of the GAP-43 gene during spatial memory. Proc. Natl. Acad. Sci. USA 101:1217-1222.[Abstract/Free Full Text]
    44. 44
  44. Perez, I., J. G. McAfee, and J. G. Patton. 1997. Multiple RRMs contribute to RNA binding specificity and affinity for polypyrimidine tract binding protein. Biochemistry 36:11881-11890.[CrossRef][Medline]
    45. 45
  45. Perrone-Bizzozero, N., and F. Bolognani. 2002. Role of HuD and other RNA-binding proteins in neural development and plasticity. J. Neurosci. Res. 68:121-126.[CrossRef][Medline]
    46. 46
  46. Polydorides, A. D., H. J. Okano, Y. Y. Yang, G. Stefani, and R. B. Darnell. 2000. A brain-enriched polypyrimidine tract-binding protein antagonizes the ability of Nova to regulate neuron-specific alternative splicing. Proc. Natl. Acad. Sci. USA 97:6350-6355.[Abstract/Free Full Text]
    47. 47
  47. Reimann, I., A. Huth, H. Thiele, and B. J. Thiele. 2002. Suppression of 15-lipoxygenase synthesis by hnRNP E1 is dependent on repetitive nature of LOX mRNA 3'-UTR control element DICE. J. Mol. Biol. 315:965-974.[CrossRef][Medline]
    48. 48
  48. Robinow, S., A. R. Campos, K. M. Yao, and K. White. 1988. The elav gene product of Drosophila, required in neurons, has three RNP consensus motifs. Science 242:1570-1572. (Erratum, 243:12, 1989.)
    49. 49
  49. Ryder, S. P., L. A. Frater, D. L. Abramovitz, E. B. Goodwin, and J. R. Williamson. 2004. RNA target specificity of the STAR/GSG domain post-transcriptional regulatory protein GLD-1. Nat. Struct. Mol. Biol. 11:20-28.[CrossRef][Medline]
    50. 50
  50. Samson, M. L., and F. Chalvet. 2003. Found in neurons, a third member of the Drosophila elav gene family, encodes a neuronal protein and interacts with elav. Mech. Dev. 120:373-383.[CrossRef][Medline]
    51. 51
  51. Shahied, L., E. H. Braswell, W. M. LeStourgeon, and A. M. Krezel. 2001. An antiparallel four-helix bundle orients the high-affinity RNA binding sites in hnRNP C: a mechanism for RNA chaperonin activity. J. Mol. Biol. 305:817-828.[CrossRef][Medline]
    52. 52
  52. Shamoo, Y., N. Abdul-Manan, and K. R. Williams. 1995. Multiple RNA binding domains (RBDs) just don't add up. Nucleic Acids Res. 23:725-728.[Abstract/Free Full Text]
    53. 53
  53. Simpson, P. J., T. P. Monie, A. Szendroi, N. Davydova, J. K. Tyzack, M. R. Conte, C. M. Read, P. D. Cary, D. I. Svergun, P. V. Konarev, S. Curry, and S. Matthews. 2004. Structure and RNA interactions of the N-terminal RRM domains of PTB. Structure (Cambridge) 12:1631-1643.
    54. 54
  54. Smith, C. L., R. Afroz, G. J. Bassell, H. M. Furneaux, N. I. Perrone-Bizzozero, and R. W. Burry. 2004. GAP-43 mRNA in growth cones is associated with HuD and ribosomes. J. Neurobiol. 61:222-235.[CrossRef][Medline]
    55. 55
  55. Soller, M., and K. White. 2003. ELAV inhibits 3'-end processing to promote neural splicing of ewg pre-mRNA. Genes Dev. 17:2526-2538.[Abstract/Free Full Text]
    56. 56
  56. Tange, T. O., C. K. Damgaard, S. Guth, J. Valcarcel, and J. Kjems. 2001. The hnRNP A1 protein regulates HIV-1 tat splicing via a novel intron silencer element. EMBO J. 20:5748-5758.[CrossRef][Medline]
    57. 57
  57. Tenenbaum, S. A., C. C. Carson, P. J. Lager, and J. D. Keene. 2000. Identifying mRNA subsets in messenger ribonucleoprotein complexes by using cDNA arrays. Proc. Natl. Acad. Sci. USA 97:14085-14090.[Abstract/Free Full Text]
    58. 58
  58. Ule, J., K. B. Jensen, M. Ruggiu, A. Mele, A. Ule, and R. B. Darnell. 2003. CLIP identifies Nova-regulated RNA networks in the brain. Science 302:1212-1215.[Abstract/Free Full Text]
    59. 59
  59. Varani, G., and K. Nagai. 1998. RNA recognition by RNP proteins during RNA processing. Annu. Rev. Biophys. Biomol. Struct. 27:407-445.[CrossRef][Medline]
    60. 60
  60. Wang, X., and T. M. Tanaka Hall. 2001. Structural basis for recognition of AU-rich element RNA by the HuD protein. Nat. Struct. Biol. 8:141-145.[CrossRef][Medline]
    61. 61
  61. Weighardt, F., G. Biamonti, and S. Riva. 1996. The roles of heterogeneous nuclear ribonucleoproteins (hnRNP) in RNA metabolism. Bioessays 18:747-756.[CrossRef][Medline]
    62. 62
  62. Yannoni, Y. M., and K. White. 1997. Association of the neuron-specific RNA binding domain-containing protein ELAV with the coiled body in Drosophila neurons. Chromosoma 105:332-341.[CrossRef][Medline]
    63. 63
  63. Yao, K.-M., M.-L. Samson, R. Reeves, and K. White. 1993. Gene elav of Drosophila melanogaster: a prototype for neuronal-specific RNA binding protein gene family that is conserved in flies and humans. J. Neurobiol. 24:723-739.[CrossRef][Medline]
    64. 64
  64. Yao, K. M., and K. White. 1991. Organizational analysis of elav gene and functional analysis of ELAV protein of Drosophila melanogaster and Drosophila virilis. Mol. Cell. Biol. 11:2994-3000.[Abstract/Free Full Text]
    65. 65
  65. Zhu, J., A. Mayeda, and A. R. Krainer. 2001. Exon identity established through differential antagonism between exonic splicing silencer-bound hnRNP A1 and enhancer-bound SR proteins. Mol. Cell 8:1351-1361.[CrossRef][Medline]

Molecular and Cellular Biology, September 2005, p. 7580-7591, Vol. 25, No. 17
0270-7306/05/$08.00+0     doi:10.1128/MCB.25.17.7580-7591.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Toba, G., White, K. (2008). The third RNA recognition motif of Drosophila ELAV protein has a role in multimerization. Nucleic Acids Res 36: 1390-1399 [Abstract] [Full Text]  
  • Lee, H. K., Kwak, H. Y., Hur, J., Kim, I. A., Yang, J. S., Park, M. W., Yu, J., Jeong, S. (2007). {beta}-Catenin Regulates Multiple Steps of RNA Metabolism as Revealed by the RNA Aptamer in Colon Cancer Cells. Cancer Res. 67: 9315-9321 [Abstract] [Full Text]  
  • David, P. S., Tanveer, R., Port, J. D. (2007). FRET-detectable interactions between the ARE binding proteins, HuR and p37AUF1. RNA 13: 1453-1468 [Abstract] [Full Text]  
  • Zhu, H., Hasman, R. A., Barron, V. A., Luo, G., Lou, H. (2006). A Nuclear Function of Hu Proteins as Neuron-specific Alternative RNA Processing Regulators. Mol. Biol. Cell 17: 5105-5114 [Abstract] [Full Text]  
  • Burry, R. W., Smith, C. L. (2006). HuD Distribution Changes in Response to Heat Shock but Not Neurotrophic Stimulation. J. Histochem. Cytochem. 54: 1129-1138 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Soller, M.
Right arrow Articles by White, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Soller, M.
Right arrow Articles by White, K.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»