These authors contributed equally to this work.
Integrative analysis revealed the molecular mechanism underlying RBM10-mediated splicing regulation
Article first published online: 22 AUG 2013
Copyright © 2013 EMBO Molecular Medicine
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
EMBO Molecular Medicine
Volume 5, Issue 9, pages 1431–1442, September 2013
How to Cite
Wang, Y., Gogol-Döring, A., Hu, H., Fröhler, S., Ma, Y., Jens, M., Maaskola, J., Murakawa, Y., Quedenau, C., Landthaler, M., Kalscheuer, V., Wieczorek, D., Wang, Y., Hu, Y. and Chen, W. (2013), Integrative analysis revealed the molecular mechanism underlying RBM10-mediated splicing regulation. EMBO Mol Med, 5: 1431–1442. doi: 10.1002/emmm.201302663
- Issue published online: 3 SEP 2013
- Article first published online: 22 AUG 2013
- Manuscript Accepted: 12 JUL 2013
- Manuscript Revised: 6 JUL 2013
- Manuscript Received: 20 FEB 2013
- Federal Ministry for Education and Research (BMBF)
- Berlin Institute of Medical Systems Biology (BIMSB) (315362A, 315362C)
- China Scholarship Council (CSC)
- GENCODYS (241995) in the European Union Framework Programme 7
As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.
|emmm201302663.reviewer_comments.pdf||PDF document||1949K||Review Process File|
Figure S1. RBM10 PAR-CLIP experiments and results. (A) Phosphorimages of SDS–PAGE that resolved 32 P-labelled RNA–FLAG/HA-RBM10 complexes immunoprecipitated (IP) from lysates of HEK293 cells cultured in media in the absence or presence of 100 μM photoactivatable 4sU and crosslinked with UV 365 nm. It was clear that RBM10 indeed bound with RNA and that the addition of 4sU greatly enhanced the crosslinking efficiency. (B) Distribution of specific mismatches in aligned PAR-CLIP reads. The predominate T to C mismatches are signature of efficient crosslinking. (C) Length distribution of RBM10 binding clusters. (D) Distribution of number of PAR-CLIP reads (left) or PAR-CLIP reads containing T–C conversions (right) within all the clusters or the consensus clusters. (E) Correlation of RBM10 binding affinities of consensus clusters measured in the two replicates (i.e. the number of PAR-CLIP reads spanning the preferred crosslinking sites).
Figure S2. RNA-seq experiments and results. (A) RBM10 knockdown (KD) efficiency. Assessment of RBM10 KD efficiency by qPCR and Western blot. Compared with control, 48 and 72 h after KD, the RBM10 protein level was decreased to approximately 24–41%. For RNA-seq experiments, we harvested the cells 24, 48 and 72 h after KD, see Supporting Information Table S2. (B) RBM10 overexpression (OE) efficiency. Assessment of RBM10 OE efficiency by Western blot. Compared with control, after OE, the RBM10 protein level was increased by two to ninefold upon stimulation with different concentration of Dox. For RNA-seq and PAR-CLIP experiments, we used the cells stimulated with 10 ng/ml Dox. (C) Gene expression changes after RBM10 KD. In the scatterplot, gene expression levels (in RPKM) of control cells (X axis) were plotted against that of cells after KD (Y axis). Differentially expressed genes were marked in blue. (D) Gene expression changes after RBM10 OE. In the scatterplot, gene expression levels (in RPKM) of control cells (X axis) were plotted against that of cells after OE (Y axis). Differentially expressed genes were marked in blue. (E) Gene expression changes (Z value) induced by RBM10 KD (X axis) were plotted against those induced by RBM10 OE (Y axis). Differentially expressed genes in either condition were marked in blue. (F) Scheme for computing the percent spliced-in (PSI) value of the middle exon. (G) Scheme for computing the percent intron retention (PIR) value of an intron. (H) MA plot with ΔPSI of exons between RBM10 OE and Control at Y axis and the number of all reads used for calculating the ΔPSI at X axis. The blue line denotes the local standard deviation (window size = 1% of exons), the red line denotes loess line.
Figure S3. qPCR validation of splicing changes detected by RNA-seq. qPCR validation of 21 splicing changes identified based on RNA-seq data. Primers targeting transcript isoforms including (I) or excluding (E) the 21 cassette exons (Table S5) were used to measure the expression level of the respective isoforms, which was then normalized based on the expression level of the neighbouring constitutive exons (C).
Figure S4. Cumulative distribution functions of splicing changes upon RBM10KD. Cumulative distribution functions of splicing changes upon RBM10 KD for different groups of cassette exons with RBM10 binding close to none or one of the four splicing sites of the adjacent introns (upper panel), or to different number of the four splicing sites (lower panel). The exons with RBM10 binding close to one of the four splice sites were more likely included upon RBM10 KD, and those with binding close to 3′ splice sites of upstream introns exhibiting the weakest and insignificant inclusion propensity. Intriguingly, exons with binding close to more of the four splice sites showed progressively stronger inclusion tendency upon RBM10 KD (please note 1) the group of exons with binding close to all four splice sites was too small to be reliable and (2) compared with RBM10 OE induced changes, the splicing changed induced by KD was overall weaker).
Figure S5. Representative agarose gel and bioanalyser gel image (Upper panel) in minigene experiments. Label is the same as Fig 4.
Figure S6. Validation of an in-frame deletion of RBM10 detected in a patient with TARP syndrome. (A) PCR on genomic DNA from the patient (III:1) and his mother (Carrier) with a combination of two primer sets, one within the deletion and the other flanking the deletion, was used to validate the deletion. Using a single primer set that flanked the deletion failed to amplify the normal allele in heterozygous females. The primer sequences were listed in Table S5. (B) RT-PCR performed on patient (III:1) and healthy control-derived lymphoblast cells demonstrating that RBM10 mRNA from the patient was expressed at similar levels, but with different size. (C) Western blot performed on patient and healthy control-derived lymphoblast cells demonstrating that RBM10 protein from the patient was expressed at 70% levels compared with the control, but with different size.
Figure S7. Photos from the two patients reported in this study (Informed consent was obtained from the parents). (A–C) photos from patient III:1; (D–F) photos from patient III:2.
Figure S8. RBM10 binding to U2 snRNA. (A) Distribution of RBM10 PAR-CLIP reads at both major and minor spliceosomal snRNAs. More than 65% of PAR-CLIP reads derived from snRNAs could be mapped to U2. (B) Distribution of RBM10 PAR-CLIP reads containing T-C conversions along the U2 consensus sequence. (C) Two strong crosslinking sites marked with red arrows were close to U2 conserved sequences responsible for branching site pairing.
Table S1. Summary of PAR-CLIP sequencing results.
Table S2. Summary of RNA sequencing results.
Table S5. Primer sequences.
Table S6. Summary of phenotypic spectrum in TARP patients and our patients.
Table S3. Differentially expressed genes upon RBM10 perturbation.
Table S4. Differentially spliced exons upon RBM10 perturbation.
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.