Comment[ArrayExpressAccession] E-GEOD-45148 MAGE-TAB Version 1.1 Public Release Date 2013-03-15 Investigation Title HITS-CLIP analysis of FMRP mRNA binding sites from P11-P25 mouse brain polysomes Comment[Submitted Name] HITS-CLIP analysis of FMRP mRNA binding sites from P11-P25 mouse brain polysomes Experiment Description FMRP loss-of-function causes Fragile X Syndrome (FXS) and autistic features. FMRP is a polyribosome-associated neuronal RNA-binding protein, suggesting that it plays a key role in regulating neuronal translation, but there has been little consensus regarding either its RNA targets or mechanism of action. Here we use high throughput sequencing of RNAs isolated by crosslinking immunoprecipitation (HITS-CLIP) to identify FMRP interactions with mouse brain polyribosomal mRNAs. FMRP interacts with the coding region of transcripts encoding pre- and postsynaptic proteins, and transcripts implicated in autism spectrum disorders Polyribosomes were prepared from UV254-crosslinked pooled littermate FVB mouse brains as described in detail in published paper PMID 21784246. Polyribosomes were dissociated under denaturing conditions using two different protocols to disrupt ribonucleoprotein complexes. In most experiments Fmr1-null littermates were used in parallel to confirm FMRP specificity. After mild RNAse treatment to reduce the size of RNA to 60-100 nucleotides, endogenous FMRP was immunoprecipitated with one of two antibody combinations, either mixed monoclonal antibodies 7G1-1 and 2F5, or polyclonal antibody ab17722 (Abcam). The ages of the mice were P11, P13, P14, P15, and P25. In addition, to assess a different neuronal polysome-associated protein, as another control, lysates prepared from two samples, the P11 and P13 mice, were split in half and one half IPed with a human anti-Hu antisera as the neuronal Hu proteins are polysome-associated in brain. In sum, 7 FMRP HITS-CLIP experiments were performed including (1) P14 mouse polysomes prepared by Protocol 1 and IPed with 7G1-1/2F5 (2) P14 mouse polysomes prepared by Protocol 1 and IPed with ab17722 (3) P15 mouse polysomes prepared by Protocol 1 and IPed with 7G1-1/2F5 (4) P25 mouse polysomes prepared by Protocol 1 and IPed with 7G1-1/2F5 (5) P25 mouse polysomes prepared by Protocol 1 and IPed with ab17722 (6) P11 mouse polysomes prepared by Protocol 2 and IPed with ab17722 and (7) P13 mouse polysomes prepared by Protocol 2 and IPed with ab17722. FMRP-bound or Hu-bound RNA tags were cloned and submitted for high throughput sequencing on the Life Science 454 platform (samples 1-5) or the Illumina platform (samples 6 and 7 and the 2 Hu samples processed in parallel with samples 6 and 7). Full details are given in PMID 21784246. Term Source Name ArrayExpress EFO Term Source File http://www.ebi.ac.uk/arrayexpress/ http://www.ebi.ac.uk/efo/efo.owl Person Last Name Darnell Darnell Zhang Darnell Person First Name Jennifer Jennifer Chaolin Robert Person Mid Initials C B Person Email darneje@rockefeller.edu Person Affiliation The Rockefeller University Person Address The Rockefeller University, 1230 York Ave., New York, NY, USA Person Roles submitter Protocol Name P-GSE45148-1 P-GSE45148-2 P-GSE45148-3 Protocol Description Mice of the indicated genotypes and ages were sacrificed by isoflurane anaesthesia and decapitation. The brain was removed and placed in ice-cold dissection buffer (1X HankM-bM-^@M-^Ys balanced salt solution (Gibco) containing 10 mM HEPES, pH 7.3 (USB) (HHBSS) and 0.1 mg/ml cycloheximide (Sigma C-7698, made fresh in methanol). Cortex and cerebellum were dissected free of underlying white matter, homogenized in 1 ml lysis buffer (10mM HEPES-KOH, pH 7.4, 150mM KCl or NaCl, 5mM MgCl2, 0.5mM dithiothreitol, 0.1 mg/ml cycloheximide, 1X Complete EDTA-free protease inhibitor cocktail (Roche), 40U/ml rRNAsin (Promega)) per brain, with 10-12 strokes at 900rpm in a motor-driven Teflon-5 ml glass homogenizer. NP-40 was added to a final concentration of 1% from a 10% stock (Calbiochem) and allowed to sit on ice 10 minutes. The homogenate was spun at 2,000xg for 10min at 4C. The supernatant (S1) was respun at 20,000xg for 10min at 4C. The resulting supernatant (S2) was loaded onto a 20-50% w/w linear density gradient of sucrose in gradient buffer (10mM HEPES-KOH pH 7.4, 150mM KCl, and 5mM MgCl2) prepared using a Gradient Master 107 (BioComp), in 14 x 89 mm polyallomer ultracentrifuge tubes (Beckman 331372). Gradients were centrifuged at 40,000 rpm for 2hrs at 4C in a Beckman SW41 rotor and sixteen fractions of 0.75ml volume were collected with continuous monitoring at 260nm using an ISCO UA-6 UV detector Crosslinking-IP (HITS-CLIP) Protocol 1: Polyribosome gradients were prepared from P14, P15 and P25 male FVB wild type and Fmr1Tm1Cgr littermate mice as described above. The S2 supernatants were UV-crosslinked in a 3.5 cm tissue culture dish on a bed of ice- slush three times at 400 mJ/square cm (254nm UV light) with swirling between each irradiation, using a Stratalinker 2400. Fractions containing polyribosome-associated FMRP were pooled, diluted 1:1 with gradient buffer and polyribosomes were pelleted by centrifugation for 2hr at 300,000xg in polycarbonate centrifuge tubes (Beckman 362305) in a TLA110 rotor (Beckman) using a tabletop ultracentrifuge (Optima MAX, Beckman) at 4C. (2) Dissociation of RNPs, RNAse treatment and IP-- Polyribosome pellets were dissociated in 1% SDS containing buffer based on a published protocol (Hinck et al., 1994) with modifications. The polyribosomal pellet was resuspended in 100 ul SDS-IP buffer (15 mM Tris, pH 7.5, 5 mM EDTA, 2.5 mM EGTA, and 1% SDS), heated to 90 degrees for 10 minutes with vigorous shaking to dissociate RNP complexes, and diluted with 9V of CSK buffer (20 mM Tris, pH 7.0, 50 mM NaCl, 3 mM MgCl2, 0.5% Triton X- 100, 300 mM sucrose) and stored on ice. Lysates were treated with 200 units/ml RNAse T1 (Applied Biosystems/Ambion AM2280) in CSK buffer for 10 minutes at 37C and stopped with the addition of 1320 units of SuperAse-In (Applied Biosystems/Ambion) so that RNAs crosslinked to FMRP were reduced to a modal size of about 60-100 nts. FMRP was IPed from these samples using Protein A Dynabeads (Invitrogen) bound to either rabbit polyclonal anti-FMRP (ab17722, Abcam) or mouse mixed monoclonal antibodies to 2 different epitopes on FMRP (7G1-1; (Brown et al., 2001) and 2F5; (Gabel et al., 2004)) described below. Following IP (described below) the beads were washed sequentially (1 ml washes) with high stringency buffer (15 mM Tris-HCl, pH 7.5, 5 mM EDTA, 2.5 mM EGTA, 1% Tx-100, 1% Na-DOC, 0.1% SDS, 120 mM NaCl, 25 mM KCl), high salt buffer (15 mM Tris-HCl, pH 7.5, 5 mM EDTA, 2.5 mM EGTA, 1% Triton X-100, 1% Na-DOC, 0.1% SDS, 1 M NaCl) and twice with low salt buffer (15 mM Tris-HCl, pH 7.5, 5 mM EDTA) followed by one wash with NT-2 buffer (50 mM Tris-HCl pH 7.4, 150 mM NaCl, 1 mM MgCl2, and 0.05% NP-40). (3) Dephosphorylation of RNA, 3M-bM-^@M-^Y linker ligation and SDS-PAGE separation of RNABP:RNA complexes-- Immunoprecipitations were treated with calf intestinal phosphatase to remove the 5M-bM-^@M-^Y phosphate from RNA crosslinked to FMRP (M-bM-^@M-^\RNA tagsM-bM-^@M-^]). Beads were resuspended in 0.08 ml of 1X dephosphorylation buffer and 3 units of calf intestinal alkaline phosphatase (CIAP, Roche) was added, incubated in the Thermomixer R at 37C for 20 min, programmed to shake at 1000rpm for 15 sec every 4 min, followed by 3X 1ml washes with NT-2, 2X PXL, and 2X NT-2 with one tube change during washing. To ligate a 32P-labelled RNA linker to the 3M-bM-^@M-^Y end of the RNA, the puromycin- blocked linker (L32, 5M-bM-^@M-^Y-OH-GUGUCAGUCACUUCCAGCGG-3M-bM-^@M-^Y-puromycin) was first labeled using T4 phosphonucleotide kinase (PNK, New England Biolabs, NEB). 50 pmol of L32 RNA, 0.015 ml 32P-M-NM-3-ATP, and 2 ul of T4 PNK in 1X PNK buffer were incubated at 37C for 30min, an additional 0.02 ml 10 mM ATP was added and the reaction incubated 5 min. Radiolabeled linker was spun through a G-25 column (Amersham) to remove free ATP. 30 pmol of the labeled 3M-bM-^@M-^Y RNA linker was added to a 0.08 ml T4 RNA ligase reaction (Fermentas), according to kit instructions, and on-bead ligation reactions were incubated at 16C for 1 hr in the Thermomixer R programmed 1000rpm for 15 sec every 4 min. After 1 hr, 80 pmol of L32 RNA linker WITH 5M-bM-^@M-^Y phosphate was added to each tube of reaction and incubated overnight followed by 3X wash with NT-2 buffer. To rephosphorylate the 5M-bM-^@M-^Y end of the RNA for 5M-bM-^@M-^Y linker ligation, 80 ul of PNK mix (6 ul of 1 mM ATP, 4 ul T4 PNK enzyme (NEB), 1 ul RNasin in 1X PNK buffer (NEB), total volume 80ul) was added to each tube and incubated for 20 min at 1000rpm shaking for 15 sec every 4 min and washed 1X with PXL and 2X with NT-2 buffers. Each tube of beads was resuspended in 40 ul NuPAGE loading buffer (LDS, Invitrogen), containing reducing agent, and incubated at 70C for 10min at 1000rpm. Supernatants were taken off the beads and run on Novex NuPAGE 10% Bis-Tris gels (Invitrogen) in MOPS running buffer (Invitrogen) for 3 hr at 175V and transferred to Protran nitrocellulose (S&S) using a Novex wet transfer apparatus (Invitrogen). After transfer, the nitrocellulose was quickly rinsed with RNAse-free PBS, blotted with Kimwipes, wrapped in plastic wrap and exposed to Biomax MR film (Kodak). (4) Recovery of complexes, protease digestion and 5M-bM-^@M-^Ylinker ligation--Nitrocellulose membranes were aligned carefully with the exposed film and filter excised with a scalpel from 105-140 kDa kDa (the size of FMRP crosslinked to RNA of approx. 100+ nucleotides). Each band of nitrocellulose membrane was further cut into smaller pieces and proteinase K treated (0.2 ml of a 1:5 dilution of proteinase K (4mg/ml, Roche) in PK buffer (100mM Tris-Cl pH 7.5, 50mM NaCl, 10mM EDTA)) at 37C, 1100rpm for 20min. Then 0.2 ml of PK+7M urea solution was added and incubated for another 20min at 37C at 1100rpm. Finally, 0.4 ml of RNA phenol (pH 6.8, Applied Biosystems/Ambion) and 0.13 ml of 49:1 CHCl3:isoamyl alcohol were added and incubated at 37C, 1100rpm for additional 20 min. Tubes were spun at 20,000xg in a microcentrifuge. Glycogen (0.5 ul, Applied Biosystems/Ambion), 50ul 3M NaOAc pH 5.2, and 1 ml of 1:1 ethanol:isopropanol were added to the aqueous phase in a fresh tube and RNA was precipitated overnight at -20C. RNA was pelleted, washed with 75% ethanol, the pellet dried and dissolved in 6 ul RNAse-free H2O. RNA ligation to add the 5M-bM-^@M-^Y linker (L51 (RNA linker 51), 5M-bM-^@M-^Y-OH- AGGGAGGACGAUGCGG-3M-bM-^@M-^Y) was performed with 1ul 10X T4 RNA ligase buffer (Fermentas), 1ul BSA (0.2ug/ul), 1ul ATP (10mM), 0.1ul T4 RNA ligase (3U, Fermentas), and 20 pmol of each L51 RNA linker in a total volume of 10 ul at 16C overnight. To the ligation reaction, 77ul H2O, 11ul 10X RQ1 DNase buffer, 5ul RQ1 DNase (Promega) and 5ul RNasin (Promega) were added and incubated at 37C for 20min. 0.3 ml H2O, 0.3 ml RNA phenol (Ambion) and 0.1 ml CHCl3 were added, vortexed, spun and the aqueous layer taken. RNA was precipitated with 50ul 3M NaOAc pH 5.2, 1ul glycogen (Ambion) and 1 ml 1:1 ethanol:isopropanol overnight at -20C. (5) RT-PCR and high-throughput sequencing of PCR products--RNA was pelleted, washed, dried, and resuspended in 8 ul of RNAse-free H2O. The RNA was mixed with 2ul of primer P32 (3M-bM-^@M-^Y DNA primer 32, 5M-bM-^@M-^Y- CTTCACTCACCTCG standard 454 GS20 base caller 454 reads collected from five different FMRP CLIP experiments were mapped to the mouse genome (NCBI37/mm9, July 2007) by BLAT (Kent, 2002), using the parameters stepSize=5 and minScore=15. A read whose best score was higher than the second best was considered to be unambiguously mappable. Reads without unambiguous mapping were excluded from further analysis. For each of the five protocol 1 CLIP experiments in which the fulllength of the RNA tag was sequenced, reads with the same start and stop positions were considered to be duplicates potentially resulting from RT-PCR amplification, and collapsed to identify unique reads used for further analysis. Two additional FMRP CLIP experiments and two Hu CLIP experiments sequenced by the Illumina Genome Analyzer are 76-nucleotide (nt) reads, including a 5-nt degenerate M-bM-^@M-^\barcodedM-bM-^@M-^] linker sequence read-able from the 5M-4 end of each CLIP tag to distinguish potential PCR duplicates from genuinely unique reads mapped to the same locations. Therefore, the sequence of actual RNA tags was determined up to 71 nt in length, and the relatively long reads maximized the chance of detecting the full length RNA fragments (modal size ~50 nt) pulled down by immunoprecipitation. To minimize sequencing errors resulting from the ultrahigh-throughput sequencing and hence alignment errors in downstream analysis, we first filtered the raw reads based on the quality scores. We required a minimum quality score 20 in the first five nucleotides (the degenerate region), and an average score 20 in the next 25 nucleotides (i.e., the first 25 nucleotides of each actual RNA tag). The filtered reads were then subject to the removal of the 5M-4 degenerate M-bM-^@M-^\barcodeM-bM-^@M-^] (whose identity was however recorded) and the 3M-4 linker to get the actual RNA tag sequences. Since RNA CLIP tags differ in size, 3M-4 linkers were detected and trimmed by aligning each read to the linker sequence using the needle program (parameters: gapopen: 100, gapext 10 for alignment; #matches-#mismatches>=4 nt, %mismatches<=15%, and #indels <1 nt for detection of the linker (Rice et al., 2000)). The linker-trimmed reads were then mapped to the mouse genome (NCBI37/mm9, July 2007) by the program Novoalign (http://www.novocraft.com), requiring $2 mismatches (substitutions or indels) in 28 nt. Only reads with unambiguous mapping were kept. Mappable reads for all four experiments were then pooled together, so that PCR duplicates and potential contaminations (e.g., between neighboring lanes in the flow cells during Illumina sequencing) were eliminated to identify unique reads, using an expectation-maximization (EM)-based algorithm, which modeled the starting coordinates of mappable tags, and the identity and copy number of their degenerate M-bM-^@M-^\barcodeM-bM-^@M-^] sequences. More detailed description is published in PMID 21784250. Genome_build: mm9 Supplementary_files_format_and_content: BED files with unique CLIP tags after removal of PCR duplicates Protocol Type specified_biomaterial_action nucleic acid library construction protocol feature_extraction Experimental Factor Name AGE DISSOCIATION PROTOCOL AND SEQUENCING PLATFORM IP ANTIBODY Experimental Factor Type age dissociation protocol and sequencing platform ip antibody Comment[SecondaryAccession] GSE45148 Comment[GEOReleaseDate] 2013-03-15 Comment[ArrayExpressSubmissionDate] 2013-03-13 Comment[GEOLastUpdateDate] 2013-03-18 Comment[AEExperimentType] RNA-seq of coding RNA Comment[AEExperimentType] other Comment[AdditionalFile:Data1] GSE45148_jendarnell_supplementalinformation_PMID_21784246tableS2C.txt Comment[SecondaryAccession] SRP019264 Comment[SequenceDataURI] http://www.ebi.ac.uk/ena/data/view/SRR772116-SRR772124 SDRF File E-GEOD-45148.sdrf.txt