FEBS Letters 588 (2014) 942–947

journal homepage: www.FEBSLetters.org

RBM10 regulates alternative splicing Akira Inoue a,b,⇑, Naoki Yamamoto c, Masatsugu Kimura d, Koji Nishio e, Hideo Yamane b, Koichi Nakajima a a

Department of Immunology, Osaka City University Graduate School of Medicine, Japan Department of Otolaryngology, Osaka City University Graduate School of Medicine, Japan c Department of Psychiatry and Behavioral Sciences, Tokyo Medical and Dental University Graduate School, Japan d Department of Radioisotope Center, Osaka City University Graduate School of Medicine, Japan e Department of Anatomy and Cell Biology, Graduate School of Medicine, Nagoya University, Japan b

a r t i c l e

i n f o

Article history: Received 22 September 2013 Revised 2 December 2013 Accepted 23 January 2014 Available online 11 February 2014 Edited by Ulrike Kutay Keywords: S1-1 RBM10 RBM5 Alternative splicing Fas, Bcl-x

a b s t r a c t RBM10, originally called S1-1, is a nuclear RNA-binding protein with domains characteristic of RNA processing proteins. It has been reported that RBM10 constitutes spliceosome complexes and that RBM5, a close homologue of RBM10, regulates alternative splicing of apoptosis-related genes, Fas and cFLIP. In this study, we examined whether RBM10 has a regulatory function in splicing similar to RBM5, and determined that it indeed regulates alternative splicing of Fas and Bcl-x genes. RBM10 promotes exon skipping of Fas pre-mRNA as well as selection of an internal 50 -splice site in Bcl-x premRNA. We propose a consensus RBM10-binding sequence at 50 -splice sites of target exons and a mechanistic model of RBM10 action in the alternative splicing. Ó 2014 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

1. Introduction RBM10, originally called S1-1, is an RNA-binding protein [1]. Loss of its function results in various deformities such as cleft palate and malformation of the heart (TARP syndrome) [2,3] and diseases such as lung adenocarcinoma [4]. S1-1/RBM10 exists in nuclear components called S1-1 nuclear bodies and S1-1 granules. When transcription is globally halted, S1-1 nuclear bodies dynamically enlarge with a concomitant diminution of hundreds of S1-1 granules. The S1-1 granules localize on perichromatin fibrils, which are the sites of transcription and splicing, and the new states resulting from these changes reversibly return to the initial states upon recovery of transcriptional activity [5]. Thus, S1-1 is likely associated with the gene expression process. RBM10 preferentially binds to G- and U-rich RNA sequences [1], and contains the domains that suggest its involvement in RNA processing [6]: two RNA-recognition motifs (RRMs) [1], a C4 Zn finger that preferentially binds to 50 -splice site-like single-stranded RNAs ⇑ Corresponding author at: Department of Immunology, Osaka City University Graduate School of Medicine, 1-4-3 Asahi-machi, Abeno-ku, Osaka 545-8585, Japan. Fax: +81 666460515. E-mail address: [email protected] (A. Inoue).

[7], and a G-patch motif found in some eukaryotic RNA-processing proteins [8]. RBM5, a close homologue of RBM10, has been shown to regulate alternative splicing of pre-mRNAs of apoptosis-related genes, FAS, c-FLIP [9], and caspase 2 [10]. To demonstrate its splicing regulatory function by RNA interference (RNAi), it was necessary to knock down not only RBM5 but also RBM10 and RBM6 at the same time, thereby suggesting that RBM5 and RBM10 have overlapping functions [9]. Furthermore, RBM10 has been shown in splicing complexes [11] and in spliceosome A [12,13] and B complexes [13,14], constituting a core protein of these complexes [15]. These lines of evidence all support the notion that RBM10 participates in alternative splicing. In this study, we examined this long-standing assumption and verified that RBM10 indeed regulates alternative splicing. We also discuss functional differences between RBM5 and 10. 2. Materials and methods 2.1. Cells and cell culture Human cell lines (HeLa cells derived from cervical carcinoma and HLE cells from hepatoma) were cultured in Dulbecco’s Modified Eagle’s Medium (Nissui) containing 100 lg/mL streptomycin, 100 IU/mL penicillin, and 5% fetal calf serum (Gibco BRL).

http://dx.doi.org/10.1016/j.febslet.2014.01.052 0014-5793/Ó 2014 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

943

A. Inoue et al. / FEBS Letters 588 (2014) 942–947

2.2. Establishment of HeLa RBM10 Tet-On cells

72 °C for 7 min using AmpliTaq Gold PCR Master Mix (Applied Biosystems) with template cDNA equivalent to 15 ng of total RNA and a high concentration (0.75 mM each) of primers (Supplementary material). The PCR conditions were semi-quantitative, and no more than 2–5% of input primers were consumed. Each primer set for caspase 2, 3, and 9 or Bcl-x simultaneously amplified two alternatively spliced isoforms. For Fas or FLIP, two primer sets were used to amplify respective isoforms, and reaction mixtures were combined before gel electrophoresis. The PCR products, resolved on 12.5% polyacrylamide gels (e-PAGEL; Atto Corp.), were stained with SYBR Green I (Lonza), and their band intensities (BIs) were measured in arbitrary units under an LAS-3000 luminescent image analyzer (Fujifilm). BIs were compared after adjusting to number of molecules. For example, in Fas splicing, the BI of exon 6-skipped DNA (PCR product size: 85 bp) was compared with that of exon 6-included DNA (118 bp) after multiplication by 118/85.

cDNA of the human RBM10 larger isoform, with an EcoRV linker immediately upstream to the 50 -initiation codon and an XbaI linker following the 30 -stop codon, was prepared by PCR amplification using primers indicated in the Supplementary material. The cDNA and vector pTRE-Tight (BD Sciences) were digested with EcoRV and XbaI and ligated to obtain pRBM10/TRE-Tight. After propagation, purification, and sequence confirmation, the plasmid was introduced into HeLa Tet-On cells (BD Sciences Clontech) using Easy Transgater (Apharmas). The culture of the transfected cells and development of a stable cell line (HeLa RBM10 Tet-On h08) were performed, as suggested by the supplier, in the presence of G418 and hygromycin for selection and maintenance. 2.3. Knockdown and overproduction of RBM10 HLE and HeLa cells were transfected with 10 nM specific human RBM10-siRNA5 or with control AllStar siRNA using HiPerFect transfection reagent (all from Qiagen) and incubated for 4 d. RBM10 expression was induced in HeLa RBM10 Tet-On cells by incubation with 2 lg/mL doxycycline (Dox) for 24 h.

2.6. Immunoblotting Western blotting was performed as previously described [5], where proteins (2.5 lg) were resolved on an 8.8% or 10% gel (e-PARGEL; Atto) by SDS–polyacrylamide gel electrophoresis (SDS–PAGE).

2.4. Isolation of RNA and reverse transcription

3. Results

Total RNA was isolated with Trizol (Invitrogen) as indicated by the supplier. Reverse transcription of 1.5 lg of total RNA was carried out at 37 °C for 17 min using a PrimeScript RT reagent kit (Takara) and mixed primers of oligo dT (0.25 lM) and random oligonucleotides (0.5 lM).

3.1. RBM10 regulates alternative splicing of Fas transcript Fas is a widely expressed cell surface receptor. The binding of Fas to its ligand on an adjacent cell initiates a cascade leading to cell death. The Fas pre-mRNA can be alternatively spliced to produce the membrane-bound, proapoptotic Fas receptor and a soluble antiapoptotic isoform with exon 6 skipped [16,17]. RBM5 regulates this alternative splicing and increases exon 6 skipping.

2.5. PCR PCR was carried out at 95 °C for 15 s, followed by 27–30 cycles of 95 °C for 15 s, 58.5 °C for 15 s, and 72 °C for 20 s, and finally at

a

RBM10 protein HeLa HLE HeLa Tet-on (aa) 930 852

1 2 Si

C

Si D(–) D(+)

1 : 0.05 1 : 0.03 1 : 1.7 (Relative abundance of RBM10 protein)

d

HLE

HeLa HeLa Tet-on *

C

exon 5

exon 7

exon 6

(1) Primers to detct exon 6 skipping

c

HLE

HeLa

HeLa Tet-on (bp) 118

(2) (1)

85

1.0

0.5

*

Primers to detct exon 6 inclusion

(2)

*

b

Exon 6 skipped / included, (1)/(2)

1.5

0 C Si

C

Si

C

C Si

D(–)D(+)

Si D(–) D(+)

Fig. 1. RBM10 regulates alternative splicing of Fas transcript. HLE and HeLa cells were transfected with siRNA and incubated for 4 d. HeLa Tet-On cells were incubated with Dox for 24 h. (a) RBM10 was analysed by immunoblotting of total proteins on an 8.8% or 10% polyacrylamide gel. The upper and lower bands are RBM10 isoforms 1 and 2, and the relative abundance of RBM10 in each group is indicated at the bottom. C, control AllStar siRNA; Si, RBM10-specific siRNA; and D, Dox. (b) cDNAs were synthesized from total RNAs, and analysed by PCR for changes in the alternative splicing of Fas pre-mRNA using the primer sets [9] indicated by arrows. The primer set (1) is for the analysis of exon 6 skipping and (2) for inclusion, the reactions producing antiapoptotic and proapoptoic isoforms, respectively. Dotted parts spanning two exons are not included in the primers. (c) PCR products were analysed by PAGE (12.5% gels). The product sizes are indicated on the right. (1) Exon 6 skipping, (2) exon 6 inclusion. (d) BIs in (c) were adjusted to number of molecules, and changes in BI ratio, (1)/(2), are presented as a column graph. Mean with S.E.M. of quadruplicate determinations from two independent experiments (Table 1a) is indicated. ⁄P < 0.05.

944

A. Inoue et al. / FEBS Letters 588 (2014) 942–947

To observe its reduced activity by RNAi in HeLa cells, simultaneous depletion of RBM5, RBM6, and RBM10 was necessary, which suggested that the corresponding proteins have overlapping functions [9]. We examined this assumption about RBM10 function, and found that depletion of RBM10 alone was enough to decrease exon 6 skipping in HeLa cells as well as HLE cells. When approximately 95% of RBM10 was silenced (Fig. 1a, Fig. S1a), exon 6 skipping was diminished by 68% in HLE cells and by 75% in HeLa cells (Fig. 1c and d). Conversely, a 1.7-fold overexpression of RBM10 in HeLa Tet-on cells (Fig. 1a, Fig. S1b) resulted in a 3.7-fold increase in the skipping (Fig. 1c and d). The results were highly reproducible (Table 1a) and indicate that RBM10 indeed regulates splicing and promotes exon 6 skipping in Fas transcripts. The similar results with two different cell lines, HeLa and HLE, suggest that the observed RBM10 function is general rather than specific to a particular cell type.

procaspase 8, while the exon 7-skipped isoform, c-FLIP(L), is less inhibitory or may act as an activator in the Fas pathway [9,18]. Here again, in contrast to the simultaneous depletions of RBM5, 6, and 10 [9], depletion of RBM10 alone decreased exon 7 skipping in HeLa cells by approximately 2%. Conversely, overproduction of RBM10 enhanced the skipping by 3% (Table 1b). However, these changes were small and not observed in HLE cells. Furthermore, as much as 89–92% of the splicing was the inclusion event irrespective of the different levels of RBM10 (Table 1b). Hence, we concluded RBM10 was not important for this reaction. 3.4. Caspase 2

We further examined whether RBM10 participates in alternative splicing of other apoptosis-related genes that produce isoforms with opposite functions.

Caspase 2 pre-mRNA produces proapoptotic caspase 2L and antiapoptotic caspase 2S. The latter mRNA includes exon 9 that harbors a premature termination codon, producing a short isoform with no caspase activity [19,20]. RBM5 has been shown to enhance exon 9 skipping in human embryonic kidney (HEK) cells and HeLa nuclear extract to produce caspase 2L [10]. Accordingly, we examined RBM10 for similar activity. However, neither knockdown nor overexpression of RBM10 produced any significant effect on exon 9 skipping of caspase 2 pre-mRNA (Table 1b).

3.3. c-FLIP

3.5. Caspase 3

The FLIP gene produces isoforms c-FLIP(S) and c-FLIP(L). The c-FLIP(S) isoform arises from exon 7 inclusion and is a strong inhibitor of the Fas pathway, preventing proteolytic activation of

Exon 6 of caspase 3 pre-mRNA undergoes alternative splicing. Skipping this exon results in a truncated antiapoptotic isoform, caspase 3s [21]. In the present study, significant changes were

3.2. RBM10 in alternative splicing of other apoptosis-related genes

Table 1 RBM10 effects on alternative splicing of some apoptosis-related genes. Changes in the alternative splicing were analyzed by PCR followed by PAGE as in the Fig. 1 legend. The primers used in PCR are indicated in Supplementary material. Band intensities (BIs) of amplified DNAs corresponding to exon skipping (1) and inclusion (2) were measured and adjusted to number of molecules. (a) Fas. Values are average with S.E.M. of quadruplicate determinations from two independent experiments (n = 4). (b) Other apoptosis-related genes. Values are average of duplicate determinations with deviation. Bcl-x was analyzed for selection of the internal (1) and external (2) 50 -splice site. (a) Alternative splicing of Fas exon 6 Knockdown with siRNA HLE

b

Control SiRNA or Doxc

Overprod. with Dox HeLa Tet-On

HeLa

Skip (1)d

inclusion (2)d

BI ratio (1)/(2)e

53.3 27.2

46.5 72.8

1.15 0.37 (0.32)f

S.E.M.

Skip (1)

inclusion (2)

BI ratio (1)/(2)

0.123 0.103

32.3 10.6

67.7 89.4

0.48 0.12 (0.25)f

(b) Alternative splicing of other apoptosis-related genes Knockdown with siRNA HLE Genes c-FLIP

Exona

Skip (1)d

inclusion (2)d

7

(L)g 10.4 10.8 2Lg 95.7 96.4 3sg 22.5 25.4 9bg 98.4 98.5 Sg 72.2 69.3

(S)g 89.6 ± 1.3 89.2 ± 2.0 2Sg 4.3 ± 0.1 4.6 ± 1.0 3g 77.5 ± 1.5 74.7 ± 1.6 9ag 1.6 ± 0.7 1.5 ± 0.4 Lg 27.8 ± 2.2 30.7 ± 0.8

Controlb siRNA or Doxc Casp-2

9 Control siRNA or Dox

Casp-3

6 Control siRNA or Dox

Casp-9

3–6 cassette Control siRNA or Dox

Bcl-x

2 Control siRNA or Dox

a b c d e f g

S.E.M.

Skip (1)

inclusion (2)

BI ratio (1)/(2)

0.106 0.036

27.6 58.7

72.4 41.3

0.38 1.42 (3.74)f

2.63 2.26 (0.86)f

0.078 0.131

Overprod. with Dox HeLa Tet-On

HeLa BI ratio (1)/(2)e

S.E.M.

Skip (1)

inclusion (2)

(L) 10.5 8.7 2L 92.6 93.7 3s 7.5 8.3 9b 92.3 92.5 S 66.3 58.8

(S) 89.5 ± 1.0 91.3 ± 1.0 2S 7.4 ± 0.2 6.3 ± 0.1 3 92.5 ± 0.8 91.8 ± 1.2 9a 7.7 ± 0.5 7.5 ± 0.1 L 33.7 ± 1.1 41.5 ± 0.8

BI ratio (1)/(2)

1.97 1.41 (0.72)f

Skip (1)

inclusion (2)

(L) 8.2 11.2 2L 95.5 96.0 3s 23.4 24.9 9b 95.6 95.7 S 63.7 72.5

(S) 91.9 ± 0.7 88.8 ± 0.6 2S 4.5 ± 0.2 4.0 ± 0.1 3 76.7 ± 0.2 75.2 ± 0.2 9a 4.5 ± 1.4 4.3 ± 0.4 L 36.3 ± 1.6 27.5 ± 1.3

BI ratio (1)/(2)

1.75 2.64 (1.51)f

Exon: target exons in (b) that undergo alternative splicing. Control: cells were either transfected with control siRNA or incubated in the absence of Dox. siRNA or Dox: cells were either transfected with specific RBM10-siRNA or incubated with Dox. Percentages of two alternative splicing events, calculated from BI values that were adjusted to number of molecules. The sum of the values (1) and (2) is 100%. (1)/(2) ratios. The (1)/(2) ratios of ‘siRNA or Dox’ were divided by those of corresponding ‘Control’. The values obtained for Fas and Bcl-x are indicated in parentheses. Name of isoform.

A. Inoue et al. / FEBS Letters 588 (2014) 942–947

not found in the alternative splicing upon knockdown or overproduction of RBM10 (Table 1b). 4. Effects of RBM10 on apoptosis-related genes undergoing other types of alternative splicing 4.1. Caspase 9 Two splice isoforms are derived from the caspase 9 gene, proapoptotic caspase 9a and truncated antiapoptotic 9b (also called caspase-9S), by inclusion or exclusion of a cassette of exons 3–6 [22]. No effect was observed by knockdown or overproduction of RBM10 in this alternative splicing (Table 1b). 4.2. Bcl-x The Bcl-x pre-mRNA is alternatively spliced at two competing 50 -splice sites present in exon 2, which is the first coding exon of the Bcl-x transcript, producing shorter proapoptotic Bcl-x(S) and longer antiapoptotic Bcl-x(L) (Fig. 2). Expression of the Bcl-x(S) isoform, which antagonizes Bcl-x(L) as well as Bcl-2, is sufficient to induce apoptotic cell death in a wide range of cell types [23,24]. RBM10 knockdown reduced Bcl-x(S) production in both HLE cells and HeLa cells, while Bcl-x(S) was conversely increased by overproduction of RBM10 in HeLa Tet-on cells, although the effects were not large (Table 1b, Fig. 2b and c). 5. Discussion In the present study, we proved the long-standing assumption that RBM10 regulates alternative splicing; RBM10 significantly

a

(1)

Bcl-x(S)

exon 2 Primers: Fw

exon 3 Rev

(2)

b

HLE

HeLa

Bcl-x(L) HeLa Tet-on

(2)

625

(1)

435

C

c

3.0

Bcl-x (S) / Bcl-x (L), (1)/(2)

(bp)

Si

C

Si D(–)D(+)

HLE

HeLa

HeLa Tet-on

C Si

C Si

D(–)D(+)

2.0

1.0

0

0

Fig. 2. RBM10 regulates 5 -splice site selection in Bcl-x transcripts. (a) Bcl-x produces proapoptotic Bcl-x(S) (1) and anti-apoptotic Bcl-x(L) (2) due to the presence of two competing 50 -splice sites in exon 2. The primers, Fw and Rev, simultaneously amplify two splicing events. (b) Bcl-x alternative splicing was examined as described in the legend to Fig. 1: subsequent to PCR, amplified DNAs were analyzed by PAGE. The product sizes are indicated on the right. (c) BIs of Bclx(S) (1) and Bcl-x(L) (2) in (b) were adjusted to number of molecules, and the BI ratios, (1)/(2), are presented as a column graph. Deviation from an average of duplicate determination is indicated.

945

enhanced the exon 6 skipping of Fas gene and, to a lesser extent, the selection of an alternative 50 -splice site in exon 2 of the Bcl-x gene. RBM5 and RBM10 are composed of more than 800 amino acid (aa) residues with 50% sequence identity. The present study showed that, in contrast to the depletion of RMB5 in HeLa cells [9], the depletion of RBM10 alone was sufficient to decrease exon 6 skipping of Fas pre-mRNA. Conversely, while RBM 5 enhances exon 9 skipping of caspase 2 pre-mRNA [10], RBM10 had no such activity. In fact, the preference of RBM10 and 5 for RNA is different: RBM10 binds to RNA homopolymers preferentially in the order poly G = poly U > poly C  poly A [1], while RBM5 does so in the order poly G > poly C > poly A P poly U [25]. Accordingly, we expect that the activities and substrate specificities of RBM10 and RBM5 in alternative splicing are considerably different. Elucidation of the target genes of RBM10 and RBM5 will help understand their roles and significance in cellular function as well as in various diseases that arise from loss of their functions. RBM10 interacts with several proteins, including the branchpoint-binding protein SF1, another spliceosome A complex protein SF4, as well as the U2-related proteins SR140 and DEAH helicase hPRP43 [15]. Interestingly, SPF45, a regulator of alternative splicing [26], interacts with the same set of proteins [15]. The common protein interaction networks support the present notion that RBM10 is a regulator of alternative splicing. From our results and previous reports, a possible mechanistic model of RBM10 action can be envisaged (Fig. 3). RBM10 has a RanBP2-type Zn finger at aa 212–243 (aa135–166 in 852 residue isoform), which has been shown to bind with high affinity to single-stranded 50 splice site-like RNA with a sequence aGGUaa (lower case letters are less conserved) [7,27]. In addition, as mentioned above, RBM10 preferentially binds to G and U polyribonucleotides and little to poly(A), suggesting that RBM10 prefers G- and U-rich sequences [1]. Intriguingly, the 50 -splice-site region of exon 6 in Fas pre-mRNA has a sequence uuguuuggG|GUaaguucuu, where the 50 -splice site gG|GUaag is heavily surrounded by U and G bases (vertical line indicates the exon–intron boundary) (Fig. 3a). Therefore, RBM10 is likely to bind to this 50 -splice site, consequently blocking the site for splicing and bringing about exon 6 skipping. This model is supported by the results of Bcl-x pre-mRNA alternative splicing, where RBM10 led to selection of internal 50 -splice site of exon 2 (Fig. 3b). The 50 -splice site of exon 2 has a sequence GG|GUAAG, which is the same as that of exon 6 in Fas pre-mRNA shown above. Hence, RBM10 also likely binds to and blocks this 50 -splice site of exon 2, thereby allowing the selection of the internal 50 -splice site at 189 nucleotides upstream. The rather week RBM10 effect in this reaction (compare Fig. 2 and Fig. 1) may be due to low G/U abundance around the 50 -splice site. Based on the information presented here, we expect that the heptamer GGGUAAG is an RBM10 consensus binding sequence and is found at 50 -splice sites of target exons, and that the RBM10 action is promoted by a sequence(s) composed of about 6–8 or possibly more G and/or U bases in the vicinity of target 50 -splice sites. Additionally, some other mechanistic component(s) may be required to collaborate in the splicing regulation. Wang et al. very recently have reported [28] that RBM10 functions primarily in regulating exon-skipping, and the conclusion is essentially the same as the present study. By photoactivatableribonucleoside-enhanced crosslinking and immunoprecipitation, they showed that as many as 6396 genes of HEK cells are targets of RBM10, and their RBM10-binding sites are enriched in exons and in the vicinity of both 50 - and 30 -splice sites. Among the 21 exon-splicing changes validated by RNA-Seq and qPCR, 74% of them were skipping events. Wang et al. postulated a mechanistic model, in which RBM10 molecules promote exon skipping by

946

A. Inoue et al. / FEBS Letters 588 (2014) 942–947

a

Fas 3’ splice site of exon 6 caaccuacAG|Gauccagaucu

exon 5

guucuuucAG|Ugaagagaaag

exon 6

aagaggaaG|GUaauuauuuu

exon 7

uuguuuggG|GUaaguucuug

5’ splice site of exon 6

b

Bcl-x Internal 5’ splice site of exon 2 uuugaacaG|GUagugaauga

exon

exon 3

2

ggcggcugG|GUaagaaccaag

5’ splice site of exon 2 Fig. 3. Mechanistic model of RBM10-mediated regulation: putative consensus RBM10-binding sequence. RBM10 is proposed to recognize GGGUAAG at the 50 -splice sites of exon 6 in Fas pre-mRNA (a) and of exon 2 in Bcl-x pre-mRNA (b). The RBM10 binding may inhibit splicing at these sites, and promote skipping of exon 6 in Fas pre-mRNA or the selection of an upstream 50 -splice site in exon 2 in Bcl-x pre-mRNA. Red indicates the 50 -splice site sequences. G|GU is the conserved 50 -splice site trinucleotide. Vertical lines indicate exon–intron boundary.

binding in the vicinity of 4 splice sites of upstream and downstream introns [28]. They verified the RBM10 function with reporter minigenes, and interestingly showed that exon skipping was brought about by a single RBM10 molecule placed near the 50 -splice site of a target exon. This RBM10 arrangement was achieved with the aid of a PUF-3 domain fused to RBM10 and a PUF domain recognition sequence located in the intronic region of the reporter close to the 50 -splice site. Notably, this system and our present model are conformationally very similar at target 50 -splice sites, although we suggest that RBM10 binds directly to these sites. Furthermore, we demonstrated that RBM10 regulates alternative splicing of Fas and Bcl-x, and that RBM10 not only leads to exon-skipping, as shown with Fas, but also to 50 -splice site selection, as shown with Bcl-x. The exon skipping of Fas pre-mRNA was regulated by RBM10, while those of caspase 2, 3 and 9 were not. These results in the same stage (i.e., exon skipping) of splicing must have arisen from pre-mRNPs with different protein constituents, and spliceosomes with different splicing regulators such as RBM10. Spliceosome heterogeneity will become an important issue in understanding the action mechanisms of individual regulators. Acknowledgements We thank Dr. Nagase for a kind gift of cDNA plasmid of the hRBM10 larger variant (KIAA0122 clone). This work was supported by the Grants-in-Aid for Scientific Research from the Japan Society for Promotion of Science to A.I. (Grant number: 21550162). Appendix A We were almost finished preparing this manuscript, when the paper by Wang et al. was published in the September issue of EMBO Molecular Medicine 5, 1431-442, 2013 [28]. Appendix B. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.febslet.2014.01. 052.

References [1] Inoue, A., Takahashi, K.P., Kimura, M., Watanabe, T. and Morisawa, S. (1996) Molecular cloning of an RNA binding protein, S1-1. Nucleic Acids Res. 24, 2990–2997. [2] Johnston, J.J., Teer, J.K., Cherukuri, P.F., Hansen, N.F., Loftus, S.K., et al. (2010) NIH Intramural Sequencing Center (NISC). Massively parallel sequencing of exons on the X chromosome identifies RBM10 as the gene that causes a syndromic form of cleft palate. Am. J. Hum. Genet. 14, 743–748. [3] Gripp, K.W., Hopkins, E., Johnston, J.J., Krause, C., Dobyns, W.B. and Biesecker, L.G. (2011) Long-term survival in TARP syndrome and confirmation of RBM10 as the disease-causing gene. Am. J. Med. Genet. A 155A, 2516–2520. [4] Imielinski, M., Berger, A.H., Hammerman, P.S., Hernandez, B., Pugh, T.J., Hodis, E., et al. (2012) Mapping the hallmarks of lung adenocarcinoma with massively parallel sequencing. Cell 150, 1107–1120. [5] Inoue, A., Tsugawa, K., Tokunaga, K., Takahashi, K.P., Uni, S., Kimura, M., Nishio, K., et al. (2008) S1-1 nuclear domains: characterization and dynamics as a function of transcriptional activity. Biol. Cell 100, 523–535. [6] Xiao, S.J., Wang, L.Y., Kimura, M., Kojima, H., Kunimoto, H., Nishiumi, F., et al. (2013) S1-1/RBM10: multiplicity and cooperativity of nuclear localisation domains. Biol. Cell 105, 162–174. [7] Nguyen, C.D., Mansfield, R.E., Leung, W., Vaz, P.M., Loughlin, F.E., Grant, R.P. and Mackay, J.P. (2011) Characterization of a family of RanBP2-type zinc fingers that can recognize single-stranded RNA. J. Mol. Biol. 25, 273–283. [8] Aravind, L. and Koonin, E.V. (1999) G-patch: a new conserved domain in eukaryotic RNA-processing proteins and type D retroviral polyproteins. Trends Biochem. Sci. 24, 342–344. [9] Bonnal, S., Martínez, C., Förch, P., Bachi, A., Wilm, M. and Valcárcel, J. (2008) RBM5/Luca-15/H37 regulates Fas alternative splice site pairing after exon definition. Mol. Cell 32, 81–95. [10] Fushimi, K., Ray, P., Kar, A., Wang, L., Sutherland, L.C. and Wu, J.Y. (2008) Upregulation of the proapoptotic caspase 2 splicing isoform by a candidate tumor suppressor, RBM5. Proc. Natl. Acad. Sci. USA 105, 15708–15713. [11] Rappsilber, J., Ryder, U., Lamond, A.I. and Mann, M. (2002) Large-scale proteomic analysis of the human spliceosome. Genome Res. 12, 1231–1245. [12] Behzadnia, N., Golas, M.M., Hartmuth, K., Sander, B., Kastner, B., Deckert, J., Dube, P., et al. (2007) Composition and three-dimensional EM structure of double affinity-purified, human prespliceosomal A complexes. EMBO J. 26, 1737–1748. [13] Kuhn, A.N., van Santen, M.A., Schwienhorst, A., Urlaub, H. and Lührmann, R. (2009) Stalling of spliceosome assembly at distinct stages by small-molecule inhibitors of protein acetylation and deacetylation. RNA 15, 153–175. [14] Deckert, J., Hartmuth, K., Boehringer, D., Behzadnia, N., Will, C.L., Kastner, B., Stark, H., et al. (2006) Protein composition and electron microscopy structure of affinity-purified human spliceosomal B complexes isolated under physiological conditions. Mol. Cell. Biol. 26, 5528–5543. [15] Hegele, A., Kamburov, A., Grossmann, A., Sourlis, C., Wowro, S., Weimann, M., Will, C.L., et al. (2012) Dynamic protein–protein interaction wiring of the human spliceosome. Mol. Cell 45, 567–580. [16] Cheng, J., Zhou, T., Liu, C., Shapiro, J.P., Brauer, M.J., Kiefer, M.C., Barr, P.J. and Mountz, J.D. (1994) Protection from Fas-mediated apoptosis by a soluble form of the Fas molecule. Science 263, 1759–1762.

A. Inoue et al. / FEBS Letters 588 (2014) 942–947 [17] Cascino, I., Fiucci, G., Papoff, G. and Ruberti, G. (1995) Three functional soluble forms of the human apoptosis-inducing Fas molecule are produced by alternative splicing. J. Immunol. 154, 2706–2713. [18] Fricker, N., Beaudouin, J., Richter, P., Eils, R., Krammer, P.H. and Lavrik, I.N. (2010) Model-based dissection of CD95 signaling dynamics reveals both a proand antiapoptotic role of c-FLIPL. J. Cell Biol. 190, 377–389. [19] Kumar, S., Kinoshita, M. and Noda, M. (1997) Characterization of a mammalian cell death gene Nedd2. Leukemia 11, 385–386. [20] Wang, L., Miura, M., Bergeron, L., Zhu, H. and Yuan, J. (1994) Ich-1, an Ice/ced-3related gene, encodes both positive and negative regulators of programmed cell death. Cell 78, 739–750. [21] Huang, Y., Shin, N.H., Sun, Y. and Wang, K.K. (2001) Molecular cloning and characterization of a novel caspase-3 variant that attenuates apoptosis induced by proteasome inhibition. Biochem. Biophys. Res. Commun. 283, 762–769. [22] Chalfant, C.E., Rathman, K., Pinkerman, R.L., Wood, R.E., Obeid, L.M., Ogretmen, B. and Hannun, Y.A. (2002) De novo ceramide regulates the alternative splicing of caspase 9 and Bcl-x in A549 lung adenocarcinoma cells. Dependence on protein phosphatase-1. J. Biol. Chem. 277, 12587–12595.

947

[23] Boise, L.H., González-García, M., Postema, C.E., Ding, L., Lindsten, T., Turka, L.A., Mao, X., Nuñez, G. and Thompson, C.B. (1993) Bcl-x, a bcl-2-related gene that functions as a dominant regulator of apoptotic cell death. Cell 74, 597–608. [24] David, C.J. and Manley, J.L. (2010) Alternative pre-mRNA splicing regulation in cancer: pathways and programs unhinged. Genes Dev. 24, 2343–2364. [25] Edamatsu, H., Kaziro, Y. and Itoh, H. (2000) LUCA15, a putative tumour suppressor gene encoding an RNA-binding nuclear protein, is down-regulated in ras-transformed Rat-1 cells. Genes Cells 10, 849–858. [26] Corsini, L., Bonnal, S., Basquin, J., Hothorn, M., Scheffzek, K., Valcárcel, J. and Sattler, M. (2007) U2AF-homology motif interactions are required for alternative splicing regulation by SPF45. Nat. Struct. Mol. Biol. 14, 620–629. [27] Loughlin, F.E., Mansfield, R.E., Vaz, P.M., McGrath, A.P., Setiyaputra, S., Gamsjaeger, R., Chen, E.S., et al. (2009) The zinc fingers of the SRlike protein ZRANB2 are single-stranded RNA-binding domains that recognize 50 splice site-like sequences. Proc. Natl. Acad. Sci. USA 106, 5581–5586. [28] Wang, Y., Gogol-Döring, A., Hu, H., Fröhler, S., Ma, Y., Jens, M., Maaskola, J., et al. (2013) Integrative analysis revealed the molecular mechanism underlying RBM10-mediated splicing regulation. EMBO Mol. Med. 5, 1431–1442.