Proximity-CLIP and expedited non-radioactive library preparation of small RNA footprints for next generation sequencing

Abstract

Along their life cycle most RNAs move between several cellular environments where they associate with different RNA Binding Proteins (RBPs). Reciprocally, a significant portion of RBPs reside in more than a single cellular compartment, where they can interact with discrete RNAs and even exert distinct biological roles. Proximity-CLIP combines proximity biotinylation of proteins with photoactivatable ribonucleoside-enhanced protein-RNA crosslinking to simultaneously profile the proteome including RBPs and the RBP-bound transcriptome in any given subcellular compartment. Here we provide a detailed experimental protocol for Proximity-CLIP with a simplified non-radioactive, small RNA cDNA library preparation protocol.

Introduction

Function and regulation of both coding and non-coding RNA molecules depends on their specific localization (Wilk et al., 2016; Kejiou and Palazzo, 2017; Lécuyer et al., 2007). Fractionation-independent methods allow high resolution mapping of protein localization within cells, such as fluorescence microscopy and more recently also proximity-labeling coupled to high-throughput proteomics (O’callaghan et al., 1999; de Boer et al., 2003; Yang et al., 2004; Fernández-Suárez et al., 2008; Hung et al., 2016).

Proximity labeling of proteins can be achieved using engineered Soy Ascorbate Peroxidase, APEX2, that can oxidize biotin-phenol in the presence of hydrogen peroxide, thus generating rapidly decaying biotin-phenoxyl radicals (t1/2<1 ms) (Hung et al., 2016). The majority of biotin-phenoxyl radicals decay by reacting with water, but some will react with aromatic amino acids in APEX2-proximal proteins and biotinylate them. Therefore, compartment-specific proteins can be biotinylated and isolated by affinity chromatography by fusing localization signals to APEX2 and targeting it to a given cellular compartment. In addition to protein focused techniques, proximity-based approaches were recently developed to study mRNA localization (Li et al., 2018; Fazal et al., 2019; Kaewsapsak et al., 2017; Padron et al., 2019).

Compared to these methods, Proximity-CLIP more inclusively detects the multitude of RNA species at various sub-cellular compartments (Benhalevy et al., 2018a; 2018b). These include short-lived RNAs that may undergo rapid cleavage or degradation (miRNA, mRNA, tRNA and rRNA precursors, enhancer and antisense lncRNAs, divergent transcription products), short RNAs, RNAs with distinct 5’ and 3’ features and RNAs susceptible to degradation after cell lysis by released de-regulated RNases. Proximity-CLIP also enables discovery of regulatory RNA elements occupied and protected by proteins at subcellular resolution. Collectively, Proximity-CLIP allows for (1.) the determination of the localized proteome in general and the RBPome in particular using mass spectrometry; (2.) the profiling of localized transcripts using RNA-seq; and (3.) the identification and quantification of RBP-occupied RNA elements within transcripts through isolation of RNase-resistant footprints that are converted into next-generation sequencing compatible cDNA libraries (Figure 1).

Figure 1.

Scheme of Proximity-CLIP and an optimized streamlined footprint library preparation. A. APEX2 is targeted to a compartment of interest by fusion to a localization element (LE), and nascent RNAs labelled with 4SU. Cells are incubated with Biotin-phenol (BP) for 30 min, and APEX2-mediated BP oxidation is induced by addition of hydrogen peroxide for 1 min. Biotin radicals are created locally and either covalently tag APEX2-proximate proteins or rapidly decay (t1/2 < 1 ms). Next, the oxidation reaction is quenched under UV light (λ > 312 nm) for protein-RNA crosslinking. Then, cells are lysed, and compartment-specific proteins and ribonucleoproteins are captured by streptavidin affinity chromatography. Purified Protein-RNA complexes are subjected for protein mass spectrometry, RNA-Seq and footprint cDNA library preparation. B. A streamlined non-radioactive footprint library preparation protocol where RNase treatment, dephosphorylation, phosphorylation and 3’ adapter ligation are performed on beads. 3’ adapter ligated, and phosphorylated footprints are then retrieved by Proteinase K digestion and a 5’ adapter is ligated on the footprints. Following reverse transcription (RT) and initial low cycle PCR with a short 5’ primer and the RT primer, products are size-selected on an agarose gel. Size-selected cDNA is used for a calibration PCR and a final PCR using long sequencing-compatible longer primers, followed by next-generation sequencing.

BASIC PROTOCOL 1

Cell culture, 4SU labeling, proximity biotinylation and crosslinking

This section describes the steps of Proximity-CLIP performed prior to cell lysis including 4SU labeling, proximity biotinylation, and crosslinking. This protocol can be adapted to any adherent mammalian cell line. We strongly recommend reading the commentary section before planning the experimental setup. In addition to the cells expressing APEX2 targeted to the compartment of interest, cells expressing APEX2 in a control compartment, as well as the parental cell line, which does not express APEX2 are required (see commentary/critical parameters and troubleshooting/cells and controls for considerations regarding the construction of cell lines).

Materials

Cells expressing APEX2 -fusion proteins targeted to control and of-interest compartments, and the parental cell line that does not express APEX2 (See commentary/critical parameters and troubleshooting/cells and controls).

500 mM 4-thiouridine (4SU) in DMSO. Store at −20°C.

500 mM biotin-phenol (BP) in DMSO (may need to be sonicated to dissolve, store 50 μl aliquots at −80°C).

1 M sodium azide. For long term store aliquots at −20°C.

Sodium ascorbate (powder, needed fresh)

Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid) (powder, needed fresh)

30% wt/wt hydrogen peroxide. Needed fresh, do not dilute solution in advance.

Any standard fixative to process cells for imaging by fluorescence microscopy. In our hands 16% PFA freshly diluted to 4% in PBS performed better than methanol at −20°C, specifically for detection of cytoplasmic V5-tagged APEX2.

Phosphate buffered saline (PBS) pH 7.4: 0.144 g/L KH₂PO₄, 9 g/L NaCl, 0.795 g/L Na₂HPO₄-7H₂O

10x PBS pH 7.4: 1.44 g/L KH₂PO₄, 90 g/L NaCl, 7.95 g/L Na₂HPO₄-7H₂O

DMSO

UV crosslinker (Spectrolinker XL-1500, Spectronics Corporation) equipped with far UV light bulbs (wavelength > 310 nm, preferably 365 nm).

1. Seed cells for experiment. For HEK293, per biological replicate split 25×10⁶ cells 36 hours prior to the experiment at equal density between one 15 cm (preparative plate), three 6 cm plates for Western blot analyses (+BP & hydrogen peroxide, -BP but + hydrogen peroxide and +BP but -hydrogen peroxide), and four cover glasses (-BP and -hydrogen peroxide controls and two +BP & + hydrogen peroxide glasses for immunofluorescence analyses).

   See commentary/critical parameters and troubleshooting/fluorescence microscopy for more details and considerations.

   Important! All cell lines should be in a similar growth state and confluence.

2. 16 h before crosslinking, add 4-thiouridine (4SU) to the cell culture media at a final concentration of 100 μM (see commentary/critical parameters and troubleshooting/4SU labeling).

3. Preparations prior to next steps: Bring cell media to 37°C and PBS to room temperature, label one 1.5 ml tube per sample and pre-chill them on ice, have liquid N₂ on hand for snap-freezing cell samples, and ensure access to a sonicator (in case Trolox does not dissolve). Weigh (but do not dissolve) required amounts of sodium ascorbate and Trolox.

4. Dilute BP stock solution 1:50 in prewarmed media to generate a 10 mM working stock, and immediately add (except to -BP controls) 50 μl of the 10 mM BP working stock per 1 ml of media to cell plates for a final concentration of 500 μM BP. Gently swirl and return plates to the incubator for 30 min.

5. During the 30 min BP incubation: (1.) Prepare 4% PFA in 1x PBS using 16% PFA and 10x PBS stock solutions. (2.) Dilute H₂O₂ (most come as 30% wt/wt, which is ~10 M) to 100 mM in PBS. (3.) Dissolve sodium ascorbate to 1 M (in water) and Trolox to 500 mM (in DMSO, sonicate if necessary). (4.) Prepare fresh quencher solution (PBS supplemented with 10 mM sodium ascorbate, 5 mM Trolox and 10 mM sodium azide). (5.) Pre-aliquot quenching solution into prechilled 50 ml tubes.

6. Bring cell plates from the incubator to the bench, add 100 mM hydrogen peroxide stock into growth medium for a final concentration of 1 mM (except for the controls without hydrogen peroxide), swirl and incubate for 1 min. Proceed immediately, do not exceed 1 min incubation to minimize the labeling time and possible diffusion of biotinylated proteins in the cells.

7. Decant the labeling cell medium (or aspirate from the cells growing on cover-glass for immunofluorescence microscopy), and quickly wash the plates 3 times in quenching solution (not less than 1 ml per 3 cm² surface area per wash).

   Quenching solution is pre-aliquoted in step 5 to allow rapid, yet careful pouring of solutions (see commentary/critical parameters and troubleshooting/cells and controls). Consider recruiting an extra pair of hands and streamline this step, in order to minimize the time gap between proximity biotinylation and UV-crosslinking.

8. For cells growing on cover glasses (for immunofluorescence analysis) that will not undergo UV-crosslinking: Leave them in the last quenching solution wash while handling the other samples.

9. Decant quenching solution from cell plates for UV-crosslinking but ensure that cells are covered by a thin film of fluid to avoid them drying out. If necessary, add 750 μl of quenching solution to cells growing in plates, remove lids and crosslink cells with at least 0.15 J/cm² of >310 nm UV light.

   Note: mark cell plates and not only the lids, to avoid any switching of samples.

10. During crosslinking, aspirate the last quenching solution off the cover glasses that will be processed for immunofluorescence, and fix cells by fixative of choice. After fixation, wash cells 3 times in PBS, and store them in PBS at 4°C for later processing and imaging (see commentary/critical parameters and troubleshooting/fluorescence microscopy).

11. At completion of UV-crosslinking cover plates and return them on ice. Use a cell scraper to collect cells with the remaining quenching solution into pre-chilled 1.5 ml tubes. Pellet cells at 300 g, remove the supernatant, snap-freeze cell pellets in liquid N₂, and store at −80°C.

BASIC PROTOCOL 2

Cell extraction, streptavidin affinity purification, and on-beads trypsinization

This section describes Proximity-CLIP steps after cell lysis, all the way to affinity purification of the localized RNPs. Since isolated RNPs will be used as input for three different analyses, beads carrying the immobilized RNPs will be split at the end of this section. This section itself will only describe the processing of RNPs for protein analysis by mass spectrometry. The processing of immobilized RNPs for RBP footprinting is described in Basic Protocol 3. Processing of intact bound RNA is also detailed In Support Protocol 1.

Materials

Cell pellets produced in Basic Protocol 1, step 11

DEPC-treated double distilled water (will be referred as water)

RIPA buffer (see recipe)

100 mM PMSF in ethanol

Protease inhibitor cocktail without EDTA (cOmplete™, Mini, EDTA-free Protease Inhibitor Cocktail, Roche 04693159001)

Sodium ascorbate (MP Biomedicals 0210289025)

Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid) (Sigma 238813)

1 M sodium azide (Sigma S8032)

Pierce 660-nm protein assay reagent (Pierce 22660)

Ponceau S solution (Sigma P7170)

Streptavidin-coupled horseradish peroxidase (ThermoFisher Scientific, S911)

Streptavidin-coupled magnetic beads (ThermoFisher Scientific, 88817)

Magnetic rack for 1.5 ml tubes

KCl (Sigma, P9541), Prepare 1 M solution

Na₂CO₃ (Sigma, S2127), Prepare 0.1 M solution

Urea (Sigma U5378)

10 mM Tris-HCl pH 8

6x protein sample buffer (see recipe)

3× protein sample buffer supplemented with 2 mM biotin and 20 mM DTT

DMSO (Sigma D2650)

Iodoacetamide, single-use (ThermoFisher Scientific, A39271)

DTT (Dithiothreitol), No-Weigh™ Format (ThermoFisher Scientific, 20291)

Sequencing Grade Modified Trypsin (Promega V5111)

NH₄HCO₃ (Sigma 09830) Prepare 250 mM stock solution

RNase T1 buffer (see recipe)

Processing of cells from 6 cm plates for Western Blot analyses

See commentary/ critical parameters and troubleshooting/cells and controls.

• 1Defrost on ice cell pellets from 6 cm plates previously stored at −80°C and resuspend cells in 300 μl of cold RIPA buffer supplemented with 1 mM PMSF, 1x protease inhibitors cocktail, 10 mM sodium azide, 10 mM sodium ascorbate, 5 mM Trolox (sodium ascorbate and Trolox should be fresh and can be weighed and directly dissolved in the appropriate amount of RIPA buffer).
• 2Incubate on ice for 2 min and centrifuge at 15,000g for 10 min at 4°C to clear the extract from debris. Carefully transfer the supernatant without disturbing the precipitated debris into new pre-chilled tubes.
• 3Quantify protein concentration in the extracts using the Pierce 660-nm protein assay. For HEK293 cells, lysis of cells from a 6 cm plate in 300 μl should yield protein concentrations of around 1.2 μg/μl.
• 4Wash streptavidin-coupled magnetic beads twice with RIPA buffer. Then incubate 150 μg of each cell extract with ~15 μl (see commentary/critical parameters and troubleshooting/Streptavidin affinity purification comments) of washed beads under rotation overnight at 4°C or for 1 h at room temperature.

  Note: Confirm good mixing of beads and extracts during rotation. If necessary, increase the volume by addition of RIPA buffer. Store the remaining cell extract for Western blot analysis.

• 5Collect the beads using a magnetic rack and transfer the supernatant (flowthrough) into pre-labeled and chilled tubes. Store the supernatant for Western blot analysis.
• 6Wash beads with 1 ml volumes of the following ice-cold solutions: RIPA buffer (repeat twice), 1 M KCl, 0.1 M Na₂CO₃, 2 M urea in 10 mM Tris-HCl pH 8 (freshly prepared), and finish with two RIPA buffer washes.
• 7Elute in 60 μl 3x protein sample buffer supplemented with 2 mM biotin and 20 mM DTT, by incubation for 10 min at 97°C with vigorous shaking. To avoid rebinding after cooling, quickly spin the tubes in a table-top centrifuge to collect liquid from the caps and place them on the magnetic rack. Collect the eluates in fresh, pre-labeled tubes.
• 8Analyze the cell extracts, bead supernatants (flowthrough), and eluates by Western blotting using streptavidin-coupled HRP. Cell extracts should also be analyzed by Western blot for the APEX2-fusion protein, to confirm expression and probe for proteolytic cleavage events. Stain the nitrocellulose membranes by Ponceau S as a loading control.

Processing of cells from preparative plates

• 9Transfer cell pellets from the 15 cm plates from −80°C to ice and resuspend each sample in 800 μl of cold RIPA buffer supplemented with 1 mM PMSF, 1x protease inhibitors cocktail, 10 mM sodium azide, 10 mM sodium ascorbate, 5 mM Trolox (sodium ascorbate and Trolox should be fresh and can be weighed and directly dissolved in the RIPA buffer).
• 10Incubate on ice for 2 min and clear the extracts by centrifugation at 15,000g for 10 min at 4°C. Carefully transfer the clarified extract without disturbing the precipitated debris into new pre-chilled tubes.
• 11Quantify protein concentration in the extracts by Pierce 660-nm protein assay. With HEK293 cells, protein concentrations are expected to average around 3 μg/μl.
• 12Wash streptavidin-coupled magnetic beads twice with RIPA buffer. Then incubate 1.5 mg total protein of the cell extracts with ~60 μl of washed beads for 1 h at room temperature with rotation. Unused cell extract should be kept and stored: 30 μl for total RNA extraction and the remaining volume for Western blot analyses.
• 13Collect the beads using a magnetic rack and transfer the supernatant (flowthrough) into pre-labeled and chilled tubes.
• 14Wash beads with 1 ml volumes of the following ice-cold solutions: RIPA buffer (repeat twice), 1 M KCl, 0.1 M Na₂CO₃, 2 M urea in 10 mM Tris-HCl pH 8 (freshly made), and finish with two washes of RNase T1 buffer.
• 15Use the last wash to split the beads from each cell sample into 3 non-equal aliquots: 30% for mass spectrometric analysis. Reduce liquid volume to minimum and keep on ice. 20% for RNA-seq analysis of intact bound RNAs; collect beads on a magnetic rack, discard supernatant and store at −80°C (note that this will damage the magnetic beads). Keep the remaining 50% of the beads for RNA footprinting and small RNA cDNA library preparation (See Basic Protocol 3). It is necessary to directly proceed with the beads for RNA footprints (Basic Protocol 3) and proteomic analysis. We recommend proceeding first with Basic Protocol 3, steps 1–11, and then return to perform steps 16–20 of on-beads trypsinization below.

On-beads proteins trypsinization

(see commentary/critical parameters and troubleshooting/Mass spectrometry comments)

• 16Transfer the beads (30% aliquot kept on ice) to the magnetic rack and discard the remaining fluid. Resuspend the beads in 30 μl freshly prepared 25 mM NH₄HCO₃ and 20 mM DTT. Shake for 30 min at 25°C, then for 20 min at 37°C, and finally for 10 min at 56°C.
• 17Add 6 μl of 200 mM iodoacetamide in 25 mM NH₄HCO₃ and shake for 1 h at 25°C.
• 18Collect liquid from tube caps by briefly spinning in a top centrifuge and transfer the tubes to a magnetic rack. Discard the supernatant and wash the beads 3 times with 200 μl of 1 mM DTT in 25 mM NH₄HCO₃, to quench any remaining iodoacetamide and to ensure full depletion of NP40.
• 19Dissolve 20 μg of trypsin in 1 ml of 25 mM NH₄HCO₃.
• 20Add 98 μl of 25 mM NH₄HCO₃ and 2 μl of the trypsin dilution (40 ng) to the beads. Shake over night at 37°C; if possible, place thermomixer lid to minimize condensation on tubes lid.
• 21Collect liquid from tube caps by briefly spinning in a tabletop centrifuge and transfer the tubes to a magnetic rack. Collect the supernatant, which contains the proteolyzed peptides to clean tubes. Peptides are ready for submission to a mass spectrometry facility.
• 22Our facility performs Mass spectrometry on an Orbitrap Fusion coupled with an Ultimate 3000 nLC (Thermo). Peptides are separated on an EASY-Spray C₁₈ column (Thermo; 75 μm × 25 cm inner diameter, 2-μm particle size and 100-Å pore size), using 5–35% linear gradient of acetonitrile + 0.1% formic acid over 120 min. Electrospray voltage of 2.1 kV is applied to the eluent via the EASY-Spray column electrode. Orbitrap Fusion operates in positive ion data-dependent mode. We perform full-scan mass spectrometry in the Orbitrap with a normal precursor mass range of 350–1,500 m/z at a resolution of 120,000. The automatic gain control (AGC) target and maximum accumulation time settings is set to 4 × 10⁵ and 50 ms, respectively. MS² is triggered by selection of the most intense precursor ions above an intensity threshold of 5 × 10³ for collision-induced dissociation (CID)-MS² fragmentation with AGC target and maximum accumulation time settings of 5 × 10² and 250 ms, respectively. Mass filtering is performed by the quadrupole with a 1.6 m/z transmission window, followed by CID fragmentation in the ion trap (rapid mode) and a normalized collision energy (NCE) of 35%. To improve the spectral acquisition rate, parallelizable time is activated. The number of MS² spectra acquired between full scans is restricted to a duty cycle of 3 s.

BASIC PROTOCOL 3

RNA footprints cDNA library preparation

This section describes a non-radioactive streamlined small RNA cDNA library preparation including RNase digestion, dephosphorylation, phosphorylation, 3’ adapter ligation, retrieval of adapter ligated-crosslinked footprints by Proteinase K digestion, 5’ adapter ligation, reverse transcription and PCR amplification. This section has been modified from the original Proximity-CLIP protocol and does not include usage of radioactivity or urea polyacrylamide gel electrophoresis. Instead, size selection is performed on an agarose gel after reverse transcription and following an initial low cycle PCR amplification.

Materials

RNase T1 buffer (see recipe)

RNase T1 (1000 U/μL)

SUPERaseIn™ RNase Inhibitor (20 U/μl) (ThermoFisher Scientific AM2696).

10x Dephosphorylation buffer (based on the Calf Intestinal Phosphatase buffer, see recipe).

Calf Intestinal Phosphatase (NEB, Quick CIP, M0525)

T4 Polynucleotide Kinase (PNK) buffer without DTT: 70 mM Tris-HCl, 10 mM MgCl₂, pH 7.6.

T4 Polynucleotide Kinase with 10x buffer

Proteinase K buffer (see recipe)

Proteinase K, recombinant (Millipore Sigma, 3115836001)

15 mg/ml GlycoBlue™ (ThermoFisher Scientific, AM9515) or 15 mg/ml glycogen

Acidic phenol-chloroform (pH 4.5)

Water-saturated chloroform

Nucleic acid low binding 1.5 ml tubes

T4 RNA Ligase 2, truncated K227Q (NEB, M0351, with 10x buffer without ATP and 50% PEG-8000)

T4 RNA Ligase (10 U/μL) (ThermoFisher Scientific, EL0021, with 10x buffer containing ATP)

Low melting point agarose

SuperScript™ IV Reverse Transcriptase (ThermoFisher Scientific, 18090050)

Platinum™ Taq DNA Polymerase with 10x buffer (ThermoFisher Scientific, 10966018).

Pippin Prep and 3% Pippin gel cassettes (Sage Science, CSD3010)

DNA Clean & Concentrator-5 (Zymo Research)

Oligo Clean & Concentrator (Zymo Research)

Magnetic rack for 1.5 ml tubes

Agilent TapeStation with DNA D1000 and RNA ScreenTapes, or Bioanalyzer

Oligonucleotides

5’ adapter (RNA): GUUCAGAGUUCUACAGUCCGACGAUC

Barcoded 3’ adapter (DNA): App-NNTGACTGTGGAATTCTCGGGTGCCAAGG-L (Underlined barcode sequence can be any of TGACTG, ACACTC, ACAGAG, GCGATA, ATAGTA, TCATAG, App: 5’ terminal adenosine residue connected via a 5’,5’-diphosphate bridge to the 5’OH of the 5’ nucleotide; L: 3’ aminohexyl blocking group)

Reverse transcription primer: GCCTTGGCACCCGAGAATTCCA

Short 5’ PCR primer: GTTCAGAGTTCTACAGTCCGACGATC

5’ PCR primer: AATGATACGGCGACCACCGAGATCTACACGTTCAGAGTTCTACAGTCCGA

3’ barcoded PCR primers (Illumina indices underlined): CAAGCAGAAGACGGCATACGAGATCGTGATGTGACTGGAGTTCCTTGGCACCCGAGAATTCCA Underlined barcode sequence can be any of CGTGAT, ACATCG, GCCTAA, TGGTCA, CACTGT, ATTGGC, GATCTG, TCAAGT

RBP-Footprint library preparation

1. Transfer the beads (50% aliquot kept on ice) to the magnetic rack. Once beads are drawn to the side, remove and discard the remaining supernatant, and resuspend the beads in 100 μl RNase T1 buffer.

2. Add RNase T1 to a final concentration of 1 U/μl and incubate at 22°C for 15 min. Then immediately cool the reaction on ice.

3. Wash beads twice with RNase T1 buffer and once with 1x dephosphorylation buffer.

4. Resuspend the beads in 60 μl dephosphorylation reaction mix: 6 μl 10x dephosphorylation 6μl of 5U/μl Calf Intestine Phosphatase (CIP), 1μl of 10 U/μl SUPERaseIn, and 47 μl H₂O. Incubate at 37°C for 10 min with shaking.

5. Wash beads twice with 1 ml of dephosphorylation buffer.

6. Wash beads twice with PNK buffer without DTT.

7. Resuspend beads in 60 μl PNK mix: 6 μl NEB PNK 10x buffer (with DTT), 6 μl PNK 6 μl 10 mM ATP, 1U/μl SUPERaseIn.

8. Incubate at 37°C for 30 min with shaking.

9. Wash the beads 2 times with 1 ml of PNK buffer without DTT.

10. Resuspend beads in 60 μl 3’ adapter ligation mix: 1 μM 3’ adapter, 1x RNA ligase buffer, 18 μl 50% aqueous PEG-8000, 10 U Rnl2(1–249) K227Q ligase, 1U/μl SUPERaseIn (optional).

11. Incubate overnight at 4°C with shaking.

12. Wash beads twice with ligation buffer without DTT.

13. Elute the RNA footprints from beads by three-step Proteinase K digestions: (1.) Add 1.2 mg/ml Proteinase K in 200 μl of Proteinase K buffer. Incubate at 50°C with vigorous shaking for 30 min. (2.) To this solution, add 0.75 mg/ml Proteinase K in 150 μl Proteinase K buffer. Incubate at 50°C with vigorous shaking for 30 min. (3.) Add 0.75 mg/ml Proteinase K in 150 μl Proteinase K buffer. Incubate at 50°C with vigorous shaking for 30 min.

14. Collect liquid from tube caps by briefly spinning in a tabletop centrifuge and place in a magnetic rack. Transfer the supernatant, which contains the eluted RNA footprints, into a new 1.5 ml low bind microcentrifuge tube.

15. To extract the RNA, add 30 μl of 5 M NaCl and 500 μl acidic phenol-chloroform (pH 4.5) to the 500 μl of supernatant, vortex well and incubate for 10 min.

16. Centrifuge at 12,000g for 10 min and transfer ~500 μl of the top aqueous phase to a new 1.5 ml microcentrifuge tube.

17. Add 500 μl water-saturated chloroform, vortex well, and centrifuge at 12,000g for 10 min. Transfer the aqueous phase to a new 1.5 ml microcentrifuge tube containing ~10 μg of glycogen or GlycoBlue and mix.

18. Precipitate the RNA by adding 3 volumes of ethanol, incubating at −80°C for >1 h (possible stopping point), and centrifuging at >12,000g for 20 min at 4°C. Remove supernatant.

19. Wash once without disturbing the pellet by adding 1 mL 75% ethanol, centrifuging at >12,000g for 5 min at 4°C. Dry the pellet by leaving the tubes open on the bench for 5 min.

    Do not let the pellet over-dry - if the pellet overdried, it will not properly solubilize in water (next step).

20. Resuspend the pellet in 9.5 μl of water and add 8.5 μl of the 5’ ligation mix containing 6 μl 50% PEG-8000, 2 μl 10x RNA ligase buffer with ATP and 0.5 μl of 50 μM RNA 5’ adapter.

21. Incubate at 90°C for 1 min and return the tubes to ice. Once chilled, add 2 μl of T4 Rnl1 (1 μg/μl) and incubate at 37°C for 1 h.

22. Clean reaction mixture using Oligo Clean & Concentrator columns. Alternatively clean the sample using phenol/chloroform extraction followed by ethanol precipitation (see Supplementary protocol steps 10–15).

23. Elute from columns in 11 μl of water. If using ethanol precipitation in step 22, resuspend pellet in 11 μl of water.

24. For reverse transcription add 1 μl 50 μM RT primer, 1 μl 10 mM dNTPs. Denature the RNA at 75°C for 3 min, then reduce the temperature to 55°C and add 7 μl mastermix comprising 1μl of 100 mM DTT, 4 μl of 5x first strand buffer (Thermo), 1 μl Superscript IV, and 1 μl 40U/μl SUPERaseIn.

25. Incubate at 55°C for 30 min.

26. Dilute the cDNA by adding 60 μl water.

27. Perform a 5 cycle PCR reaction consisting of 10 μl 10 mM dNTPs each, 12 μl MgCl₂ (50 mM stock), 40 μl 10x Platinum Taq DNA Polymerase buffer, 2 μl of 100μM Short 5’ PCR primer, 2 μl of 100μM RT primer, 1.6 μl Platinum Taq DNA Polymerase, 40 μl of the diluted cDNA and 292.4 μl H₂O. Split reaction into 8 PCR tubes and set PCR with following parameters: 94°C for 2 minutes then 5 cycles of 94°C for 30 seconds, 50°C for 30 seconds, and 72°C for 15 sec followed by final extension of 1 min at 72°C.

28. Combine the 8 tubes, purify and concentrate the PCR reaction with DNA Clean & Concentrator kit.

29. Size select, for a range of 75–95 bp, on a 3% Pippin gel cassette. Alternatively, PCR products can be size selected based on a DNA ladder on a standard 3% agarose gel. Adjust your size-selected purified PCR product by adding H₂O for a final volume of 40 μl.

30. Use 10 μl of the size-selected low cycle PCR for calibration PCR by adding 10 μl 10x Platinum Taq DNA Polymerase buffer, 0.5 μl of 100 μM 3’ RNA Index Primer, 0.5 μl 100 μM Long 5’ PCR primer, 2.5 μl 10 mM dNTPs each, 3 μl MgCl₂ (50 mM stock), 0.4 μl Platinum Taq DNA and 73.1 μl H₂O.

31. Split the 100 μl reaction mix into 7 tubes and load on a PCR 19 cycles program. Remove one of the tubes after 7, 9, 11, 13, 15, 17, and 19 cycles, and load the PCR products on a 2.5% agarose gel for electrophoresis.

32. Image the gel to select the optimal number of PCR cycles, where amplification of the library is favored relative to that of linker-linker products derived from directly ligated 3’ and 5’ adapters. Amplified linker-linker products should be at 126 while small RNA cDNA libraries should be between 146 and 166.

33. Use the same reagents used for the calibration PCR to set up an identical PCR with a reaction volume of 300 μl using 30 μl of the size selected PCR as template. Split the reaction to 6 50 μl tubes and run with the previously determined optimal number of cycles.

34. Merge, clean and concentrate the PCR product using a standard column-based purification kit and elute in 30 μl of water.

35. Load 30 μl of the purified PCR product for a Pippin Prep size selection to deplete adapter-adapter ligation products (126 bp long) and enrich insert-containing library products.

    For footprints of lengths 20–40 nt, the expected library size is 146–166 bp.

36. Use TapeStation with a D1000 ScreenTape to measure the library final concentration and average length.

37. The library is now ready for Illumina sequencing. Dilute the library as required by your sequencing provider or facility.

Support Protocol 1

Preparation of RNA-seq libraries from intact RNA

This support protocol describes steps for transforming total cell extract RNA and RNPs immobilized on beads, (sections 12 and 15 respectively in Basic Protocol 2) into cDNA libraries ready for RNA-seq. Sequencing total extract RNA contributes insight on gene expression levels. Sequencing of the RNA component in immobilize dRNPs without an RNase treatment step represents full length RNAs that localize to the compartment of interest: Unlike cDNA libraries produced in Basic Protocol 3 from the short RNA footprints protected by UV-crosslinked RBPs, reads distribution from RNA-seq libraries do not give insight on precise RBPs binding sites, and T-to-C conversions cannot be used to filter experimental background. However, given the relative ease of RNA-seq library generation, we have learned that creating this additional data set can be useful: By adding confidence in localized transcripts that are detected both in full-length and as RBP-bound RNA footprints, by allowing comparison of the signal obtained by the two techniques per locus, increasing the confidence in RNA quantification. Other useful information that sequencing of the full-length RNA may yield can be identification of localized splice isoforms.

Additional materials

UltraPure™ Phenol:Chloroform:Isoamyl Alcohol (25:24:1, v/v).

Ribosomal RNA depletion kit (NEBNext rRNA depletion Kit or similar).

RNA Library Prep Kit (NEBNext Ultra™ II directional RNA Library prep Kit for Illumina or similar)

Multiplex Oligoes for Illumina (NEBNext Index Primers Set or similar)

Magnetic rack for 96-well plate

1. Retrieve the −80°C-stored 20% of beads for RNA-seq of intact bound RNAs (Basic Protocol 2, step 15).

2. Elute the bound RNA by three step Proteinase K digestions: (1.) Add 1.2 mg/ml Proteinase K in 200 μl of Proteinase K buffer. Incubate at 50°C with vigorous shaking for 30 min. (2.) To this solution add 0.75 mg/ml Proteinase K in 150 μl Proteinase K buffer. Incubate at 50°C with vigorous shaking for 30 min. (3.) Add 0.75 mg/ml Proteinase K in 150 μl Proteinase K buffer. Incubate at 50°C with vigorous shaking for 30 min.

3. Collect liquid from tube caps by briefly spinning on a tabletop centrifuge and place the tubes in a magnetic rack. Transfer the supernatant, which contains the eluted RNA, into a new 1.5 ml low bind microcentrifuge tube.

4. To extract the RNA, add 30 μl of 5 M NaCl and 300 μl of acidic phenol-chloroform (pH 4.5) to the 500 μl supernatant, vortex well and incubate for 10 min.

5. Centrifuge at 12,000g for 10 min and transfer ~300 μl of the top aqueous phase to a new 1.5 ml microcentrifuge tube.

6. Add 300 μl water-saturated chloroform, vortex well, and centrifuge at 12,000g for 10 min. Transfer the aqueous phase to a new 1.5 ml microcentrifuge tube containing ~10 μg of glycogen and mix.

7. Precipitate the RNA by adding 3 volumes of ethanol, incubating at −80°C for >1 h, and centrifuging at >12,000g for 20 min at 4°C.

8. Discard the supernatant and add 0.5 ml of 75% ethanol. Without mixing, centrifuge at >12,000g for 7 min, thoroughly remove the supernatant and air-dry the pellet by leaving the tubes open on the bench for 5 min.

   Do not let the pellet over-dry - if the pellet overdried, it will not properly solubilize in water (next step).

9. Resuspend the bound intact RNA pellet in 20 μl of water. Optional, use TapeStation and an RNA ScreenTape to analyze the size distribution and concentration of the eluted RNA. Store the RNA at −80°C.

10. Retrieve the 30 μl aliquots containing total cell extract saved for total RNA analysis (Basic Protocol 2, step 12).

11. To these aliquots, add 370 μl water and immediately after 400 μl of phenol:chloroform:isoamyl alcohol 25:24:1 mixture. Vortex for 15 sec, incubate on the bench for 15 min, and centrifuge at maximum speed for 10 min.

12. Transfer 200 μl of the top aqueous phase to a new tube, add the same volume of water-saturated chloroform, vortex and centrifuge at maximum speed for 10 min.

13. Transfer 100 μl of the top aqueous phase to a new tube, add 7 μl of 3 M sodium acetate pH 5.3, ~10 μg of glycogen and 400 μl of cold 100% ethanol, vortex, and incubate at −80°C for >1 h.

14. Spin at maximum speed for 15 min in a pre-chilled centrifuge. Discard the supernatant and add 0.5 ml of 75% ethanol. Without mixing, centrifuge at maximum speed for 7 min, thoroughly remove the supernatant and air-dry the pellet by leaving the tubes open for 5 min. Do not let the pellet over-dry.

15. Dissolve the total RNA pellet in 20 μl of water. Optional, use TapeStation and an RNA ScreenTape to analyze the integrity of eluted RNA. Store the RNA at −80°C.

16. Use the kits mentioned in “additional materials”, or other kits if preferred, to transform the total and bound intact RNA samples into cDNA libraries for RNA-seq. For the bound RNA samples avoid ribosomal RNA depletion and use half of the recommended time for RNA fragmentation.

Reagents and Solutions

RIPA buffer

50 mM Tris-HCl, 150 mM NaCl, 0.1% (wt/vol) SDS, 0.5% (wt/vol) sodium deoxycholate, 1% (wt/vol) Triton X-100. Adjust pH to 7.5 with HCl. Can be stored at 4°C for many months.

6x protein sample buffer

300 mM Tris-HCl pH 6.8, 30% glycerol, 6% SDS, 600 mM DTT, 0.01% bromophenol blue. Can be stored at −20°C for many months.

RNase T1 buffer

20 mM Tris pH 7.4, 150 mM NaCl, 2 mM EDTA, 1% NP40. Can be stored at 4°C for many months.

10x Dephosphorylation buffer (based on the Calf Intestinal Phosphatase buffer)

500 mM potassium acetate, 200 mM Tris acetate, 100 mM magnesium acetate, 0.1 mg/ml BSA, pH 7.9. Can be stored at 4°C for many months.

Proteinase K buffer

50 mM Tris pH 7.5, 75 mM NaCl, 6.25 mM EDTA, 1% SDS. Can be stored at room temperature for many months.

Commentary

Background Information

Proximity-CLIP relies on the well-supported assumption that most cellular RNAs are protein-bound throughout their life cycle, including transcription, processing, transport, translation and degradation (Choder, 2011; Khong and Parker, 2019). Proximity-CLIP combines cellular compartment-specific protein biotinylation (Hung et al., 2016) with photoreactive ribonucleoside-enhanced crosslinking to covalently and irreversibly crosslink RNA with RNA-binding proteins (RBPs) in intact cells (Hafner et al., 2010; Benhalevy et al., 2016) (Figure 1). Our approach enables determination of the localized proteome that includes RNA-binding proteins (RBPs) using mass spectrometry, and the profiling of localized transcripts using RNAseq. Importantly, Proximity-CLIP reveals hotspots of protein occupancy along the RNA sequence, which often relate to elements of crucial functional and regulatory significance and facilitates studying short-lived and degradation-susceptible RNAs due to the stabilizing effect of crosslinking to their associated RBPs. Our proof-of-concept experiments in HEK293 cells recapitulated many RNA-related phenomena that were previously studied using specialized approaches, and, most importantly, allowed us to generate hypotheses about RNA biology in the cell-cell interface (Benhalevy et al., 2018a; Koch, 2019).

The covalent bonds between biotin, RBPs, and RNA renders the RNP complexes resistant to stringent purification steps, maximizing the signal-to-noise ratio in the downstream high-throughput proteomic and transcriptomic analyses. The approach is fractionation-independent and allows for the isolation of compartments that are inaccessible to biochemical purification. Furthermore, the use of stringent extraction conditions promotes the preservation of the isolated cellular components. Finally, UV-crosslinking of 4-thiouridine (4SU)-labeled RNA to interacting proteins leads to a structural change at the photoreactive nucleoside, resulting in nucleotide misincorporation during reverse transcription and a characteristic T-to-C mutation in the corresponding cDNA libraries (Ascano et al., 2012). This feature allows for efficient computational removal of contaminating sequences derived from non-crosslinked fragments of abundant cellular RNAs, further increasing the specificity of Proximity-CLIP by reducing the false-positive detection rate.

Critical Parameters and Troubleshooting

The workflow of Proximity-CLIP comprises the following steps (Figure 1): (1.) 4SU labeling of RNAs in living cells expressing specifically localized APEX2 (engineered ascorbate peroxidase); (2.) biotinylation of APEX2-proximate proteins by incubation of cells with biotin-phenol, followed by activation of the peroxidase reaction with hydrogen peroxide (H₂O₂) for 1 min, and reaction quenching using sodium ascorbate, Trolox, and sodium azide; (3.) in-vivo crosslinking of RNA and proteins using UVA or UVB light ( λ > 310 nm) during the quenching step; and (4.) isolation of localized, biotinylated, and crosslinked ribonucleoprotein (RNP) complexes by affinity chromatography.

Cells and controls

The protocol is currently designed for adherent cells, and we worked on weakly adherent HEK293 cells that required additional caution during the numerous washing steps, to avoid loss of cells. In principle, adjustments could be made to fit cells in suspension. To employ Proximity-CLIP in other systems, either non-mammalian or non-cell-culture, the following requirements must be met: (1.) Cells need to be expressing APEX2 in a compartment-specific manner. (2.) 4-thiouridine or 4-thiouracil need to be taken-up by cells and metabolized into 4-thioUTP. (3.) Cells need to be accessible for UV-crosslinking and for administration of BP, hydrogen peroxide, and antioxidative quenching. Mediating proximity biotinylation by a biotin ligase such as TurboID (Branon et al., 2018) instead of APEX2 may be preferable in cases where a shorter labeling radius is required, which is particularly relevant in smaller cell systems, such as yeast and bacteria. It would also obviate the requirement for BP, hydrogen peroxide and the administration of quenching solution, but will add the requirement of accessibility for administration of biotin. Essentially, a good indication that a system is amenable to Proximity-CLIP would be the identification of successful PAR-CLIP and either APEX2- or biotin ligase-mediated labeling experiments in the literature.

In terms of cells density, to balance the need for high cell number with the requirement of cells to be rapidly growing for efficient 4SU incorporation we aimed for 90% confluence on the day of the experiment. This requires careful calibration per cell line. We aimed to have at least ~25×10⁶ HEK293 cells in a similar growth stage 36 hours prior to the experiment. This sufficed in our hands for seeding one 15 cm preparative plate, three 6 cm plates (+BP & hydrogen peroxide, and -BP and -hydrogen peroxide controls) for Western blot analyses, and four cover glasses (-BP and -hydrogen peroxide controls and two +BP & hydrogen peroxide glasses) for immunofluorescence analyses. The additional +BP & hydrogen peroxide glass is usually required for “no primary antibody” or another technical control.

The compartment serving as control should be selected based on where signal from the compartment of interest is expected to diffuse to; in most cases this would be the cytoplasm. APEX2 fusion proteins should be detectable by fluorescence microscopy, by immunostaining or by fusion to a fluorescent protein. Staining procedures should be optimized beforehand, and homogenic APEX2 expression as well as proper localization in all expressing cells should be verified. Multiple APEX2-fusion constructs are currently available on Addgene, including plasmids #107597 and #107596 submitted by our laboratory, which enable expression of V5-tagged nuclear and cytoplasmic APEX2, respectively, and also harbor the hygromycin resistance gene for cell selection.

Seeding three 6-cm plates per cell line enables biochemical control of the proximity biotinylation: (1.) Western blot analysis of cell extracts and eluates with HRP-coupled streptavidin controls for the dependence of the reaction in APEX2, BP and hydrogen peroxide, (2.) Analysis of the unbound material and eluates will confirm that biotinylated proteins were fully collected from the cell extracts and are concentrated on the beads, and finally, (3.) Western blot analysis to confirm the expression of the APEX2-fusion protein is vital to rule out expression of truncated versions of the protein.

4SU labeling

Nascent RNAs incorporate 4SU during transcription, therefore in most cases overcrowding of cells should be avoided to maintain a metabolically active culture. The 16 h labeling time window can be increased for slow growing cells, but the active concentration of cellular 4-thioUTP may be decreasing after more than 16 h. Alternatively, if focused on transient events, the labeling time can be decreased. However, note that the active concentration of nuclear 4-thioUTP may take some time to build up. Therefore, even for labeling transient and rapid nuclear events 2–4 h incubation with 4SU is usually required.

Size selection during Basic Protocol 3

Pippin Prep instrument and 3% Pippin gel cassettes are used for agarose gel size separation and extraction of the amplified sRNA cDNA sequencing library. This step is necessary because directly ligated 3’ and 5’ adapters (“empty library”) are a common side reaction that can overwhelm the library during PCR. If Pippin Prep is not available, it is possible to use standard 3% agarose gel electrophoresis to size-separate the two PCR products. An alternative to gel extraction of the amplified library would be to follow the following steps: (1.) Run the gel at 90 V for ~90 min. (2.) position the gel on a UV imager and manually carve a well, same size as a loading well, right below the band corresponding to the library. (3.) Return the gel to the running apparatus, lower the buffer height so it is just below the top of the gel, empty the liquid off the carved well and fill it with clean running buffer. (4.) Continue to run the gel for 5 min while every 1 min collecting the buffer from the well into a fresh tube and quickly replacing it with fresh buffer. (5.) Re-image the gel to validate that the library band no longer appears and use TapeStation and a D1000 ScreenTape to asses in which of the tubes the library eluted, and to measure its concentration and average length.

Fluorescence microscopy

Fluorescence microscopy imaging should be performed according to the standard procedures applied per lab, cells and experimental system. It is essential to image both the APEX2-fusion protein as well as biotinylated proteins in all conditions (labeled and -BP, -hydrogen peroxide and -APEX2 controls). Note that detection of the APEX2-fusion protein may vary after labeling due to self biotinylation. Biotinylated proteins can be labeled by fluorophore-coupled neutravidin or an alternative (we use Alexa-Fluor-647-coupled Neutravidin). The main goals of imaging are: (1.) to confirm APEX2 is expressed and localized as expected. (2.) to confirm biotinylation is dependent upon APEX2, BP and hydrogen peroxide. (3.) to probe how well localization of biotinylated proteins correlates to APEX2 localization.

Streptavidin affinity purification

It is recommended to calibrate the optimal ratio of beads volume to cell extract total protein for each APEX2-fusion protein expressing cell line (calibration could be incorporated into Basic Protocol 2 steps 1–8). While insufficient quantity of beads may result in loss of biotinylated material that remains in solution, too many beads can increase the experiment background. Only ~15% of bound material is eluted from streptavidin beads upon incubation with protein loading dye (see section 2.3 step 18 and section 3.3 step 7). Therefore, Western blot analysis of the unbound material after incubation of cell extracts with the beads may be the best indication for full depletion of biotinylated proteins from the cell extracts. Finally, when working with beads or resin, make sure either wide or cut tips are used to maintain their integrity.

Mass Spectrometry Comments

Although mass spectrometry is often outsourced, we recommend performing on-beads trypsinization and the preceding steps in-house. Due to relatively low amounts of input material, it is advisable to minimize the amount of added trypsin, to strictly avoid contamination of samples with skin proteins and to use high grade reagents. We used Pierce single-use/no-weigh iodoacetamide and DTT (ThermoFisher Scientific 90034 and 20291, respectively) and Sequencing Grade Modified Trypsin (Promega V5111).

Understanding the Results

Checkpoint in Basic Protocol 3

Avoidance of radioactive labeling results in loss of many of the checkpoints along the protocol. The main checkpoint remaining for whether a proper production of sequencing library is being achieved is the calibration PCR (steps 30–32 in basic protocol 3). Imaging the PCR product separated on an agarose gel will show the ratio between cDNA-containing library and empty linker-linker product, at increasing numbers of PCR cycles (Figure 2). [8 Copy Editor: Author query - “ What is a reasonable number of PCR cycles? If bands can’t be seen, what steps could have gone wrong?”]

Figure 2.

Agarose gel electrophoresis of pilot PCR products. Typically, desired products appear at approximately 160 bp and linker-linker products appear at 130 bp. Ideal cycle number for a large scale PCR in this example would be 12 cycles.

Data analysis

Proximity-CLIP provides 3 data sets that require distinct initial processing – Mass spectromentry data, RNA-seq data, and RNA footprints data. The initial processing of proteomic data should allow for label-free quantification of proteins captured in the compartment of interest as well as in control copartments. From there, more than a single way exists to approach proteomic data, but label-free quantification parameters should enable defining which proteins are enriched in the compartment of interest in a statistically-significant and reproducible manner. We initially relied on the proteomic facility for these analyses, and later learned to independently analyze the data using MaxQuant (Cox and Mann, 2008).

RNA-seq data produced in the support protocol includes two subsets – total RNA that reflects relative gene expression levels, and intact RNA isolated from the captured RNPs (without RNase T1 treatment). The initial processing of this dataset is standard and should result in proper alignment of reads to the genomic sequence and in assigning of reads to genes. There is a multitude of tools to perform this initial processing, and we recommend sticking to what is routinely performed by you or your facility.

Initial processing of RNA footprints data is performed using the PARalyzer program (Corcoran et al., 2011) that will align the reads to the genomic sequence, quantify T-to-C conversions, and define the coordinates of RBP binding sites (a.k.a. “clusters”) along the transcriptome. On top of the alignment file, PARalyzer will produce a spreadsheet that will indicate for each cluster, its coordinates, the number of reads and of reads with T-to-C conversions (which we consider crosslinked reads), and the functional genomic element the coordinates comprise. From that point the analysis is dependent on the biological questions, and can go in multiple ways. We found NGSplot (Shen et al., 2014) and metaplotR (Olarerin-George and Jaffrey, 2017) useful to ask questions on how RBP footprints in our compartments of interest distribute relative to known functional elements that we retrieved from other sources. We find ssHMM (Heller et al., 2017) and K-mers analysis useful in characterizing sequence and structure motifs of RBP footprints.

Finally, we strongly recommend investing in manual browsing of the reads alignment. Many sequence browsers enable visualization of the alignment files, we use IGV (Thorvaldsdóttir et al., 2013). The visual integration and free browsing of RNA-seq and RBP footprints data in different compartments is essential to discover new phenomena that become detectable by Proximity-CLIP. Some of the observations we obtained in lncRNAs loci are depicted in Figure 3.

Figure 3.

Comparative visualization of Proximity-CLIP and RNA-seq reads and of different compartments enables discovery of RNA phenomenae. A. The mRNA CNN3 and the lncRNA LOC729970 are encoded on opposite strands in adjacent loci. While RNA-seq data indicates their transcription imenates from distinct promoters, Proximity-CLIP data hints that their transcription might emerge from the same locus in the form of divergent transcription. B. KCNQ1OT1 is an antisense lncRNA of the gene KCNQ1. While RNA-seq data suggests KCNQ1OT1 is not expressed, Proximity-CLIP data indicated that its transcription is initiated but decays prior completion, and the integration of the two data sets suggests these partial transcripts are not stable. C. Comparison of cytoplasmic and nuclear Proximity-CLIP data suggests that for lncRNA RAB30-AS1 distinct splicing isoforms localize to each of the compartments. D. The lncRNA XLOC_012192 is encoded downstream of CDK12. Nuclear Proximity-CLIP data shows transcription occurs almost continuously through the space between the two genes, suggesting the lncRNA transcription might not be independently regulated. In addition, cytoplasmic Proximity-CLIP suggests that only the CDK12 mRNA is exported to the cytoplasm.

Time Considerations

Setting up: Two weeks to three months

Setup time is dependent on how much modification your favorite experimental system requires for adaptation to Proximity-CLIP. For example, if you already perform proximity biotinylation for proteomic analyses, the setup for Proximity-CLIP should be straightforward. Setup includes thorough reading and full comprehension of the protocol, establishment of cells expressing APEX2 localized to the compartment of interest and a control compartment, detection of the APEX2 construct and of biotinylated proteins by both Western Blot and fluorescence microscopy, obtaining equipment and reagents, and preparation of solutions and buffers.

Basic Protocol 1: Three days

Basic Protocol 1 comprises splitting the cells for the experiment, followed by 36 hours of growth. At some point during this time 4SU is added to the media (see above). Most of the following protocol steps comprise a single, short yet intense, work day. At the end of that day cells for Western Blot analysis and of the preparatory plate are frozen, and cells for immunofluorescence analysis are fixed and ready for further staining. We recommend completing the immunofluorescence analysis before proceeding to Basic Protocol 2.

Basic Protocol 2 and steps 1–11 of Basic protocol 3: 4 days

Steps 1–8 of Basic Protocol 2 include processing of 6 cm plates for Western Blot analysis, and we recommend full completion of the analysis prior to proceeding with processing of the preparative plates. Altogether these should take 2 days. The rest of protocol 2 takes additional 2 days of work that include also steps 1–11 in Basic Protocol 3.

Basic Protocol 3 steps 12-end: 3 to 7 days

Basic Protocol 3 comprises multiple consequtive molecular manipulations, with many possible stopping points. It can be rushed with longer work days, and fast submission of libraries for sequencing. However, it can be safely split to shorter work days, and often the submission of libraries for sequencing may take another day.

Support Protocol: 3–5 days

The support protocol comprises the release and extraction of intact RNA from captured RBPs (steps 1–9) as well as extraction of total RNA (steps 10–15). Steps 10–15 could be performed in parallel to steps 4–9 or on a separate day. Step 16 directs the usage of standard kits for RNA-seq library preparation, which should take 2 days of work. Finally, the submission of libraries for sequencing might take another day.

Data analysis: Weeks to several months

The initial processing of each of the data sets - Mass spectromentry data, RNA-seq data, and RNA footprints data (see “Understanding Results”), should take no more than 1–2 days. However, familiarization with each of the data sets, integration of the different data sets with each other and with previously available data, and tackling specific biological questions may last an indefinite time.