GTC
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Genes to Cells (2005) 10, 531-541. doi:10.1111/j.1365-2443.2005.00855.x
© 2005 Blackwell Publishing or its licensors
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
Right arrow Citation Map
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
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fushimi, K.
Right arrow Articles by Tsukahara, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fushimi, K.
Right arrow Articles by Tsukahara, T.

NSSRs/TASRs/SRp38s function as splicing modulators via binding to pre-mRNAs

Kazuo Fushimi1, Noriko Osumi2 and Toshifumi Tsukahara1,*

1 Center for Nano Materials and Technology, Japan Advanced Institute of Science and Technology, Nomi-City, Ishikawa 923-1292, Japan
2 Division of Developmental Neuroscience, Center for Translational and Advanced Animal Research on Human Diseases, Tohoku University Graduate School of Medicine, Aoba-ku, Sendai 980-8575, Japan


   Abstract
Top
 Abstract
Introduction
 Results
 Discussion
 Experimental procedures
 Supplementary material
 References
 
The genes for neural-salient serine/arginine-rich (NSSR) proteins 1 and 2 have been cloned from the neuronal differentiated embryocarcinoma cell line, P19. NSSRs contain an RNA recognition motif (RRM) at the N-terminal and several SR rich regions at the C-terminal resembling RS domains. We found that NSSRs associated with U1-70k, and determined the exon inclusion activity of NSSRs’ C-terminals. First, the RRM was changed to the MS2 coat protein (MS2CP) and then, MS2 RNA stem-loops were inserted in the middle of the exon N of the clathrin light chain B minigene as an artificial exonic splicing enhancer to be recognized by the MS2CP. The modified exon N of the pre-mRNA was included by the MS2CP switched NSSR 1, but it was excluded by the MS2CP switched NSSR 2. The deletion analysis of the MS2CP switched NSSR 1 showed that the middle SR rich region was responsible for the activity of the modified exon N inclusion. Furthermore, the RRM domain of NSSRs recognized mRNAs. NSSRs were expressed in the nervous system, especially in cerebellar and hippocampal primordia, ventricular zone of the neocortex and olfactory bulb primordia, retina, and olfactory epithelium at E15.5, all containing undifferentiated neural stem cells. Taken together, our results showed that NSSRs modulate alternative splicing via binding to premRNAs during neural differentiation.


   Introduction
Top
 Abstract
 Introduction
Results
 Discussion
 Experimental procedures
 Supplementary material
 References
 
Alternative splicing is one way of generating multiple gene products from a single gene at the post-transcriptional level. A survey of alternative splicing with exon junction microarrays suggests that at least 74% of human multiexon genes are alternatively spliced (Johnson et al. 2003). In fact, many neurotransmitter receptors, such as dopamine D2 receptor (Khan et al. 1998), mGlu5 (Mion et al. 2001), NMDA R1 (Kuppenbender et al. 1999), GlyR{alpha}2 (Polydorides et al. 2000), GABA(A) receptor {gamma} 2 (Jensen et al. 2000), serotonin receptor (Heidmann et al. 1998) and GluR1-4 (Lambolez et al. 1996) genes, have alternative splicing products that are expressed by specific types of neurons. In addition, many analyses indicated that the alternative splicing products of a single gene have different functions (Valenzuela et al. 1993; Lopez 1995; Eide et al. 1996).

Serine/arginine-rich (SR) proteins have RNA binding domains (RRM domains) at the N-terminal and an arginine/serine-rich domain (RS domain) at the C-terminal. They regulate alternative splicing by interacting with U1-70k and U2AF35, which determine 5' and 3' splicing sites, respectively (Ge & Manley 1990; Wu & Maniatis 1993; Cao & Garcia-Blanco 1998; Graveley 2000). On the other hand, exonic cis-acting elements can be often found in pre-mRNAs. The consensus sequences of many SR proteins have been determined, and they act as exonic cis-acting elements, which are known as exonic splicing enhancers (ESEs). In addition, after SR proteins recognize ESEs, the RS domains contact with the pre-mRNA branchpoints (Shen et al. 2004).

Previously, we reported two SR protein-like gene products, neural-salient serine/arginine-rich proteins 1 and 2 (NSSR 1 and 2), which are generated from a single gene, and are thought to be splicing isoforms (Komatsu et al. 1999). NSSR 1 and 2 mRNAs are present in the brain and testis at higher levels than in other tissues. NSSR 1 is expressed in the neural stage during the neuroectoderm differentiation of embryocarcinoma cells. The expression of NSSR 2 prevents the inclusion of either the Flip or Flop exons in the splicing of the GluR-B gene, resulting in an increase in the abnormal exon-skipping product. However, transient transfection with NSSR 1 promotes the inclusion of the Flip exon in the splicing of the GluR-B gene (Komatsu et al. 1999). Although NSSRs affect alternative splicing, the consensus sequence of RRM in NSSRs has not been determined.

Previously, Yang et al. (1998, 2000) cloned the genes for the SR proteins, TASR-1 and 2, as translocated in liposarcoma (TLS) associated proteins using the yeast two-hybrid system, and showed that both gene products can process E1a pre-mRNA. Their results indicate that NSSR 1 and 2 are identical to TASR-2 and 1, respectively. TLS-ERG leukemia fusion protein interacts with both TSARs and inhibits the RNA splicing mediated by TASRs (Yang et al. 2000). Shin & Manley (2002) cloned SR proteins with unexpected properties called SRp38. The dephosphorylated SRp38 accumulates by heat shock and inhibits splicing at an early step (Shin et al. 2004). In addition, during the M phase SRp38 is dephosphorylated as well (Shin & Manley 2002). SRp38 and SRp38-2 are also identical to NSSR 1 and 2, respectively.

Phage MS2 is a bacterial RNA virus that infects E. coli. The MS2 coat protein (MS2CP) controls the sequence-specific RNA encapsidation and repression of the replicase translation by binding to the RNA stem-loop structure of the 19 nucleotides in the viral genome (Peabody & Ely 1992). Recently, the interaction between MS2CP and MS2 RNA has been applied to the analysis of RNA-protein interaction, splicing, etc. For example, spliceosomes are isolated using the interaction between MS2CP and MS2 RNA (Zhou et al. 2002). According to the analysis of fusion proteins consisting of MS2CP and SR dipeptide repeat, a hybrid of Drosphila doublesex pre-mRNA and the MS2 stem loop RNA, dsx(70)M1, undergoes splicing in the presence of the fusion protein consisting of MS2CP and 14 RS dipeptides in vitro. However, it is not possible to activate the splicing by the fusion protein including seven RS dipeptides (Philipps et al. 2003).

Clathrin light chain B is one of the components of clathrin coated vesicle. There are two alternative splicing isoforms, a ubiquitous and a brain-specific form. The brain-specific form has 54 additional nucleotides as a result of the exon N inclusion. A minigene consisting of the exon IV, N and V of the clathrin light chain B gene has been constructed to study neuron-specific alternative splicing (Stamm et al. 1992).

In this paper, we demonstrated the associations of NSSRs with U1-70k. The switching of the NSSRs’ RRM to MS2CP suggested that NSSRs can modulate alternative splicing via binding to pre-mRNAs. Deletion analyses of the MS2CP switched NSSR 1 indicated that the middle of the NSSR 1 specific SR rich region regulates the inclusion of the modified exon N of the clathrin light chain B minigene. Furthermore, the expression of NSSRs in undifferentiated neural stem cells during the development was confirmed by in situ hybridization.


   Results
Top
 Abstract
 Introduction
 Results
Discussion
 Experimental procedures
 Supplementary material
 References
 
The association of NSSRs with U1-70k

One of the functions of SR proteins is the selection of 5' splicing sites by binding to U1-70k through their SR domains. The hybrid protein consisting of MS2CP and seven RS dipeptides is ineffective in the stimulation of a splicing activity against the target in vitro (Philipps et al. 2003). However, the RS like domains of the C-terminals of NSSRs consist of several three RS dipeptides (Fig. 4A). Therefore, we examined the association between NSSRs and U1-70k by yeast two-hybrid assays. In the presence of NSSR 1 and 2 fused with the GAL4 DNA binding domain (BD), U1-70k fused with the GAL4 activation domain (AD) rescued the auxotrophy, as well as in the presence of ASF/SF2 fused with the GAL4 DNA BD on the highest stringent condition (–Ade and –His). All of the growing colonies were ß-galactosidase positive (data not shown). In contrast, BD and AD themselves did not rescue the auxotrophy in the presence of U1-70k fused with AD and NSSR fused with BD, respectively (Fig. 1B).



View larger version (32K):
[in this window]
[in a new window]
  		Figure 4  Deletion constructs of NSSR 1 C-terminal. (A) The primary structure of NSSR 1. The RRM is shown in gray and three SR rich regions are underlined. (B) Five partial fragments of the NSSR 1 C-terminal were fused with MSCP. The 79–183 fragment of NSSR 1 is the common region among NSSR 1 and 2 C-terminal.

 


View larger version (49K):
[in this window]
[in a new window]
  		Figure 1  Schematic structures of NSSR proteins and interaction of NSSRs with U1-70k. (A) NSSR 1 and 2 have an RNA recognition motif (RRM) at the N-terminal and several serine and arginine rich regions that are located at the C-terminal. (B) NSSRs were inserted into pAS2-1 to fuse with the GAL4 BD and introduced into AH109. Y187 expressing the U1-70k fused with the GAL4 AD was mated with the transformed AH109. The obtained yeasts were grown in the liquid media until cells were saturated. The media was diluted 10, 100, and 1000 times, and 5 µL of each medium was plated on to an SD agar plate lacking adenine, histidine, leucine, and tryptophan. The interaction of U1-70K with ASF/SF2 was detected as previously reported. The co-expression of NSSR 1 and U1-70K resulted in the slow growth of the yeasts, but no growth in the negative controls was observed under the highest stringency (–His, –Ade). NSSR 2 also rescued the growth of the yeast in the presence of U1-70k. (C) Myc tagged U1-70k and FLAG tagged NSSRs were over-expressed in COS 7 cells by transient transfection, and the cells were lyzed to perform immunoprecipitation with anti-myc antibody. The immunocomplexes included tagged NSSRs in the presence of U1-70k (top panel). The lysates were analyzed by Western blot (WB) analysis to confirm the expression of FLAG-tagged NSSRs (middle panel) and myc-tagged U1-70k (bottom panel).

 
The results of the two-hybrid assays were confirmed in the COS7 cells. The expression vectors for myc-tagged U1-70k and FLAG-tagged NSSRs were introduced into COS7 cells, and immunoprecipitation with anti-myc tag agarose was carried out in the presence of RNase A. NSSR 1 and 2 were detected in the immunocomplexes of U1-70k (Fig. 1C).

Regulation of the clathrin light chain B minigene containing stem loops of MS2 RNA by MS2 coat protein fused NSSRs’ C-terminals

In order to determine how NSSRs contribute to alternative splicing, the clathrin light chain B minigene was applied as a model for neuronal specific alternative splicing, because the exon N of the clathrin light chain B gene is included during the neuronal differentiation of P19 cells (Stamm et al. 1992). NSSRs themselves do not affect the clathrin light chain B minigene (Supplementary Fig. S1). In addition, the recognition sequence of NSSRs’ RRM has not been determined. Therefore, the RRM motif of NSSR 1 and 2 was replaced with an RNA binding protein, MS2CP. In order to localize in nuclei, three nuclear localization signals were inserted between the MS2CP and NSSR C-terminals (Fig. 2A). Furthermore, a myc-tag was added to the C-terminals. As a result of the introduction of these expression vectors into COS7 cells, bands of appropriate sizes were detected by Western blot analysis of the lysates using a myc antibody. These results suggested that these expression vectors work in mammalian cells (Fig. 2C).



View larger version (45K):
[in this window]
[in a new window]
  		Figure 2  Effects of MS2 coat fused NSSRs’ C-terminals on neural specific exon. (A) The RRM region of NSSRs was switched to the RNA bacteriophage coat protein (MSCP) in order to detect effects on alternative splicing (MS2CNS1 and MS2CNS2). To assure the nuclear localization of the RRM switched NSSRs, three nuclear localization signals (NLS) were inserted between MSCP and the C-terminals. (B) To study the effects of NSSRs on the neuron specific alternative splicing, the clathrin light chain B minigene (pJS74), which contains exon IV, neuron specific exon N and exon V, was modified as a model for neuronal alternative splicing. MS2 RNA stem loops were inserted into the middle of the exon N to be recognized by MS2CNS1 and 2. (C) The protein expression by MS2CNS1 and MS2CNS2 expression vectors was confirmed by transfection into COS7 cells. The cell lysates were loaded on to SDS-PAGE, and transferred on to a sheet of PVDF membrane to detect the expressed proteins by anti-myc antibody. Appropriate sizes of bands were observed from the lysates other than the negative control (pDSRED). (D) The NSSR C-terminals fused with MSCP and the clathrin light chain B minigenes were transfected into NIH3T3 and COS7 cells. The processed pre-mRNAs were amplified by RT-PCR with the minigene specific primers. The exon N of pJS74 was excluded from NIH3T3 cells according to other non-neural cells as reported previously (pJS74 lanes 1–3). Some of the modified exons N were included in the mRNA in the absence of MSCP fused with C-terminals (pJSMS2r, lane 3). The NSSR 1 C-terminal activated the inclusion of the modified exons N further, but the NSSR 2 C-terminal inhibited the inclusion in NIH3T3 and COS7 cells. The results were reproducible and the detected bands were confirmed by sequencing.

 
The stem loops of MS2 RNA, which are recognized by MS2CP, were inserted into the middle of the exon N of the clathrin light chain B minigene driven by an SV40 promoter (Fig. 2B). These expression vectors were transiently transfected into COS7, NIH 3T3 and N2a cells. Then, total RNA was extracted from these cells followed by RT-PCR with minigene specific primers to determine the splicing forms of their minigene's mRNA. No signal was detected when total RNA were extracted from untransfected cells (Fig. 3, Supplementary Fig. S1). In the case of the original minigene, the neural-specific form was not detected in the absence of the NSSRs’ C-terminals. However, both the neural and non-neural forms were detected, when the MS2 stem-loops were inserted into the exon N (Fig. 2D). The expression of the NSSR 1's C-terminal fused with MS2CP further activated the inclusion of the exon N containing MS2 RNA, but the original exon N was not included at all (Fig. 2D). In contrast, the NSSR 2's C-terminal fused with MS2CP repressed the inclusion of the exon N containing MS2 RNA (Fig. 2D). In addition, the activities of NSSRs’ C-terminals were in a dose-dependent manner (Fig. 3) and observed in NIH3T3, COS7 and N2a cell lines (Figs 2D and 5).



View larger version (21K):
[in this window]
[in a new window]
  		Figure 3  The dose-dependency of the activity of NSSRs’ C-terminals. The modified minigene and various amounts of the expression vectors were co-transfected into NIH3T3 cells. The total amounts of DNA were adjusted with the empty vector. (A) The mRNAs derived from the minigene were amplified by RT-PCR followed by 3% agarose gel electrophoresis. The products were stained with ethidum bromide. The proportion of the fragments that included exon N in the detected bands by the electrophoresis was shown in (B) (MS2CNS1) and (C) (MS2CNS2). The experiments were triplicated and the standard deviations are shown by error bars.

 


View larger version (40K):
[in this window]
[in a new window]
  		Figure 5  The exon inclusion activity of the deletion mutants of NSSR 1 C-terminal. Each deletion mutant was introduced into (A) N2a and (B) NIH3T3 cells with modified clathrin light chain B minigene to detect their splicing regulation as described in Figure 2. The numerals at the bottom of lanes indicate the proportion of the fragments that included exon N in the detected bands. MS2CNS1 and MS2CNS2 possessed the inclusion and exclusion activity, respectively, in N2a cells as well as CCS7 and NIH3T3 cells (A). The 143–234 region of NSSR 1 was necessary for the exon N inclusion in both of N2a and NIH3T3 cells. Half of the cells into which the plasmids were introduced were lyzed with the SDS sample buffer containing 5 M urea and then, the lysates were analyzed by Western blotting. The deletion mutants were detected with anti-myc antibody. The experiments were performed twice, and similar results were obtained.

 
Deletion analysis of the NSSR 1 C-terminal

Deletion mutations of the C-terminal of NSSR 1 were fused to MS2CP, and then applied to the clathrin light chain B minigene analysis using N2a and 3T3 cells. The C-terminal RS region of NSSR 1 (amino acids 222–262) had no effect on either the inclusion or exclusion of the minigene. Although the common sequence among NSSR 1 and 2, the C-terminal RS region of NSSR 1 or both were deleted, the inclusion activity of the modified exon N was retained in 3T3 and N2a (Figs 4 and 5).

The RNA binding activity of the RRM domain in NSSRs

Since SR proteins recognize ESEs by their RRM, the binding activity of NSSRs to exons is one of our interests. In order to elucidate this issue, a bacterial recombinant NSSRs’ RRM was produced. The recombinant contained a cellulose binding domain (CBD)-tag at the N-terminal. Therefore, the NSSRs’ RRM was trapped on cellulose resin. Western blot analysis revealed that about 0.5 µg of the full length of the recombinant protein was absorbed on the resin (Fig. 6). As a result of the incubation on the resin with 20 µg of mRNA from murine brains, 10 ng of mRNAs were obtained from the resin (Fig. 6B, Table 1). RNAs at 89 bases in length were detected in both tRNAs and the RNAs obtained from the resin (Fig. 6B,C). Because the resin was precoated with yeast tRNAs, a non-specific interaction probably occurred between tRNAs and the cellulose resin. The electropherogram of the murine brain mRNAs showed obvious contamination of 18 S ribosomal RNA, but the RNAs obtained from the resin contained considerably less than 18 S ribosomal RNA (Fig. 6B,C). The results suggest that the NSSRs’ RRM recognizes mRNA in a sequence specific manner.



View larger version (16K):
[in this window]
[in a new window]
  		Figure 6  The RNA binding activity of the NSSRs’ RRM. The bacterial recombinant of NSSRs’ RRM tagged by a cellulose binding domain (CBD) at the N-terminal was absorbed on 30 mg of cellulose resin. One twentieth of the resin was analyzed by Western blotting using anti-CBD antibody. The intensity of the band for the full length CBD-tagged NSSRs’ RRM was equivalent to 16 ng of the purified CBD protein (A). Twenty micrograms of mRNA from murine brains was added on to the resin which was precoated with 20 µg of yeast tRNAs. After an incubation for 10 min, the resin was washed four times. The binding RNAs were extracted and analyzed with Agilent 2100 bioanalyzer (B). The electropherograms of the yeast tRNAs and murine brain mRNA used are shown in (C).

 

View this table:
[in this window]
[in a new window]
  		Table 1 The absorbed RNA by the NSSRs’ RRM on the resin
 
In situ hybridization analysis of NSSR 1 and 2 in embryos

Since NSSRs fused with MS2CP regulated neuronal-specific alternative splicing, it is possible that NSSRs are components of a spliceosome that plays a role in neuron-specific alternative splicing. In addition, the expression of NSSR 1 is observed in P19 cells during neuronal differentiation, suggesting participation of NSSRs in neural differentiation (Komatsu et al. 1999). Therefore, we examined the expression of NSSRs in vivo by in situ hybridization. For the in situ hybridization, three probes (NSSRC, NSSA1 and NSSA2) were used. NSSRC recognizes mRNAs of both NSSR 1 and 2, while NSSA1 and NSSA2 are specific for NSSR 1 and 2, respectively. NSSR 1 and 2 mRNAs were prominently distributed in the ventricular zone of the cortical, cerebellar, hippocampal and olfactory bulb primordia, while the rest of the CNS showed only a weak expression of NSSRs at E15.5 (Fig. 7). In addition, an intense signal was observed in the retina, olfactory epithelium and tooth germ mesenchyme, all containing the undifferentiated neural stem cells (Fig. 8). Futhermore, the epithelium of the intestine and salivary gland also showed high levels of the mRNA of NSSRs by NSSRC and NSSA1 (Fig. 8). The expression of NSSR mRNAs was not only observed in the neural retina, but also in the lens epithelium, where proliferation and differentiation occur (Fig. 8A). Taken together, the expression of NSSRs in various tissues was highly specific to undifferentiated cells. Overall, a similar distribution of hybridization signals was detected by NSSRC, NSSA1 or NSSA2. However, the signal for NSSA2 was very faint, in total (Fig. 8A,B). A weak expression of NSSRs was detected in muscle, but the expression of NSSR mRNAs was less than the detection limit in other tissues (data not shown). The sense probe as a negative control demonstrated that NSSRC, NSSA1 and NSSA2 hybridized to NSSR mRNAs specifically (data not shown).



View larger version (41K):
[in this window]
[in a new window]
  		Figure 7  In situ hybridization analysis of NSSRs in the central nervous system at E15.5.Complementary RNAs were prepared as a probe for the in situ hybridization from common (NSSRC), NSSR 1 specific (NSSA1) and NSSR 2 specific (NSSA2) regions. (A) The sequence shared with NSSR 1 and 2 is shown in black. The gray area indicates the specific regions. (B) The sagittal section of the telencephalon at E15.5 was hybridized by the NSSA1 probe. The hybridization signal in the cortical plate (arrow) was weaker than that in the ventricular zone of the telencephalon (arrowhead). The asterisk indicates the lateral ventricle, and the striatum is enclosed by the line. (C) The NSSRC probe was hybridized to the cerebellar primordium (arrows). The asterisk indicates the fourth ventricle. (D) The transverse section of the hippocampal primordium was hybridized by the NSSRC probe. The area indicated by the arrow would form the hippocampus. The asterisk indicates the lateral ventricle.

 


View larger version (76K):
[in this window]
[in a new window]
  		Figure 8  The distrubution of NSSRs other than the central nervous system at E15.5 embryo. (A) The NSSRC probe was hybridized to the eye primordium. In addition to the intense signal in the retina, the lens epithelium expressed NSSR mRNAs, as indicated by the arrows. (B) The hybridization to the retina by the NSSA2 probe resulted in a similar staining pattern to the NSSRC probe. (C) The coronal section of the head was hybridized by the NSSA1 probe. The ventricular zone of the olfactory bulbs (arrows) and nasal epithelium surrounding the nasal cavity (asterisks) were positive. (D) NSSR mRNA was detected in the dental mesenchyme (arrows) of the tooth germs at the bell stage, which are enclosed by the lines. Contrary to the mesenchyme, the epithelium (arrowheads) are negative for NSSR mRNA. The salivary gland (E) and intestine (F) hybridized by the NSSRC probe were showing a strong signal in the epithelium.

 

   Discussion
Top
 Abstract
 Introduction
 Results
 Discussion
Experimental procedures
 Supplementary material
 References
 
U1, U2, U4, U5 and U6 snRNPs play a major role in pre-mRNA splicing. U1 and U2 snRNPs bind to a pre-mRNA at the 5' splicing site and branch point region, respectively, and form an early spliceosome. This suggests that U1 and U2 snRNP determine the splice site. U1-70k is a component of U1 snRNP and interacts with the SR protein, ASF/SF2. Therefore, the recognition of ESEs by SR proteins is important for the exon inclusion and selection of 5' splicing sites (Graveley 2000). In the presence of the GAL4 AD fused U1-70k, the GAL BD fused NSSR 1 and 2 rescued the auxotrophy. Although the activity of NSSR 1 was much weaker compared to NSSR 2 and ASF/SF2, the results could be confirmed by the immunoprecipitation from over-expressing cell lysates. Furthermore, the interaction between U1-70k and NSSR 1 is shown by the recombinant proteins (Shin et al. 2004). According to the interaction analysis by the recombinants, the dephosphorylation form of NSSR 1 binds to U1-70k much stronger than the phosphorylated form does (Shin et al. 2004). Therefore, NSSR 1 might be phosphorylated in the yeasts. One of the functions of dephosphorylated NSSR 1 is the repression of splicing (Shin et al. 2004). However, the NSSR 1 C-terminal stimulated the inclusion of the modified exon. Since the dephosphorylated form cannot be detected in the normally growing cells (Shin & Manley 2002; Shin et al. 2004), the inclusion activity of the NSSR 1 C-terminal is likely to be caused by the phosphorylated form.

Since the consensus sequence of the RRM of NSSRs is unknown, it is difficult to show the exon inclusion activity by direct binding to the ESE on pre-mRNAs. Therefore, the RRM of NSSRs was switched to MS2CP and then, the stem loops of the MS2 RNA were inserted into the middle of the exon N of the clathrin light chain B minigene. In the case of the original minigene, the neural-specific form was not detected in NIH3T3 cells. However, both the neural and non-neural forms of the mRNA were detected when MS2 stem-loops were inserted into the exon N. As a result, both inclusion and exclusion activities were detected (Fig. 2D). The replacement of the RRM of NSSRs by MS2CP revealed the exon inclusion and exclusion activities of the NSSR 1 and 2 C-terminals in all cell lines examined, respectively (Figs 2D and 5). Although NSSR 1 has short SR dipeptide repeats, NSSR 1 is thought to act as an SR protein. Correspondingly, a fusion protein consisting of MS2CP and the RS domain of U1-70k has the efficient splicing activity of dsx(70)M1 despite a rather low content of SR (Philipps et al. 2003). The sequences other than the SR dipeptides in the RS domain of U1-70k may be capable of functioning as a splicing activation domain.

Since the switching experiment of RRMs by MS2CP was successful, we speculate that an ESE-dependent splicing regulation can be detected by this system even though the consensus sequences of the RNA binding domains are unknown. Recently, contributions of RNA binding proteins that do not contain any RS domains to alternative splicing have been reported. Therefore, this system can be applied to those proteins.

The deletion analyses revealed that the amino acids 143–234 of NSSR 1 contribute to the exon inclusion. Although SR rich sequences are thought to participate in their protein-protein interaction among them, the third C-terminal SR rich region of NSSR 1 (the amino acids 222–262) was not necessary for the exon inclusion. Since the amino acids 143–234 of NSSR 1 consist of the SR rich region and an additional sequence, this additional sequence seems to play an important role in the inclusion of exons. For example, because SR repeats are too simple to recognize their specific partners, the additional sequence may engage in the specific recognition of NSSR 1 binding proteins which contain SR rich sequences. One of the targets for the amino acids 143–234 of NSSR 1 may be U1-70k, but we do not have evidence of a direct binding between NSSR 1 and U1-70k. Since the association between U1-70k and NSSR 1 was observed in yeasts and COS cells, it is possible that NSSR 1 brings U1-70k in the neighborhood of the exons, which are recognized by NSSRs’ RRM, and activates the 5' splicing sites. On the other hand, the C-terminal sequence of NSSR 2 contains the SR rich sequence and NSSR 2 associated with U1-70k. However, the C-terminal sequence of NSSR 2 inhibited the inclusion of the modified exon. Therefore, NSSR 2 may block the access of U1-70k to the 5' splicing site by recognizing the exons.

In this study, the bacterial recombinant NSSRs’ RRM bound mRNAs specifically. We also showed the association of NSSR with U1-70k. In addition, NSSRs localize in nuclear speckles (Komatsu et al. 1999). Taken together, NSSRs potentially bind exons of pre-mRNAs and participate in the regulation of alternative splicing.

Finally, the in situ hybridization indicated that NSSRs were expressed in the brain during development, which is consistent with the expression of NSSR 1 during neuronal differentiating P19 cells (Komatsu et al. 1999). In the ventricular zone, there are cells that are differentiating into neurons and glial cells. The dental bulb of tooth germs and salivary gland also express NSSRs, and cells in these regions are in the middle of differention at E15.5. The results of the Northern blot analysis of human mature tissues showed higher expression of NSSR 1 and 2 in skeletal muscle than brain (Clinton et al. 2002). Since many types of cells are under differentiation in fetal bodies, NSSRs may contribute to the differentiation of the cells rather than the maintenance of these cells. In contrast, skeletal muscles seem to utilize NSSRs for its maintenance.

NSSR 1 was cloned as an SR protein that was expressed in neural differentiated P19 cells (Komatsu et al. 1999). However, we found that many tissues express NSSR 1 during their differentiation. Thus, we speculate that NSSR 1 regulates transcriptional regulators shared in NSSR 1 expressing cells to control their differentiation.

In this paper, we demonstrated the association of NSSR with U1-70k. In addition, we showed roles which NSSRs play in the regulation of neural alternative splicing by binding to pre-mRNAs directly. Interestingly, the expression of NSSRs in proliferating undifferentiated cells in nervous systems imply crucial roles of NSSRs during the development. The C-terminal of NSSR 2 and middle RS region of NSSR 1 can participate in the exclusion and inclusion of exons, respectively. These results suggested that NSSRs are SR proteins despite their atypical RS domain and regulate alternative splicing during neuronal differentiation. Furthermore, we demonstrated that the domain swapping analysis would be useful to analyze the exon inclusion and exclusion activity of splicing associated proteins. In fact, other than SR proteins, many RNA binding proteins have been found as splicing regulators.


   Experimental procedures
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
Supplementary material
 References
 
Plasmid construction

High fidelity DNA polymerase (Pyrobest DNA polymerase, TAKARA) was used to construct plasmids. The primers used in this study are listed in Table 2. In order to construct pMS2CNS1, 2 and their deletion mutants, pCMV/myc/nuc and pHybLex/Zeo-MS2 were purchased from Invitrogen. DNA fragments of the NSSR C-terminals and MS2CP were obtained by PCR with primers shown in Table 1. The fragments for MS2CP were inserted between the NotI and PstI sites of pCMV/myc/nuc followed by linearization by PCR with the primers, MycnucF and MycnucR. The linearized plasmid was digested with NotI and the C-terminal of NSSRs was inserted. The clathrin light chain B minigene expression vector, pJS74, which was a generous gift from Dr Stamm (Stamm et al. 1992), was amplified with the primers, CLR and CLF, to linearize the plasmid. The MS2 stem loops, which were obtained by digesting pRH3' with EcoRI and SmaI, were ligated with the linearized pJS74.


View this table:
[in this window]
[in a new window]
  		Table 2 Oligonucleotides used in this study
 
For the yeast two-hybrid assays, NSSRs and ASF were inserted into pAS2.1 in frame between the EcoRI and SalI sites in their full lengths. U1-70k was inserted into pACT2 in frame at the SmaI site.

In order to produce the NSSRs’ RMM recombinant, the DNA fragments that code the RRM were amplified by high fidelity PCR with the primers, RRMF and RRMR. The fragments were subcloned at the EcoRI site of pYesTrp3. The obtained plasmids were digested with HindIII and XhoI to insert into pET-34b(+) that carried a cellulose binding domain tag (CBD tag, Novagen).

RT-PCR

Total RNAs were isolated from cells using TRIzol reagent (Invitrogen). One microgram of total RNA was used for each reaction. Complementary DNAs were synthesized from oligo dT primer using ImProm-ll Reverse Transcription System (Promega). The mRNAs derived from the clathrin light chain B minigene were amplified by PCR with the specific primers, SV40A and SS031, as described by Stamm et al. (1992).

Yeast two-hybrid assay

The GAL4 DNA binding domain and GAL4 activation domain fusions were constructed with pAS2.1 and pACT2 (Clontech), respectively. The plasmids derived from pAS2.1 and pACT2 were introduced into AH109 and Y189 by the Yeast Transformation System (Clontech), respectively. The obtained clones were mated and grown on SD plates lacking tryptophan, leucine, adenine and histidine to test the interaction between NSSRs and U1-70k. In this two-hybrid system each reporter gene in yeasts was driven by different promoter elements, UAS-TATA. Furthermore, the reporter gene for Ade– was tighter than that for His–. The Yeast Protocol Handbook (Clontech) was used as a guide to handle yeasts.

Mammalian cell culture

COS7 and NIH3T3 cells were grown in DMEM supplemented with 10% heat inactivated FCS. Neuro2a cells were grown in DMEM supplemented with 10% heat inactivated FCS and non-essential amino acids solution (Sigma). The transfection was performed with Lipofectamine and plus regent (Invitrogen), as described in the product manual.

Immunoprecipitation and immunoblot analysis

Cells were suspended in PBS containing 0.05% Triton X-100 RS, 10 µg/mL RNase A and protein inhibitor cocktails for mammalian cells (Nacalai tesque), and then frozen once to lyze. Myc-tagged U1-70K was collected with anti-myc monoclonal antibody agarose affinity gel (clone 9E10, Santa Cruz). The agarose gel was washed four times by the lysis buffer.

SDS-PAGE was performed as described by Laemmli, followed by blotting on to the PVDF membranes. Anti-FLAG M2 monoclonal antibody (Sigma), anti-myc monoclonal antibody (Cell Signaling Technology) and anti-CBD Tag rabbit polyclonal antibody (Novagen) were used as primary antibodies. HRP linked anti-mouse IgG and anti-rabbit IgG (Chemicon) were used as a secondary antibodies. FLAG, myc and CBD tagged proteins were detected by ECL (Amersham Biosciences).

Production and extraction of the NSSRs’ RRM recombinant protein

The NSSRs’ RRM recombinant was produced in 100 mL culture of E. coli, Rosetta(DE3)pLysS, by induction with 1 mM IPTG in the LB medium for 3 hours at 37 °C. The bacterial pellet was frozen once and then lyzed with 50 mM Tris-HCl buffer (pH 8.0) containing 500 mM NaCl, 0.1% Triton X-100 and 100 µM PMSF. The lysate was dialyzed against 50 mM Tris-HCl buffer (pH 8.0) containing 500 mM NaCl and 100 µM PMSF by Spectra/Por CE (MWCO 300 000, SPECTRUM). The external solution was concentrated by ultrafiltration using Amicon Ultra-4 (MWCO 10 000, MILLIPOR).

Messenger RNA binding assay

Total RNA was extracted from seven-day-old murine brains by TRIzol and then, the mRNA was purified with the mTRAP total kit (ACTIVE MOTIF). The extracted NSSRs’ RRM recombinant protein was absorbed on 30 mg of cellulose resin, CBinD 100 (Novagen). After three washes with 10 mM Tris-HCl (pH 8.0) containing 500 mM NaCl, 1 mM MgCl2 and 0.2 U/µL SUPER-ase-In (Ambion), the resin was resuspeneded in 300 µL of the same buffer. Twenty micrograms of yeast tRNA were added to the suspension followed by an incubation of 5 min. Then, 20 µg of the murine brain mRNA were added. After an incubation of 10 min, the resin was washed four times with the same buffer. Next, the resin was resuspended with 100 µL of 0.5% SDS and then, 500 µL of TRIzol was added on to the resin to purify the mRNA on the resin. The obtained RNA was analyzed by BIOANALYZER 2100 with RNA 6000 pico reagent kit (Agilent Technologies).

In situ hybridizaion

The NSSR 1 and 2-specific probes for the in situ hybridization were constructed using the fragments comprised of the bp 831–2669 of NSSR 1 cDNA and the bp 681–2227 of NSSR 2 cDNA, respectively. The common probe recognizing NSSR 1 and 2 was produced from the bp 1–229 fragment of NSSR 1 cDNA (Fig. 4A). ICR mice were purchased from Charles River, Japan. DIG-labeled cRNAs were synthesized by T7 RNA polymerase. The in situ hybridization on frozen cryosections was performed using methods previously described (Takahashi & Osumi 2002).


   Supplementary material
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 Supplementary material
References
 
The following material is available from:

http://www.blackwellpublishing.com/products/journals/suppmat/GTC/GTC855/GTC855sm.htm

Supplementary Figure S1


   Acknowledgements
 
We thank Dr Stefan Stamm for the generous gift of his minigene constructs and Petra Ruick for grammatical corrections. This work was supported in part by a Research Grant-in-Aid for Scientific Research on Priority Areas from the Ministry of Education, Culture, Sports, Science and Technology, a grant from the National Center of Neurology and Psychiatry of the Ministry of Health, Labour and Welfare, Japan.


   Footnotes
 
Communicated by: Yoshikazu Nakamura

* Correspondence: E-mail: tukahara{at}jaist.ac.jp


   References
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 Supplementary material
 References
 
Cao, W. & Garcia-Blanco, M.A. (1998) A serine/arginine-rich domain in the human U1-70k protein is necessary and sufficient for ASF/SF2 binding. J. Biol. Chem. 273, 20629–20635.[Abstract/Free Full Text]

Clinton, J.M., Chansky, H.A., Odell, D.D., Zielinska-Kwiatkowaska, A., Hickstein, D.D. & Yang, L. (2002) Characterization and expression of the human gene encoding two translocation liposarcoma protein-associated serine-arginine (TASR) proteins. Gene 284, 141–147.[CrossRef][Medline]

Eide, F.F., Vining, E.R., Eide, B.L., Zang, K., Wang, X.Y. & Reichardt, L.F. (1996) Naturally occurring truncated trkB receptors have dominant inhibitory effects on brain-derived neurotrophic factor signaling. J. Neurosci. 16, 3123–3129.[Abstract/Free Full Text]

Ge, H. & Manley, J.L. (1990) A protein factor, ASF, controls cell-specific alternative splicing of SV40 early pre-mRNA in vitro. Cell 62, 25–34.[CrossRef][Medline]

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

Heidmann, D.E., Szot, P., Kohen, R. & Hamblin, M.W. (1998) Function and distribution of three rat 5-hydroxytryptamine 7 (5-HT7) receptor isoforms produced by alternative splicing. Neuropharmacology 37, 1621–1632.[CrossRef][Medline]

Jensen, K.B., Dredge, B.K., Stefani, G., et al. (2000) Nova-1 regulates neuron-specific alternative splicing and is essential for neuronal viability. Neuron 25, 359–371.[CrossRef][Medline]

Johnson, M.J., Castle, J., Garrett-Engele, P., et al. (2003) Genome-wide survey of human alternative pre-mRNA splicing with exon junction microarrays. Science 302, 2141–2144.[Abstract/Free Full Text]

Khan, Z.U., Mrzljak, L., Gutierrez, A., de la Calle, A. & Goldman-Rakic, P.S. (1998) Prominence of the dopamine D2 short isoform in dopaminergic pathways. Proc. Natl. Acad. Sci. U S A 95, 7731–7736.[Abstract/Free Full Text]

Komatsu, M., Kominami, E., Arahata, K. & Tsukahara, T. (1999) Cloning and characterization of two neural-salient serine/arginine-rich (NSSR) proteins involved in the regulation of alternative splicing in neurones. Genes Cells 4, 593–606.[Abstract]

Kuppenbender, K.D., Albers, D.S., Iadarola, M.J., Landwehrmeyer, G.B. & Standaert, D.G. (1999) Localization of alternatively spliced NMDAR1 glutamate receptor isoforms in rat striatal neurons. J. Comp. Neurol. 415, 204–217.[CrossRef][Medline]

Lambolez, B., Ropert, N., Perrais, D., Rossier, J. & Hestrin, S. (1996) Correlation between kinetics and RNA splicing of alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid receptors in neocortical neurons. Proc. Natl. Acad. Sci. U S A 93, 1797–1802.[Abstract/Free Full Text]

Lopez, A.J. (1995) Developmental role of transcription factor isoforms generated by alternative splicing. Dev. Biol. 172, 396–411.[CrossRef][Medline]

Mion, S., Corti, C., Neki, A., et al. (2001) Bidirectional regulation of neurite elaboration by alternatively spliced metabotropic glutamate receptor 5 (mGluR5) isoforms. Mol. Cell. Neurosci. 17, 957–972.[CrossRef][Medline]

Peabody, D.S. & Ely, K.R. (1992) Control of translational repression by protein–protein interactions. Nucleic Acids Res. 20, 1649–1655.[Abstract/Free Full Text]

Philipps, D., Celotto, A.M., Wang, Q., Tarng, R.S. & Graveley, B.R. (2003) Arginine/serine repeats are sufficient to constitute a splicing activation domain. Nucleic Acids Res. 31, 6502–6508.[Abstract/Free Full Text]

Polydorides, A.D., Okano, H.J., Yang, Y.Y., Stefani, G. & Darnell, R.B. (2000) A brain-enriched polypyrimidine tract-binding protein antagonizes the ability of Nova to regulate neuron-specific alternative splicing. Proc. Natleic Acad. Sci. U S A 97, 6350–6355.

Shen, H., Kan, J.L.C. & Green, M.R. (2004) Arginine-Serine-rich domains bound at splicing enhancers contact the branchpoint to promote prespliceosome assembly. Mol. Cell 13, 367–376.[CrossRef][Medline]

Shin, C., Feng, Y. & Manley, J.L. (2004) Dephosphorlylated SRp38 acts as a splicing repressor in response to heat shock. Nature 427, 553–558.[CrossRef][Medline]

Shin, C. & Manley, J.L. (2002) The SR protein SRp38 represses splicing in M phase cells. Cell 111, 407–417.[CrossRef][Medline]

Stamm, S., Casper, D., Dinsmore, J., Kaufmann, C.A., Brosius, J. & Helfman, D.M. (1992) The clathrin light chain B: gene structure and neuron-specific splicing. Nucleic Acids Res. 20, 5097–5103.[Abstract/Free Full Text]

Takahashi, M. & Osumi, N. (2002) Pax6 regulates specification of ventral neurone subtypes in the hindbrain by establishing progenitor domains. Development 129, 1327.

Valenzuela, D.M., Maisonpierre, P.C., Glass, D.J., et al. (1993) Alternative forms of rat TrkC with different functional capabilities. Neuron 10, 963–974.[CrossRef][Medline]

Wu, J.Y. & Maniatis, T. (1993) A striking organization of a large family of human neural cadherin-like cell adhesion genes. Cell 75, 1061–1070.[CrossRef][Medline]

Yang, L., Embree, L.J. & Hickstein, D.D. (2000) TLS-ERG leukemia fusion protein inhibits RNA splicing mediated by serine-arginine proteins. Mol. Cell. Biol. 20, 3345–3354.[Abstract/Free Full Text]

Yang, L., Embree, L.J., Tsai, S. & Hickstein, D.D. (1998) Oncoprotein TLS interacts with serine-arginine proteins involved in RNA splicing. J. Biol. Chem. 273, 27761–27764.[Abstract/Free Full Text]

Zhou, Z., Licklider, L.J., Gygi, S.P. & Reed, R. (2002) Comprehensive proteomic analysis of the human spliceosome. Nature 419, 182–185.[CrossRef][Medline]

Received: 11 January 2005
Accepted: 28 February 2005





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
Right arrow Citation Map
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
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fushimi, K.
Right arrow Articles by Tsukahara, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fushimi, K.
Right arrow Articles by Tsukahara, T.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS