Dilated cardiomyopathy-associated RNA-binding motif protein 20 regulates long pre-mRNAs in neurons

The CLIP experiments were performed according to the seCLIP protocol (Van Nostrand et al., 2017a) with some minor modifications (Traunmüller et al., 2023). Olfactory bulbs from seven mice were pooled for each biological replicate, and for heart tissue, one heart was used per biological replicate. Samples were flash-frozen and ground on dry ice first in a metal grinder and a porcelain mortar. The frozen powder was transferred into a plastic Petri dish and distributed in a thin layer. The samples were UV-cross-linked three times at 400 mJ/cm² on dry ice with a UV-cross-linker (Cleaver Scientific). The powder was mixed and redistributed on the Petri dish before each UV exposure. After cross-linking, the powder was collected in 3.5 ml (olfactory bulbs) or 5.5 ml (heart) in lysis buffer (50 mM Tris-HCl pH 7.5, 100 mM NaCl, 1% NP-40, 0.1% SDS, 0.5% sodium deoxycholate, complete protease inhibitors, Roche) and 4 U/ml Turbo-DNase (Thermo Fisher). Samples were further processed as described in detail below.

For the clip samples preparation, the lysate was transferred into a glass homogenizer and homogenized by 30 strokes on ice. 1 ml aliquots of homogenized tissue were transferred to 2 ml tubes, 10 µl of RNaseI (Thermo Fisher) diluted in PBS (1:5–1:40) were added to each tube. Samples were incubated at 37°C with shaking for 5 min at 1200× rpm and then put on ice. 10 µl RNasin RNase inhibitor (40 U/µl, Promega) was added to each tube. Samples were mixed and centrifuged at 16,000 × g for 15 min at 4°C. The supernatants were transferred to a new tube, and 60 µl from each sample was taken and further processed for sized-matched INPUT (SMIn). 10 µl HA-magnetic beads (Pierce) was added to each sample and incubated at 4°C for 4 hr in a rotating shaker. Following incubation, the beads were washed 2× with a high salt wash buffer (50 mM Tris-HCl pH 7.5, 1 M NaCl, 1 mM EDTA, 1% NP-40, 0.1% SDS, 0.5% sodium deoxycholate), 2× with the lysis buffer, 2× with low salt wash buffer (20 mM Tris-HCl pH 7.5, 10 mM MgCl₂, 0.2% Tween-20) and 1× with PNK buffer (70 mM Tris-HCl pH 6.5, 10 mM MgCl₂). Beads were resuspended in 100 µl PNK mix (70 mM Tris-HCl pH 6.5, 10 mM MgCl₂, 1 mM DTT, 100 U RNasin, 1 U TurboDNase, 25 U Polynucleotide-Kinase [NEB]) and incubated at 37°C for 20 min on a shaking termomixer (1200× rpm). Upon RNA dephosphorylation, the beads were washed (2× high salt, 2× lysis, and 2× low salt buffers as before) and additionally with 1× Ligase buffer (50 mM Tris-HCl pH 7.5, 10 mM MgCl₂). Beads were then resuspended in 50 µl ligase mix (50 mM Tris-HCl pH 7.5, 10 mM MgCl₂, 1 mM ATP, 3% DMSO, 15% PEG8000, 30 U RNasin, 75 U T4 RNA-ligase [NEB]). 10 µl of the beads/ligase mix was transferred to a new tube, and 1 µl of pCp-Biotin (Jena Bioscience) was added to validate IP of the RNA-protein complexes by western blot. 4 µl of the RNA-adaptor mix containing 40 µM of each InvRiL19 & InvRand3Tr3 (IDT) was added to the remaining of the samples (40 µl). Samples were incubated at room temperature for 2 hr for adaptor ligation. Samples were washed 2× with high salt, 2× with lysis, and 1× with low salt buffers. Finally, beads were resuspended in 1× LDS sample buffer (Thermo Fisher) supplemented with 10 µM DTT and incubated for 10 min at 65°C, shaking on a thermomixer at 1200× rpm. Eluates or inputs were loaded on 4–12% Bis-Tris, 1.5 mm gel (Thermo Fisher) and separated at 130 V for ~1.5 hr. Proteins were transferred overnight at 30 V to a nitrocellulose membrane (Amersham). The membranes were placed in a 15 cm Petri dish on ice, and an area between 55 and 145 kDa was cut out into small pieces and transferred into a 2 ml tube.

RNA extraction, reverse transcription using InvAR17 primer, cDNA clean-up using silane beads (Thermo Fisher), second adaptor ligation (InvRand3Tr3), and cDNA purification steps were performed as previously described (Van Nostrand et al., 2016). The sequencing libraries were amplified using Q5-DNA polymerase (NEB) and i50X/i70X Illumina indexing primers (IDT). Final libraries were amplified with 14 cycles. Libraries were purified and concentrated with ProNEX size-selective purification system (Promega) using sample/beads ratio of 1/2.4. Samples were loaded on a 2% agarose gel, and the area corresponding to the size between 175 and 350 bp was cut out. The amplified and purified libraries were then extracted from the gel using a gel extraction kit (Machery&Nagel) and eluted with 16 µl.

The concentrations and the size distributions of the libraries were determined on the Fragment Analyzer system (Agilent). 75 bp single-end sequencing was performed on the NextSeq500 platform using Mid Output Kit v2.5 (75 cycles).

Adaptor and primer sequences used in this study:

X* = Phosphorthioated base

seCLIP data processing was performed as described (Van Nostrand et al., 2016; Van Nostrand et al., 2017b; Van Nostrand et al., 2020). In brief, raw reads were processed to obtain unique CLIP tags mapped to mm10 using Clipper (https://github.com/YeoLab/clipper ; Lovci et al., 2025; https://github.com/YeoLab/eclip ; Yee and Domissy, 2022). Reads from replicates 1 and 2 from the olfactory bulb were concatenated. Peak normalization was performed by processing the SMInput samples using the same peak calling pipeline. Irreproducible discovery rate (IDR) analysis was performed to identify reproducible peaks across biological replicates (Li et al., 2011). IDR (https://github.com/nboley/idr; Boley, 2017) was used to rank seCLIP peaks by the fold change over the size-matched input. Clip peaks were called based on IDR<0.05. We observed some short highly represented sequences that were not specific to RBM20 seCLIP isolations, which were excluded based on peak shape and width (<30 bp) using StoatyDive (Heyl and Backofen, 2021).

For motif discovery, cross-link-induced truncation sites (CITS) were called using the CTK pipeline (Shah et al., 2017). Briefly, unique tags from replicates were combined, and CITS were called by requiring FDR<0.001. Sequences from –10 bp to +10 bp from called CITS were used as input sequences for DREME software (Bailey et al., 2009; Bailey et al., 2015; Nystrom and McKay, 2021). As a control, sequences of the same length located 500 bp upstream of the CITS site (–510 to –490 bases) were used. Enrichment of the UCUU motif at the CITS sites was calculated.