Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow A corrigendum has been published
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (29)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Nasim, M. T.
Right arrow Articles by Eperon, I. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nasim, M. T.
Right arrow Articles by Eperon, I. C.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Human Molecular Genetics, 2003, Vol. 12, No. 11 1337-1348
DOI: 10.1093/hmg/ddg136
© 2003 Oxford University Press

HnRNP G and Tra2ß: opposite effects on splicing matched by antagonism in RNA binding

M. Talat Nasim, Tatyana K. Chernova, Hasnin M. Chowdhury, Bai-Gong Yue and Ian C. Eperon*

Department of Biochemistry, University of Leicester, Leicester LE1 7RH, UK

Received November 6, 2002; Revised March 6, 2003; Accepted March 21, 2003


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The hnRNP G family comprises three closely related proteins, hnRNP G, RBMY and hnRNP G-T. We showed previously that they interact with splicing activator proteins, particularly hTra2ß, and suggested that they were involved in regulating Tra2-dependent splicing. We show here that hnRNP G and hTra2ß have opposite effects upon the incorporation of several exons, both being able to act as either an activator or a repressor. HnRNP G acts via a specific sequence to repress the skeletal muscle-specific exon (SK) of human slow skeletal alpha-tropomyosin, TPM3, and stimulates inclusion of the alternative non-muscle exon. The binding of hnRNP G to the exon is antagonized by hTra2ß. The two proteins also have opposite effects upon a dystrophin pseudo-exon. This exon is incorporated in a patient to a higher level in heart muscle than skeletal muscle, causing X-linked dilated cardiomyopathy. It is included to a higher level after transfection of a mini-gene into rodent cardiac myoblasts than into skeletal muscle myoblasts. Co-transfection with hnRNP G represses incorporation in cardiac myoblasts, whereas hTra2ß increases it in skeletal myoblasts. Both the cell specificity and the protein responses depend upon exon sequences. Since the ratio of hnRNP G to Tra2ß mRNA in humans is higher in skeletal muscle than in heart muscle, we propose that the hnRNP G/Tra2ß ratio contributes to the cellular splicing preferences and that the higher proportion of hnRNP G in skeletal muscle plays a role in preventing the incorporation of the pseudo-exon and thus in preventing skeletal muscle dystrophy.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Heterogeneous nuclear ribonucleoprotein (hnRNP) G was the Cinderella of the hnRNP family. Like other hnRNP proteins, it is associated with nascent RNA (1–3); indeed, despite its relatively low abundance, it was detected on most loops of lampbrush chromosomes, suggesting that it is a general component of transcription complexes (4). It contains an RRM-type RNA-binding domain (RRM), implying a direct association with RNA. However, it was until recently one of the few hnRNP proteins to which no function had been ascribed (5).

A possible function for hnRNP G in splicing has emerged from work on RBMY (originally YRRM). RBMY is a Y chromosome gene that was identified as a candidate azoospermia factor (6). The predicted sequence of the RBMY product showed striking similarity to the RNA-binding domain of hnRNP G, and later amendment of the hnRNP G sequence revealed 60% identity overall (7). Indeed, hnRNP G is encoded by RBMX, which is the X-chromosome homologue of RBMY (8,9). Yeast two-hybrid screens led to the findings that hnRNP G and RBMY interact with each other and with a third member of the family, hnRNP G-T, and that they also interact both with splicing activators (Tra2ß and SRp30c) and with two members of the STAR family of potential RNA-binding and signalling proteins, Sam68 and T-STAR (10,11). The significance of these observations for splicing was demonstrated by functional assays, in which splicing factors were removed or sequestered by their interaction with RBMY; specifically, SR proteins were depleted from extracts by immobilized murine RBMY (12), and addition of the protein interaction domain of RBMY to in vitro splicing reactions inhibited the splicing of a substrate containing an enhancer that bound Tra2ß but not that of a control substrate (11).

Mammalian Tra2ß is very similar to Drosophila Transformer-2, which has an important role as a splicing activator in sex determination and sex-specific neural splicing (13–15), and as a repressor of its own splicing in the male germline (16). Human Tra2ß was shown to activate splicing of a generic substrate with tandem optimal binding sequences in HeLa nuclear extract, albeit weakly (17). It was therefore expected that the interaction of hnRNP G and RBMY with Tra2ß would play an important role in the regulation of splicing in vivo. However, it has proven to be difficult to find genes regulated by Tra2ß. To the best of our knowledge, apart from the reported failures, Tra2ß has been reported to activate splicing in only two human genes (exon 7 of SMN1 and 2) (18), and to repress splicing weakly of exons in clathrin light chain B and tau (19,20). In the case of SMN, it has been suggested that Tra2ß recruits SRp30c, which itself binds negligibly to the RNA (21), and hnRNP G, which was found to have no intrinsic RNA-binding specificity (22), with the result in both cases that splicing is stimulated.

We describe here a different set of functional interactions between hnRNP G and Tra2ß. We have shown recently that Tra2ß binds to the exonic splicing enhancer in the alternative version of tropomyosin (TPM3) exon 5 that is incorporated in skeletal muscle (exon SK), but that overexpression does not affect the level of splicing; in contrast, it represses the incorporation of the mutually exclusive non-muscle exon, NM (Ayres et al., submitted). We show here that hnRNP G has independent or opposite effects upon exon inclusion in these and other exons, demonstrate that it has distinct RNA-binding preferences, and report that the antagonistic effects of Tra2ß and hnRNP G upon an example of cardiac-enhanced splicing of a pseudo-exon in a mutant dystrophin gene are consistent with a role in tissue-specific splicing and the development of disease.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Effects on the mutually exclusive exons of TPM3
The human {alpha}s-tropomyosin gene, TPM3, contains two mutually exclusive versions of exon 5. The upstream exon, NM, is expressed in non-muscle cells, and the downstream exon, SK, is expressed in skeletal muscle cells. We have demonstrated elsewhere that these exons are regulated independently, and that hTra2ß represses exon NM and forms part of the exon enhancer complex on exon SK (Ayres et al., manuscript submitted). To compare the action of hnRNP G with that of hTra2ß, minigenes containing either exon NM (HC4) or exon SK (HC5; Fig. 1A) were expressed in rodent C2C12 myoblasts or myotubes in the presence of a co-transfected plasmid expressing either Tra2ß or hnRNP G. RT–PCR showed that exon NM was incorporated to a significant extent in myoblasts, and that co-transfection with hTra2ß reduced incorporation. In contrast, co-transfection with hnRNP G stimulated incorporation (Fig. 1B). The negligible level of incorporation of exon SK was not affected significantly by co-transfection. Reducing the lengths of the flanking introns did not block the opposite actions of the two proteins on exon NM, although it did virtually eliminate incorporation of exon SK.



View larger version (49K):
[in this window]
[in a new window]
  		Figure 1. Effects of exogenous hnRNP G compared with hTra2ß on two mutually exclusive exons of {alpha}s-tropomyosin. (A) Diagram of the natural TPM3 (human {alpha}s-tropomyosin) gene around exons 5NM (non-muscle isoform) and 5SK (skeletal muscle isoform). Plasmids pHC4–pHC12 were constructed by cloning either exon NM or SK with various lengths of flanking intron sequences (lengths shown in nt) into pcDys. The latter is an exon-trapping vector based on pcDNA3.1, into which exons 11 and 12 of dystrophin were cloned together with 116 and 262 nt, respectively, of intron 11, leaving a set of cloning sites in the intron. (B) Analysis by RT–PCR and gel electrophoresis of mRNA that had skipped (lower bands, open arrows) or included (filled arrows) exon 5 after transfection of pHC4-pHC12 into C2C12 (skeletal muscle) myoblasts. Plasmids expressing tagged hnRNP G and hTra2ß were co-transfected as shown. (C) As (B), but C1C12 cells were allowed to differentiate into myotubes before analysis of the mRNA. The effects of differentiation and Tra2ß on HC4 and HC5 are described elsewhere (Ayres et al., submitted), but the results of different experiments are shown here because they were done in parallel with the tests of hnRNP G.

 
A very similar pattern was seen in myotubes (Fig. 1C). Even though the use of exon NM was much reduced, co-transfection with hnRNP G increased its use. The incorporation of exon SK was higher after differentiation, but hnRNP G acted negatively. Interestingly, the reduction in the length of the introns flanking exon SK in mutants HC11 and HC12 blocked the induction of exon SK in myotubes as well as eliminating the leakage in myoblasts.

We conclude that exon NM usage is stimulated in myoblasts and myotubes alike by hnRNP G, and repressed by hTra2ß. These effects are unaffected by substantial changes in the length of the flanking introns. In contrast, neither hnRNP G nor hTra2ß affect exon SK usage in myoblasts, but hnRNP G represses it in myotubes.

Opposite effects on TPM3 exon SK
The constructs used for the experiment in Figure 1 are subject to tissue-specific regulation, and it is possible that the contribution of a single protein to the splicing of the SK exon would be overridden by the contributions of other regulatory elements and proteins. With a splicing reporter construct containing only exon SK and a small part of the upstream intron (Fig. 2A), Tra2ß stimulated splicing of exon SK (23). To test whether hnRNP G acts antagonistically, the reporter construct was co-transfected with the plasmid expressing hnRNP G. The results showed that this was the case in all three cell lines tested (Fig. 2B). Thus, in this context the two proteins have opposite effects on exon SK, just as they did on exon NM in Figure 1, but the effects of each protein are the converse of those on exon NM.



View larger version (31K):
[in this window]
[in a new window]
  		Figure 2. Effects of hnRNP G on the splicing efficiency of TPM3 exon SK. (A) Diagram of the double reporter construct, which contains the SK exon and the preceding 30 nt of intron. Unspliced but exported mRNA is translated to produce the polypeptide with reporter activity 1, whereas spliced RNA is translated to produce a fusion polypeptide with both activities. XXX indicates translation termination signals in all frames (23). (B) The effects of co-transfection with hnRNP G and hTra2ß in three cell lines. The ratios of activities of reporter 2 to reporter 1, which report the extent of splicing, are shown relative to the ratios obtained in the absence of co-transfection, which are normalized at 100 (Cont). Averages and standard deviations of the ratios were calculated from four to 12 experiments. The data for HEK293 cells was shown previously (23) and is included for comparison. (C) Effect of mutation of SK exon nt 2–8 upon the ability of hnRNP G to reduce the efficiency of splicing. The portion of intron from TPM3 upstream of the SK exon was absent from both the wt and (2–8) mutant. The values were normalized to the wt exon in the absence of hnRNP G.

 
Several mutations in exon SK in the reporter construct were tested to determine whether specific sequences confer sensitivity to hnRNP G. Omitting the tropomyosin portion of the intron preceding exon SK reduced the effectiveness of hnRNP G (data not shown), and additional changes to the sequence at the 5' end of the exon (nt 2–8) abolished it (Fig. 2C). Changes in the downstream purine-rich sequences of the exon did not have this effect (data not shown). We conclude that at least one sequence element in the exon contributes to the effects of hnRNP G, either directly or indirectly.

Binding of hnRNP G to exon SK
To isolate hnRNP G for binding studies, lysates from cells transiently expressing hnRNP G tagged with an epitope from T7 gene 10 were incubated with antibody-coated agarose beads; the beads were washed extensively, and the retained protein was used for RNA-binding studies. The protein was eluted from the antibody for cross-linking studies. The level of purification achieved by this procedure was tested with [35S]methionine-labelled cell cultures. The results in Figure 3A show that only one major band was detectable either in the material eluted from the beads (lane 6) or in the elution-resistant material (lane 10). The band was absent in lysates from mock-transfected cells (lanes 5 and 9) and it is of an appropriate size, from which we infer that it is hnRNP G. Co-purifying proteins were present only at much lower levels. To illustrate this, a plasmid expressing hTra2ß was co-expressed. Tra2ß is known to associate with hnRNP G in vivo (11). The protein was recovered only at relatively low levels (Fig. 3A, lane 11). Contamination by proteins that do not interact appears to be negligible. This was tested by the co-expression of GFP; even though it was expressed abundantly (lane 4), none was retained by the beads (lanes 8 and 12), indicating a high level of selectivity.



View larger version (76K):
[in this window]
[in a new window]
  		Figure 3. Preferential binding of hnRNP G to SK exon RNA. T7-tagged hnRNP G was expressed in HEK 293 cells, and recovered on antibody-coupled beads. (A) Analysis by SDS–PAGE of the affinity-selected protein from cell cultures labelled with [35S]L-methionine. Flow-through, material that was not bound by the antibody-coupled beads (lanes 1–4); eluate, bound protein that was eluted by 0.1 M citric acid, pH 2.2 (lanes 5–8); retained, protein remaining on the beads after elution (lanes 9–12). G, cells transfected with plasmid expressing T7-tagged hnRNP G; Tra2, cells transfected also with plasmid expressing GFP-Tra2ß; GFP, cells transfected with plasmid expressing GFP; mock, untransfected cells. 14C-labelled marker proteins and their molecular masses in kDa are shown, and arrows indicate the tagged proteins expressed after transfection. (B) Two experiments showing binding of SK exon RNA to bound tagged hnRNP G. Beads displaying hnRNP G were incubated with ~0.05 pmol 32P-labelled SK exon RNA in the presence of the competitor sequences shown. After binding and washing, the RNA was released and analysed by gel electrophoresis. The unlabelled competitor sequences comprised SK exon RNA, the corresponding sequence from the non-muscle exon, and tRNA. In all cases, 1, 2, 3 and 5 µg of competitor were added. C, RNA bound to beads not exposed to cell extract; M, RNA bound to beads that had been incubated with lysate from untransfected cells.

 
To measure RNA binding, beads containing retained protein were incubated with 32P-labelled exon SK RNA in the presence of various concentrations of unlabelled competitor RNA. The labelled RNA retained by the beads was detected by PAGE. There was considerable variation in the level of binding and competition from experiment to experiment, reflecting the method of recovering the protein. However, exon NM invariably competed less well than exon SK. This is illustrated by the two experiments in Figure 3B. When the labelled RNA comprised exon SK plus the preceding 30 nt from the intron, unlabelled RNA of the same sequence competed effectively (Fig. 3B). However, exon NM with 30 nt of preceding intron competed less well, even though hnRNP G stimulates the splicing of this exon. Transfer RNA produced results similar to exon NM. As a control, radiolabelled RNA was incubated either with antibody-coated beads (lane C) or with beads that had been incubated with a lysate from untransfected cells (lane M).

Although hnRNP G and Tra2ß have opposite effects on the inclusion of both of the mutually exclusive exons, it does not follow that their binding to RNA is antagonistic. Given that both proteins bind to the SK exon, one possible model for the action of hnRNP G is that its binding is synergistic with that of Tra2ß and that it sequesters Tra2ß in an RNA-bound form. This was tested in two ways. First, labelled SK exon RNA was incubated with the immobilized T7-tagged hnRNP G, as in Figure 3, but in the presence of various concentrations of GST-tagged Tra2ß (Fig. 4A). The presence of Tra2ß reduced the retention of RNA by the beads. In a second method, T7-tagged hnRNP G was eluted from the beads and cross-linked to exon SK RNA in the presence of GST-Tra2ß (Fig. 4B). Increasing concentrations of GST-Tra2ß appeared to displace hnRNP G, whereas GST alone was inactive. We conclude that the proteins bind competitively.



View larger version (51K):
[in this window]
[in a new window]
  		Figure 4. Competition with Tra2ß for binding to exon SK. (A) Labelled SK RNA was incubated with hnRNP G on beads, in the presence of increasing concentrations of GST-Tra2ß (0, 2, 20, 40 and 60 pmol in 10 µl). (B) HnRNP G eluted from beads was incubated with RNA in the presence of recombinant GST-Tra2ß or GST, and cross-linked with UV before digestion of the RNA and SDS–PAGE. Proteins were present at 0 (lane 1), 0.2, 2, 20 and 60 pmol in 10 µl (lanes 3–6 and 7–10). (-) An extract from untransfected cells was incubated with beads, eluted, incubated with RNA and cross-linked in parallel.

 
Opposite effects of hTra2ß and hnRNP G on a dystrophin pseudo-exon
The effects of hnRNP G and Tra2ß were tested upon another example of muscle-specific splicing, one in which skeletal and cardiac muscles exhibit different splicing preferences for a pseudo-exon. The deletion of ~12 kb from intron 11 of dystrophin is responsible for a familial X-linked dilated cardiomyopathy (24) (Fig. 5A). The deletion brings into close proximity sequences that resemble 3' and 5' splice sites, resulting in the creation of a pseudo-exon which, if incorporated, disrupts the proper reading frame. A higher proportion of mRNA containing this exon is produced in heart muscle than skeletal muscle, and this is presumed to be the cause of the cardiomyopathy (24). Since the levels of protein are lower in heart than skeletal muscle (24), it is probable that the higher level of mRNA including the exon in heart muscle does not represent a reduced efficiency of nonsense-linked degradation but reflects splicing preferences.



View larger version (27K):
[in this window]
[in a new window]
  		Figure 5. Effects of hnRNP G and Tra2ß on incorporation of a patient-derived dystrophin pseudo-exon. (A) Construction of the mini-genes used for transfection: (i) part of the human DMD gene between exons 11 and 12; (ii) site of the deletion of 11.6 kb in a patient with XLDC that brought two pseudo-splice sites into closer proximity, producing a pseudo-exon of 159 nt that is incorporated during splicing in cardiac muscle; (iii) sequences of various lengths including the pseudo-exon were amplified by PCR; (iv) cloning of these sequences into pcDys (Fig. 1A). In effect, the construction recreates the natural dystrophin gene between exons 11 and 12, but with the 11.6 kb deletion and deletions of 1.84 and 14.9 kb in the flanking portions of the intron. Sequences derived from the DNA of the patient are shown in black; pcDys sequences are grey; the pseudo-exon is shown with a dotted border. (B) Analysis by RT–PCR of pseudo-exon incorporation in C2C12 and H9C2 cells. The mini-genes containing insertion 1.3 kb was transfected into C2C12 or H9C2 myoblasts, and inclusion or skipping of the pseudo-exon during splicing was detected by RT–PCR. Co-transfections with plasmids encoding GFP-Tra2ß or T7-hnRNP G are as shown. White lines indicate where the order of lanes from the same gel was altered. The diagrams alongside indicate the exons in the corresponding PCR products. (C) Bar chart showing the change in the percentage of mRNA incorporating the pseudo-exon after co-transfection with plasmids expressing hnRNP G or Tra2ß. The data derive from seven separate experiments, with constructs containing various lengths of intron flanking the pseudo-exon.

 
We have cloned the pseudo-exon and flanking intron sequences into a mini-gene between dystrophin exons 11 and 12 (Fig. 5A). Expression of this construct in C2C12 (skeletal muscle) and H9C2 (cardiac muscle) myoblasts generally showed a higher level of incorporation of the pseudo-exon in H9C2 cells, although the exact ratio varied with the length of the flanking intron portions (Fig. 5B; see below). Co-expression of hnRNP G reduced incorporation in H9C2 cells, whereas expression of Tra2ß increased incorporation in C2C12 cells (Fig. 5B). A quantitative comparison confirming these results is shown in Figure 5C, which includes data from seven different experiments, with three constructs with different length introns and widely different baseline levels of incorporation.

The results were affected strongly by the sequence of the pseudo-exon. When all but the five nucleotides at each end of the exon were replaced by sequences from a constitutive exon (ß-globin exon 2), the repressive effect of hnRNP G in H9C2 cells was lost and the stimulatory effect of Tra2ß in C2C12 cells became inhibitory (Fig. 6A). If the ß-globin sequence was inserted in the opposite orientation, the exon was unaffected by either protein. We conclude that the effects of these proteins are not generally synergistic or antagonistic, and that their results cannot be attributed to a general increase or decline in splicing efficiency. Instead, both proteins can either stimulate or suppress exon inclusion, depending upon the specific pre-mRNA substrate and the cellular context.



View larger version (36K):
[in this window]
[in a new window]
  		Figure 6. Dependence of responses to hnRNP G and Tra2ß on exon sequence. (A) The pseudo-exon sequence of a mini-gene containing 0.96 kb (Fig. 5A) was replaced by rabbit ß-globin exon 2 in the sense or antisense orientations. The splicing patterns after transfection were detected by RT–PCR as in Figure 5; co-transfections with plasmids encoding GFP-Tra2ß or T7-hnRNP G are as shown. (B) The 0.96 kb mini-gene shown in Figure 5A was mutated in the region 29–21 nt from the 3' end of the pseudo-exon, and transfected into H9C2 cells. The level of incorporation is the ratio of the intensity of the PCR band derived from amplification of the mRNA containing the exon to the sum of the intensities of the bands derived from incorporation and skipping. The value is a mean from eight experiments, with error bars showing the S.D. WT, wild-type 0.96 kb construct; +G, co-transfection of hnRNP G.

 
The ß-globin substitutions showed an additional important result. Whether the level of inclusion was high or low, the substitution almost eliminated the differences in exon incorporation between C2C12 and H9C2 cells. This suggested that the differences in incorporation of the wild-type pseudo-exon might be linked to responsiveness to hnRNP G or Tra2ß. To determine whether this correlation was significant, a number of smaller substitutions were made in the pseudo-exon. If the effect of either protein was mediated by a discrete and small sequence, then it would be predicted that loss of responsiveness would be associated with a loss of cell-specific differences. Among these mutations, one was of particular interest. Nine nucleotides near the 3' end of the pseudo-exon (nt 131–139) were altered from AAGTAACAA to CTAGAATTC. The response to hnRNP G in H9C2 cells became insignificant (Fig. 6B). This change substantially increased the use of the exon in both cell lines. Whereas the level of incorporation of the wild-type exon was 48% in H9C2 cells (Fig. 6B) and 26% in C2C12 cells, the mutant exon was incorporated at 80% in H9C2 cells and 60–70% in C2C12 cells. There is a clear link between loss of hnRNP G responsiveness in the mutant exon and increased incorporation in the absence of protein co-expression.

In view of the ability of human hnRNP G to suppress inclusion of the human dystrophin pseudo-exon in the mini-genes and of human Tra2ß to enhance it, the levels of mRNA encoding these proteins were compared in different human tissues using a northern blot. A direct comparison of intensities detected by a single probe on a northern blot with RNA from various tissues is of uncertain value, particularly for skeletal muscle. RNA loadings are often adjusted such that they are normalized for actin mRNA, but the levels of actin vary and are greatly altered in skeletal muscle (25). However, comparisons between tissues based on the ratio of intensities of two different mRNA sequences can be useful. In this case, the ratio of hnRNP G to Tra2ß signals is nearly twice as high in mRNA derived from skeletal muscle as it is in mRNA from human heart (Fig. 7). This ratio may have a direct bearing on the lower level of inclusion of the pseudo-exon in skeletal muscle and thus on the failure of the patient to develop skeletal muscle dystrophy as well as cardiomyopathy. Interestingly, a northern blot of murine RNA showed no difference between the ratios in heart and skeletal muscle (not shown).



View larger version (52K):
[in this window]
[in a new window]
  		Figure 7. Northern blot analysis of expression of hnRNP G and Tra2ß in human tissues. A multiple tissue northern blot (Amersham) was probed with labelled cDNA complementary to the mRNA as shown. The intensities of the bands were measured by phosphorimager, and the hnRNP G/Tra2ß ratio is plotted below. The positions of RNA size markers are shown (kb). The multiple bands for Tra2ß have been attributed to alternative polyadenylation sites and both were included in the quantification (43).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
The interaction of hnRNP G and RBMY with hTra2ß was identified originally by yeast two-hybrid methods, and validated by both in vitro methods and co-immunoprecipitation (11). The possibility that this interaction was indirect, via RNA bridges, was excluded in several ways. In addition, we showed that the interaction between the protein-interaction domain of RBMY and Tra2ß was enhanced by phosphorylation of Tra2ß. This raised the possibility that the two proteins might act synergistically to modulate RNA splicing (11). Indeed, this possibility was borne out recently by experiments in which overexpression of hnRNP G was shown to act additively with Tra2ß to stimulate splicing of SMN2 exon 7 (22).

The results we describe here (summarized in Table 1) show that hnRNP G does not necessarily act as an auxiliary factor that augments or mediates the actions of Tra2ß. Whereas Tra2ß represses the NM exon 5 of TPM3 and stimulates the inclusion of the SK exon in a single intron construct, hnRNP G has the opposite effects: it enhances NM inclusion, and represses splicing of the SK exon construct. With a more extensive exon SK minigene, the exon is almost completely repressed in myoblasts even without hnRNP G addition, but after differentiation there is sufficient use for repression by hnRNP G to be detected.


View this table:
[in this window]
[in a new window]
  		Table 1. Summary of results from tests of hnRNP G and Tra2ß co-transfections with mini-genes. The data represent a compilation from Figures 1, 2 and 5. A stimulation of exon incorporation or (in the case of the Ad-SK constructs in Fig. 2) the efficiency of splicing is represented by + (weak) to +++ (strong); inhibition is correspondingly represented by - to - - -. The absence of a significant effect is shown by 0. For the ease of interpreting data from experiments of various types, only subjective indications are shown
 
We have investigated further the basis of the apparent antagonism in the effects of the two proteins. The ability of hnRNP G to reduce the efficiency of splicing of the pre-mRNA double reporter construct was ablated by changes in the sequence of SK exon nt 2–8 (Fig. 2C). This effect might result from the loss of an RNA-binding site for hnRNP G, or it might result from the loss of an essential component. Biochemical investigations of the RNA-binding properties of the protein have been hampered by the difficulty of preparing soluble, native protein. Such difficulties may account for the inability to demonstrate any binding preferences previously (22). To circumvent this, we used a T7 tag to isolate protein from transfected cells and to measure binding of exon SK in the presence of competitor sequences. There was clearly a preference for exon SK compared with NM. It is impossible to exclude the possibility at present that low levels of protein associated with hnRNP G mediated the binding, particularly since there was also an abundant non-specific and weak component to the binding curves (data not shown). However, it seems most likely that hnRNP G itself was responsible for the binding seen because it was by far the most abundant protein recovered, and the cross-linking assays with eluted protein demonstrated that the protein did interact directly with the RNA. Other proteins linked to the recovery of hnRNP G were detected by cross-linking. Although one of these gave a strong signal, it showed a similar response to competition with Tra2ß (Fig. 4B), and we infer that it cross-linked efficiently rather than that it was either particularly abundant or bound more strongly than hnRNP G. It may be a minor component that bound synergistically with hnRNP G. Importantly, by both assays, Tra2ß binding inhibited the binding of hnRNP G.

The antagonistic effects of hnRNP G and Tra2ß on binding to exon SK are consistent with the splicing data with the reporter construct, which contained the same tropomyosin sequences. However, the mechanism is not clear yet. The proteins might compete for binding to a single site, they might interact in their free forms, preventing interaction with RNA, or they might each sequester the RNA in such a way as to prevent binding by the other protein. Sequestration might involve the formation of large protein arrays or changes to the RNA secondary structure.

We can provisionally exclude the first mechanism, simple competition for an RNA target site, because Tra2ß seems to bind to the GAAGAA region of the enhancer (Ayres et al., submitted), which does not affect repression by hnRNP G (data not shown). The second mechanism, in which the proteins interact and the overexpressed protein would reduce the concentration of the other protein available to bind RNA, implies that the actions of the two proteins would always be antagonistic. This is at odds with our results with the SK minigene expressed in myotubes, and also with the results from the study with SMN (22). Thus, we favour a model in which the binding of Tra2ß takes place at a different site on the SK exon from that of hnRNP G; after binding, Tra2ß either nucleates local propagation along the RNA or causes local folding of RNA secondary structure in such a way that the highest affinity binding site for hnRNP G is blocked. Similarly, binding by hnRNP G might prevent the binding of Tra2ß. This model is local and specific, in that the effects of binding by one protein will depend very much on the specific RNA structures or binding sites for other proteins, and it is possible to envisage models in which binding might be synergistic.

The only other proteins whose opposing effects have been studied are SF2/ASF, often seen as the archetypal SR protein, and hnRNP A1. The SR proteins generally stimulate inclusion of exons, via binding to exonic splicing enhancers, and affect 5' splice site selection (26–29). HnRNP A1 is usually found to cause exon skipping, again via specific sequences in the exon or flanking introns (30–33), and, in contrast to SF2/ASF, it shifts 5' splice site preferences to the upstream alternative site in most model pre-mRNA substrates (34). The effects of SF2/ASF and hnRNP A1 on 5' splice site selection are the results of opposite and apparently indiscriminate effects upon the binding of U1 snRNPs to 5' splice sites (35–38). The functional antagonism of SF2/ASF and hnRNP A1 on exon incorporation or 5' splice site selection appears to be the consequence of competition for binding to the pre-mRNA (36,39). As with Tra2ß and hnRNP G, SF2/ASF can displace all hnRNP A1 from a target RNA (36,39), and it was suggested that high-affinity sites for hnRNP A1 could nucleate cooperative binding or loop formation, marking out a repressed domain, the extent of which is limited by SF2/ASF binding to its preferred sites (36,39).

Insufficient is yet known to determine whether Tra2ß and hnRNP G behave like SF2/ASF and hnRNP A1. The lack of recombinant, active hnRNP G has limited attempts to determine whether it has high specificity or, like hnRNP A1 (36,40), can bind cooperatively throughout a long RNA molecule. Indeed, the mere diversity of responses to hnRNP G that we have described here suggests that the effects of both hnRNP G and Tra2ß are very much dependent upon the context of a specific target sequence and may not involve the widespread suppression of binding by other factors that seems to be the hallmark of hnRNP A1. A possible target sequence for hnRNP G is revealed by comparing the sequences of the short regions of SK exon nt 2–8 (AAGTGTT) and the dystrophin pseudo-exon nucleotides 131–139 (AAGTAACAA) that are implicated in the response to hnRNP G. Both contain a short motif, AAGT. While we have tested only one replacement sequence in the dystrophin case, the loss of hnRNP G responsiveness in this case but not when an adjacent block of 10 nucleotides was replaced by an EcoRI site is consistent with the possibility that hnRNP G binds AAGT.

The influence of hnRNP G and Tra2ß on the use of a pseudo-exon in dystrophin showed that increased concentrations of either protein can have an effect in one cellular background but not another (Table 1). The explanation of this is not clear. One possibility is that hnRNP G, for example, is more abundant in C2C12 myoblasts and that its target sites are fully occupied even in the absence of additional exogenous protein. However, the tropomyosin constructs do respond to exogenous hnRNP G, and western blotting showed no evidence for higher levels of protein in C2C12 myoblasts (data not shown). Unfortunately, we have been unable to determine the hnRNP G/Tra2ß ratio by western blotting because the quality of several antibodies against Tra2ß was inadequate. Tra2ß expression in C2C12 myoblasts stimulated the use of the dystrophin pseudo-exon but markedly inhibited incorporation when most of the exon had been replaced by a ß-globin sequence. This confirms that the effect was not due to a generalized stimulation of splicing but involved interactions with either the exon sequences or components bound to it.

Our results suggest a possible explanation for the familial X-linked dilated cardiomyopathy produced by the deletion within intron 11 of dystrophin. Pseudo-exon incorporation is lower in the patient's skeletal muscle than it is in heart (24). We have shown that the level of incorporation of the human pseudo-exon in a mini-gene depends upon the sequence of the exon itself and upon the expression of exogenous human hnRNPG and Tra2ß, which act via the exon sequence to decrease and increase incorporation, respectively. Analysis by northern blot suggests that in human tissues there is a higher ratio of hnRNP G mRNA/Tra2ß mRNA in skeletal muscle compared with heart (Fig. 7). While we do not yet know if the difference is sufficient, it appears that the ratio of hnRNP G and Tra2ß plays a role in determining cellular splicing preferences and we speculate that repression by hnRNP G in skeletal muscle may be important in protecting the family with the dystrophin pseudo-exon from pseudo-exon incorporation in muscle and thus from skeletal muscle dystrophy. Presumably, this would not be the case if the mutant dystrophin gene were expressed in mice.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Plasmid construction
The exon-trapping vector, pcDys, was made by cloning (i) the 3' 131 nt of dystrophin exon 11 and the adjacent 116 nt of intron 11 into the AflII and KpnI sites of pcDNA3.1, and (ii) a sequence comprising the 3' 261 nt of intron 11 and the 5' 104 nt of exon 12 into the XbaI and ApaI sites. This produced a CMV-driven expression construct containing dystrophin exons 11 and 12 separated by a much-reduced intron with cloning sites at its centre. Plasmids pHC4 and pHC5 contained TPM3 sequences that had been amplified by PCR and cloned into the BamHI and XbaI sites in pcDys. For pHC4, the sequence cloned extended from 654 nt 5' of exon NM to 402 nt 3' of the exon; for pHC5, the sequence was from 391 nt 5' of exon SK to 553 3' of the exon. Deletion mutants of pHC4 and pHC5 were generated using the following primer combinations: pHC7, (Bam) 5'-CGCGGATCCTGGCTGGGTGTGGTGGC-3' and (Xho) 5'-CTCGAGAAGTCATAACTGCTACC-3'; pHC8, (Bam) 5'-CGCGGATCCGATATTAACTGCTCTCC-3' and (Xho) 5'-CCGCTCGAGAGCAACTAGGAAAGAAG-3'; pHC9, (Bam) 5'-CGCGGATCCCAGACATAACATATCATG-3' and (Xho) 5'-CCGCTCGAGGAGGCAGCTGCAAAAC-3'; pHC10, (Bam) 5'-CGCGGATCCGTTCCTGGTAGTGGATTTC and (Xho) 5'-CCGCTCGAGCCTGCTTGATGAGCAAG-3'; pHC11, (Bam) 5'-CGCGGATCCGGGTTGATGCTTGCTC AG-3' and (Xho) 5'-CCGCTCGAGGCACATGACTCCAGTA AC-3'; pHC12, (Bam) 5'-CGCGGATCCCTCTACATATAG GTT C-3' and (Xho) 5'-CCGCTCGAGTGGAATGCGTG TCTCC-3'. The PCR products were cloned in BamHI and XhoI sites of pcDys.

For the dystrophin mini-gene, DNA from the patient with cardiomyopathy was amplified and inserted into the BamHI and NotI sites of pcDys. The sequences amplified were from 367 nt upstream of the pseudo-exon (primer sequence 5'-TCCCGCGAATCCCAGTCTTCCTTGCCACTTTGAC) to 422 nt downstream (primer sequence 5'-TACGCGGCGGCCGCGGCCTTCTCTTCATTCTCTC; 0.96 kb mini-gene), or from 367 nt upstream to 825 nt downstream (primer sequence 5'-TACGCGGCGGCCGCCTGGTGTTTCCAATGTTGGGTAC; 1.3 kb mini-gene). The nine nucleotide substitution was made using inverse PCR to incorporate EcoR1 cleavage sites.

Constructs for determining splicing efficiencies (pTN23 and pTN24) were described previously (23). The mutants of exon SK are as follows: [GGA], (GGA)3 to TTTTGTGTC; [GAA], (GAA)2 to CGCTCC; [2–8], AAGTGTT to TGCATCC.

Transfection assays
The coding sequence of hnRNPG was inserted into pCG-T7 (41). The plasmid expressing Tra2ß contained the entire reading frame of the major isoform (11), fused to the C-terminus of a reading frame for green fluorescent protein in peGFP.C1 (Clontech). For co-transfections, 1–2 µg of plasmid encoding the mini-gene was added with 1–2 µg of expression vector and transfected into HEK293 cells in a 10 cm2 well GeneJammer (Stratagene). Transfections of C2C12 and H9C2 cells were done similarly when cultures were 40% confluent. RNA was extracted after 48 h with Tri Reagent (Sigma) and assayed by RT–PCR using primers directed to pcDNA3.1 sequences. To assay expression in myotubes, 50% confluent myoblasts were transfected and, after 24 h, when cells were confluent, the serum component of the medium was changed from 10% fetal calf serum to 5% horse serum; RNA was extracted after 14 days, when the cells had the morphology of myotubes. One primer was 5' end-labelled for the assays of dystrophin pseudo-exon splicing, and in this case the PCR products were resolved on a denaturing polyacrylamide gel and detected by a phosphorimager. Cycling curves were done with RNA from transfected HEK293 cells. After 20, 25, 35, 40 or 50 cycles of PCR there was no apparent difference in the ratio of incorporated and skipped products. The assays of reporter activities were done as described (23).

HnRNP G binding assays
HEK 293 cells (6x106) were transfected with 12 µg of pCG-T7/hnRNP G. After 48 h, cells were harvested and total cell lysate was prepared (23). The lysate in 0.4 ml was bound to 0.4 ml T7 Tag Affinity Antibody Agarose (Novagen,) according to the manufacturer's protocol. Washes were carried out using T7 Tag binding buffer (Novagen) and buffer DGlu–0.1% Tween 20 (Sigma). DGlu is buffer D with 80 mM potassium glutamate instead of 0.1 M KCl (42). Transcripts labelled during transcription with 32P were incubated with beads (5 µl) in BX buffer (10 mM Hepes pH 7.5, 4 mM creatine phosphate, 2 mM ATP, 5 mM MgCl2, 0.5 mM DTT, 1% glycerol) in a total volume of 10 µl for 30 min at 30°C. Samples were transferred into Spin-X filters (Costar), and centrifuged at 800g (3500 rpm in a microfuge). The beads were washed four times with 500 µl of ice-cold DGlu–0.1% Tween 20, each wash being allowed to stand for 5–10 min at ambient temperature before centrifugation as above. RNA was eluted by adding 10 µl of formamide dye per sample and incubating at 70°C for 10 min. The RNA was collected after a 2 min centrifugation at 10 000g, resolved on an 8% urea–polyacrylamide gel and detected by a phosphorimager.

For the analysis of proteins recovered by the agarose beads, cells were transfected with 20 µg of pCG-T7/hnRNP G. After 48 h, the medium was replaced with 200 µCi (0.17 nmol) [35S]L-methionine (ICN) in 2.5 ml methionine-free MEM (Sigma). After 4 h, the cells were lysed and affinity selection was done as described above.

The labelled RNA was transcribed with a cap analogue (GpppG), but unlabelled competitor RNA was not capped. The unlabelled RNA was quantified by absorbance and staining after gel electrophoresis. Phosphorylated GST-Tra2ß protein was prepared as described elsewhere (Ayres et al., submitted), and used at 2, 20, 40 or 60 pmol in each 10 µl reaction.

For UV cross-linking assays, the T7-tagged protein was recovered on beads as described above, and eluted in two batches by incubating the beads at room temperature in 0.1 M citric acid, pH 2.2. The eluate was neutralized by the addition of 0.15 vols of 2 M Tris, pH 10.4, dialysed against DGlu–0.1% Tween, and concentrated in a Centricon 30 (Amicon). UV crosslinking was performed as described previously (36), except that BX buffer was used.

Northern blots
A Hybond northern blot (Amersham, RPN4801) containing ~2 µg of poly(A)+ RNA per lane (normalized approximately with ß-actin) was probed according to the protocol supplied with PCR fragments corresponding to the complete coding sequences of Tra2ß or hnRNP G. The probes were labelled by inclusion of [{alpha}-32P]dCTP in the PCR. Signals were detected and quantified by use of a phosphorimager.


   ACKNOWLEDGEMENTS
 
We are very grateful to Dr A. Ferlini, Dr M. Dunckley and Professor F. Muntoni for gifts of genomic DNA with the X-linked dilated cardiomyopathy lesion in dystrophin, and we thank them also for support and advice. We thank Dr S.L. Chew for reading the manuscript. This work was supported by the Wellcome Trust and a CEC Framework 5 RTD project grant co-ordinated by Dr A. Ferlini.


   FOOTNOTES
 
* To whom correspondence should be addressed. Tel: +44 1162523482; Fax: +44 1162523369; Email: eci{at}le.ac.uk Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Soulard, M., Barque, J.P., Della Valle, V., Hernandez-Verdun, D., Masson, C., Danon, F. and Larsen, C.J. (1991) A novel 43-kDa glycoprotein is detected in the nucleus of mammalian cells by autoantibodies from dogs with autoimmune disorders. Exp. Cell Res., 193, 59–71.[CrossRef][ISI][Medline]

  2. Dreyfuss, G., Swanson, M.S. and Pinol-Roma, S. (1988) Heterogeneous nuclear ribonucleoprotein particles and the pathway of mRNA formation. Trends Biochem. Sci., 13, 86–91.[CrossRef][ISI][Medline]

  3. Dreyfuss, G., Matunis, M.J., PinolRoma, S. and Burd, C.G. (1993) hnRNP proteins and the biogenesis of mRNA. A. Rev. Biochem., 62, 289–321.[CrossRef][ISI][Medline]

  4. Soulard, M., Della Valle, V., Siomi, M.C., Pinol-Roma, S., Codogno, P., Bauvy, C., Bellini, M., Lacroix, J.C., Monod, G., Dreyfuss, G. et al. (1993) HnRNP G: sequence and characterization of a glycosylated RNA-binding protein. Nucl. Acids Res., 21, 4210–4217.[Abstract/Free Full Text]

  5. Krecic, A.M. and Swanson, M.S. (1999) HnRNP complexes: composition, structure, and function. Curr. Opin. Cell Biol., 11, 363–371.[CrossRef][ISI][Medline]

  6. Ma, K., Inglis, J.D., Sharkey, A., Bickmore, W.A., Hill, R.E., Prosser, E.J., Speed, R.M., Thomson, E.J., Jobling, M., Taylor, K. et al. (1993) A Y chromosome gene family with RNA-binding protein homology: candidates for the azoospermia factor AZF controlling human spermatogenesis. Cell, 75, 1287–1295.[CrossRef][ISI][Medline]

  7. Delbridge, M.L., Ma, K., Subbarao, M.N., Cooke, H.J., Bhasin, S. and Graves, J.A.M. (1998) Evolution of mammalian HNRNPG and its relationship with the putative azoospermia factor RBM. Mammal. Genome, 9, 168–170.[CrossRef][ISI][Medline]

  8. Delbridge, M.L., Lingenfelter, P.A., Disteche, C.M. and Graves, J.A. (1999) The candidate spermatogenesis gene RBMY has a homologue on the human X chromosome. Nat. Genet, 22, 223–224.[CrossRef][ISI][Medline]

  9. Mazeyrat, S., Saut, N., Mattei, M.G. and Mitchell, M.J. (1999) RBMY evolved on the Y chromosome from a ubiquitously transcribed X–Y identical gene. Nat. Genet, 22, 224–226.[CrossRef][ISI][Medline]

  10. Venables, J.P., Vernet, C., Chew, S.L., Elliott, D.J., Cowmeadow, R.B., Wu, J., Cooke, H.J., Artzt, K. and Eperon, I.C. (1999) T-STAR/ETOILE: a novel relative of SAM68 that interacts with an RNA- binding protein implicated in spermatogenesis. Hum. Mol. Genet., 8, 959–969.[Abstract/Free Full Text]

  11. Venables, J.P., Elliott, D.J., Makarova, O.V., Makarov, E.M., Cooke, H.J. and Eperon, I.C. (2000) RBMY, a probable human spermatogenesis factor, and other hnRNP G proteins interact with tra2beta and affect splicing. Hum. Mol. Genet., 9, 685–694.[Abstract/Free Full Text]

  12. Elliott, D.J., Bourgeois, C.F., Klink, A., Stevenin, J. and Cooke, H.J. (2000) A mammalian germ cell-specific RNA-binding protein interacts with ubiquitously expressed proteins involved in splice site selection. Proc. Natl Acad. Sci. USA, 97, 5717–5722.[Abstract/Free Full Text]

  13. Tian, M. and Maniatis, T. (1993) A splicing enhancer complex controls alternative splicing of doublesex pre-mRNA. Cell, 74, 105–114.[CrossRef][ISI][Medline]

  14. Lynch, K.W. and Maniatis, T. (1996) Assembly of specific SR protein complexes on distinct regulatory elements of the Drosophila doublesex splicing enhancer. Genes Dev., 10, 2089–2101.[Abstract/Free Full Text]

  15. Heinrichs, V., Ryner, L.C. and Baker, B.S. (1998) Regulation of sex-specific selection of fruitless 5' splice sites by transformer and transformer-2. Mol. Cell. Biol., 18, 450–458.[Abstract/Free Full Text]

  16. Chandler, D.S., McGuffin, M.E. and Mattox, W. (2001) Functionally antagonistic sequences are required for normal autoregulation of Drosophila tra-2 pre-mRNA splicing. Nucl. Acids Res., 29, 3012–3019.[Abstract/Free Full Text]

  17. Tacke, R., Tohyama, M., Ogawa, S. and Manley, J.L. (1998) Human Tra2 proteins are sequence-specific activators of pre-mRNA splicing. Cell, 93, 139–148.[CrossRef][ISI][Medline]

  18. Hofmann, Y., Lorson, C.L., Stamm, S., Androphy, E.J. and Wirth, B. (2000) Htra2-beta 1 stimulates an exonic splicing enhancer and can restore full-length SMN expression to survival motor neuron 2 (SMN2). Proc. Natl Acad. Sci. USA, 97, 9618–9623.[Abstract/Free Full Text]

  19. Arikan, M.C., Memmott, J., Broderick, J.A., Lafyatis, R., Screaton, G., Stamm, S. and Andreadis, A. (2002) Modulation of the membrane-binding projection domain of tau protein: splicing regulation of exon 3. Brain Res. Mol. Brain Res., 101, 109–121.[Medline]

  20. Stamm, S., Casper, D., Hanson, V. and Helfman, D.M. (1999) Regulation of the neuron-specific exon of clathrin light chain B. Brain. Res. Mol. Brain Res., 64, 108–118.[Medline]

  21. Young, P.J., DiDonato, C.J., Hu, D., Kothary, R., Androphy, E.J. and Lorson, C.L. (2002) SRp30c-dependent stimulation of survival motor neuron (SMN) exon 7 inclusion is facilitated by a direct interaction with hTra2beta1. Hum. Mol. Genet., 11, 577–587.[Abstract/Free Full Text]

  22. Hofmann, Y. and Wirth, B. (2002) hnRNP-G promotes exon 7 inclusion of survival motor neuron (SMN) via direct interaction with Htra2-beta1. Hum. Mol. Genet., 11, 2037–2049.[Abstract/Free Full Text]

  23. Nasim, M.T., Chowdhury, H.M. and Eperon, I.C. (2002) A double reporter assay for detecting changes in the ratio of spliced and unspliced mRNA in mammalian cells. Nucl. Acids Res., 30, E109–9.[Medline]

  24. Ferlini, A., Galie, N., Merlini, L., Sewry, C., Branzi, A. and Muntoni, F. (1998) A novel Alu-like element rearranged in the dystrophin gene causes a splicing mutation in a family with X-linked dilated cardiomyopathy. Am. J. Hum. Genet., 63, 436–446.[CrossRef][ISI][Medline]

  25. Hsiao, L.L., Dangond, F., Yoshida, T., Hong, R., Jensen, R.V., Misra, J., Dillon, W., Lee, K.F., Clark, K.E., Haverty, P. et al. (2001) A compendium of gene expression in normal human tissues. Physiol. Genom., 7, 97–104.[Abstract/Free Full Text]

  26. Graveley, B.R. (2000) Sorting out the complexity of SR protein functions. RNA, 6, 1197–1211.[CrossRef][ISI][Medline]

  27. Caceres, J.F. and Kornblihtt, A.R. (2002) Alternative splicing: multiple control mechanisms and involvement in human disease. Trends Genet., 18, 186–193.[CrossRef][ISI][Medline]

  28. Cartegni, L., Chew, S.L. and Krainer, A.R. (2002) Listening to silence and understanding nonsense: exonic mutations that affect splicing. Nat. Rev. Genet., 3, 285–298.[CrossRef][ISI][Medline]

  29. Hastings, M.L. and Krainer, A.R. (2001) Pre-mRNA splicing in the new millennium. Curr. Opin. Cell Biol., 13, 302–309.[CrossRef][ISI][Medline]

  30. Mayeda, A., Helfman, D.M. and Krainer, A.R. (1993) Modulation of exon skipping and inclusion by heterogeneous nuclear ribonucleoprotein A1 and pre-mRNA splicing factor SF2/ASF. Mol. Cell Biol., 13, 2993–3001.[Abstract/Free Full Text]

  31. Caputi, M., Mayeda, A., Krainer, A.R. and Zahler, A.M. (1999) hnRNP A/B proteins are required for inhibition of HIV-1 pre-mRNA splicing. EMBO J., 18, 4060–4067.[CrossRef][ISI][Medline]

  32. Blanchette, M. and Chabot, B. (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][ISI][Medline]

  33. Del Gatto-Konczak, F., Olive, M., Gesnel, M.C. and Breathnach, R. (1999) HnRNP A1 recruited to an exon in vivo can function as an exon splicing silencer. Mol. Cell Biol., 19, 251–260.[Abstract/Free Full Text]

  34. Mayeda, A. and Krainer, A.R. (1992) Regulation of alternative pre-mRNA splicing by hnRNP A1 and splicing factor SF2. Cell, 68, 365–375.[CrossRef][ISI][Medline]

  35. Eperon, I.C., Ireland, D.C., Smith, R.A., Mayeda, A. and Krainer, A.R. (1993) Pathways for selection of 5' splice sites by U1 snRNPs and SF2/ASF. EMBO J., 12, 3607–3617.[ISI][Medline]

  36. Eperon, I.C., Makarova, O.V., Mayeda, A., Munroe, S.H., Caceres, J.F., Hayward, D.G. and Krainer, A.R. (2000) Selection of alternative 5' splice sites: role of U1 snRNP and models for the antagonistic effects of SF2/ASF and hnRNP A1. Mol. Cell Biol., 20, 8303–8318.[Abstract/Free Full Text]

  37. Kohtz, J.D., Jamison, S.F., Will, C.L., Zuo, P., Luhrmann, R., GarciaBlanco, M.A. and Manley, J.L. (1994) Protein–Protein interactions and 5'-splice-site recognition in mammalian mRNA precursors. Nature, 368, 119–124.[CrossRef][Medline]

  38. Xiao, S.H. and Manley, J.L. (1997) Phosphorylation of the ASF/SF2 RS domain affects both protein–protein and protein–RNA interactions and is necessary for splicing. Genes Dev., 11, 334–344.[Abstract/Free Full Text]

  39. Zhu, J., Mayeda, A. and Krainer, A.R. (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][ISI][Medline]

  40. CasasFinet, J.R., Smith, J.J., Kumar, A., Kim, J.G., Wilson, S.H. and Karpel, R.L. (1993) Mammalian heterogeneous ribonucleoprotein A1 and its constituent domains. Nucleic acid interaction, structural stability and self-association. J. Mol. Biol., 229, 873–889.[CrossRef][ISI][Medline]

  41. Caceres, J.F., Misteli, T., Screaton, G.R., Spector, D.L. and Krainer, A.R. (1997) Role of the modular domains of SR proteins in subnuclear localization and alternative splicing specificity. J. Cell Biol., 138, 225–238.[Abstract/Free Full Text]

  42. Black, D.L. (1992) Activation of c-src neuron-specific splicing by an unusual RNA element in vivo and in vitro. Cell, 69, 795–807.[CrossRef][ISI][Medline]

  43. Nayler, O., Cap, C. and Stamm, S. (1998) Human transformer-2-beta gene (SFRS10): complete nucleotide sequence, chromosomal localization, and generation of a tissue-specific isoform. Genomics, 53, 191–202.[CrossRef][ISI][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Hum Mol GenetHome page
I. Ehrmann, C. Dalgliesh, A. Tsaousi, M. P. Paronetto, B. Heinrich, R. Kist, P. Cairns, W. Li, C. Mueller, M. Jackson, et al.
Haploinsufficiency of the germ cell-specific nuclear RNA binding protein hnRNP G-T prevents functional spermatogenesis in the mouse
Hum. Mol. Genet., September 15, 2008; 17(18): 2803 - 2818.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
P. Cloutier, J. Toutant, L. Shkreta, S. Goekjian, T. Revil, and B. Chabot
Antagonistic Effects of the SRp30c Protein and Cryptic 5 ' Splice Sites on the Alternative Splicing of the Apoptotic Regulator Bcl-x
J. Biol. Chem., August 1, 2008; 283(31): 21315 - 21324.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
M. T. Nasim, A. Ghouri, B. Patel, V. James, N. Rudarakanchana, N. W. Morrell, and R. C. Trembath
Stoichiometric imbalance in the receptor complex contributes to dysfunctional BMPR-II mediated signalling in pulmonary arterial hypertension
Hum. Mol. Genet., June 1, 2008; 17(11): 1683 - 1694.
[Abstract] [Full Text] [PDF]

Home page
 RNAHome page
C. Paradis, P. Cloutier, L. Shkreta, J. Toutant, K. Klarskov, and B. Chabot
hnRNP I/PTB can antagonize the splicing repressor activity of SRp30c
RNA, August 1, 2007; 13(8): 1287 - 1300.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
K.-H. Shin, M. K. Kang, R. H. Kim, R. Christensen, and N.-H. Park
Heterogeneous Nuclear Ribonucleoprotein G Shows Tumor Suppressive Effect against Oral Squamous Cell Carcinoma Cells.
Clin. Cancer Res., May 15, 2006; 12(10): 3222 - 3228.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
D. O. Watermann, Y. Tang, A. zur Hausen, M. Jager, S. Stamm, and E. Stickeler
Splicing Factor Tra2-{beta}1 Is Specifically Induced in Breast Cancer and Regulates Alternative Splicing of the CD44 Gene.
Cancer Res., May 1, 2006; 66(9): 4774 - 4780.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
A. Disset, C.F. Bourgeois, N. Benmalek, M. Claustres, J. Stevenin, and S. Tuffery-Girau