Enriching s⁴U-RNA using methane thiosulfonate (MTS) chemistry

Abstract

Metabolic labeling of cellular RNA is a useful approach to study RNA biology. 4-thiouridine (s⁴U) is a convenient nucleoside for metabolic labeling because it is cell-permeable, incorporated into newly transcribed RNA, and the sulfur moiety provides a handle for biochemical purification. However, a critical step in the purification of s⁴U-RNA is the efficiency of the chemistry used to enrich s⁴U-RNA. Here, we present a protocol for s⁴U-RNA enrichment that includes efficient and reversible covalent chemistry to biotinylate s⁴U-RNA using the activated disulfide methane thiosulfonate conjugated to biotin (MTS-biotin), followed by enrichment on streptavidin beads. The efficiency of this chemistry reduces enrichment bias and requires less starting material, thereby expanding the utility of s⁴U to study RNA biology.

INTRODUCTION

Metabolic labeling is a useful tool to understand RNA dynamics without perturbing RNA transcription or processing. This protocol describes how to enrich 4-thiouridine-labeled RNA from HEK293T cells following metabolic labeling. First, cells are cultured using standard techniques, followed by the addition of 4-thiouridine nucleoside to the cell culture media. Labeling is stopped by removing culture media and lysing the cells in TRIzol reagent, which also prevents cellular RNases from degrading cellular RNA. The RNA is then be purified away from genomic DNA and proteins. DNA contamination may interfere with downstream analyses, and proteins contain cysteine residues whose thiols can react with the activated disulfide reagent). We have included an optional RNA shearing step at the end of the RNA purification that decreases nonspecific background in the enrichment. Reducing agents are present throughout the RNA purification protocol in order to prevent s⁴U-RNA oxidation and potential disulfide formation.

• The s⁴U-RNA is then reacted with an activated disulfide conjugated to biotin, which specifically biotinylates s⁴U. The biotinylated RNA is then enriched on streptavidin coated magnetic beads, nonspecific background is washed away, and s⁴U-RNA can be specifically eluted from beads by reducing the disulfide bond between biotin and s⁴U. In this way, biotin remains bound to the beads and the RNA eluent is collected. Finally, s⁴U-RNA enrichment is quantified relative to any nonspecific background and the enriched samples are prepared for downstream analysis (e.g., RT-qPCR or RNASeq). In addition, protocols for positive control in vitro transcribed RNAs that contain s⁴U are included to help researchers gauge the efficiency of s⁴U-RNA enrichment steps independently from the success of a metabolic labeling experiment. In summary, here we present protocols for enrichment of metabolically labeled s⁴U-RNA and in vitro transcribed (IVT) s⁴U-RNA. This protocol has been successful at enriching s⁴U-RNA from cells that have been treated with media containing s⁴U at concentrations ranging from 100 μM to 1 mM and at times ranging from minutes to days.

Strategic planning

• Extent of s⁴U incorporation: When enriching s⁴U-RNAs from a population, there are two related considerations: what fraction of the RNAs has any s⁴U, and of the RNA that contains s⁴U, what fraction of the uridine nucleotides has been substituted with s⁴U.

• Amount of starting RNA: This protocol is designed for quantities of total RNA ranging from 1-50 μg of metabolically labeled RNA.

• Length of input RNAs: This protocol is focused on the purification of long RNAs (>200 nt), but can be adapted for RNAs <200 nt (for example, miRNAs). For long RNAs, pre-shearing the RNA is not necessary but can improve enrichment levels.

• Handling of reagents and RNA: The s⁴U and s⁴U-RNA is light sensitive and can also oxidize upon handling. Without precautions, we have found that these properties can lead to low enrichment and poor reproducibility. This protocol provides handling suggestions to increase the integrity of the s⁴U-RNA prior to enrichment. Similarly, some of the chemicals used in this protocol can hydrolyze or oxidize, and so care should be taken to use fresh reagents and to store the stocks under appropriate conditions as described.

BASIC PROTOCOL

ISOLATION OF METABOLICALLY LABELED 4-THIOURIDINE-CONTAINING RNA USING ACTIVATED DISULFIDE METHANE THIOSULFONATE BIOTIN (MTSBIOTIN) FOLLOWED BY ENRICHMENT ON STREPTAVIDIN BEADS

There are a variety of approaches to chemically enrich different sub-populations of RNA. These techniques are particularly powerful for examining RNA turnover (Tani and Akimitsu, 2012), acute changes in transcription (Fuchs et al. 2014, Schwalb et al. 2016), and tissue specific expression (TU-tagging) (Miller et al 2009). Critical to the success of these approaches is the choice of chemical label and the chemistry used to enrich the labeled RNA. While there are multiple nucleosides that can be metabolically incorporated into cellular RNA, the non-canonical nucleoside 4-thiouridine (s⁴U) has several advantages compared to alternative chemistries such as 5-bromouridine and 5-ethynyluridine. For example, s⁴U is structurally very similar to uridine, displays low toxicity at concentrations necessary for s⁴U-RNA enrichment (even after prolonged treatments), and can be chemically modified using sulfur chemistry that is orthogonal to the canonical RNA nucleotides. Chemical approaches that have been used to enrich s⁴U RNA include mercury columns, chemical alkylation of the sulfur, and disulfide chemistry. Disulfide chemistry has the advantage that disulfide bonds formed with s⁴U can be easily reversed through reduction with relatively gentle reducing reagents. While the majority of reports use a pyridylthio-activated disulfide to retrieve s⁴U RNA, we have found these reagents to be inefficient (Duffy et al. 2015). As an alternative, methanethiosulfonate (MTS)-activated disulfides have also been found to modify s⁴U (Qin et al. 2003, Wunnicke et al. 2011) and we have developed conditions in which these reagents are efficient at modifying s⁴U-RNA for the purposes of enriching subpopulations of RNA (Duffy et al. 2015). In summary, MTS chemistry is efficient, reversible, and provides a facile means of enriching s⁴U-RNA.

This protocol describes steps to isolate cellular RNAs that have been metabolically labeled with the 4-thiouridine (s⁴U). First, cellular RNA is metabolically labeled by culturing cells in the presence of 4-thiouridine nucleoside for the desired amount of time. Total cellular RNA is isolated, and the 4-thiouridine-containing subpopulation of RNA is biotinylated using the activated disulfide MTS conjugated to biotin. The biotinylated RNAs are enriched on streptavidin resin and eluted from resin via reduction of the disulfide bond. This protocol yields enriched s⁴U RNA samples that are suitable for various analyses including RT-qPCR and RNA-Seq. Appropriate controls for this experiment include RNA from cells that have not been exposed to s⁴U, and input RNA prior to enrichment. The protocol also includes a description of how to make s⁴U-RNAs using in vitro transcription (IVT), and the use of these IVT s⁴U-RNAs as positive controls for enrichment. These protocols are similar to ones previously described (Duffy et al. 2015) with added tips and improvements for enriching long cellular RNAs from microgram amounts of starting total RNA. As many applications exist for this chemistry beyond the metabolic labeling described here, we conclude with a discussion of considerations and limitations when adapting this protocol for other uses.

Equipment and Materials

• Cultured cells (e.g., HEK293T cells; at least 1 × 10⁶)

• DMEM high glucose (Invitrogen, cat. no. 11965-092)

• Fetal bovine serum (Invitrogen, cat. no. 16000-044)

• Penicillin/streptomycin (Millipore, cat. no. TMS-AB2-C)

• 4-thiouridine (MP Biomedicals, cat. no. 0215213405

• TRIzol reagent (Life Technologies, cat. no 15596-026)

• Chloroform (AmericanBio, cat. no. AB00350-00500)

• Kimwipes (Kimberly Clark, cat. no. 730830)

• RNeasy Mini Kit (Qiagen, cat. no. Q74106)

• 100% ethanol (AmericanBio, cat. no. AB04010-00500)

• 10x TURBO DNase reaction buffer

• TURBO DNase (Life Technologies, cat. no. AM2238)

• DTT (VWR International, cat. no. EM-3860)

• DEPC-Treated Water (Life Technologies, cat. no. AM9906)

• HEPES free acid (DOT Scientific, cat. no. DSH75030-250)

• EDTA disodium salt (Sigma Aldrich, cat. no. E5134-500G)

• Dimethyl formamide (J.T. Baker, cat. no. 9344-13)

• MTSEA-biotin-XX (Biotium, cat. no. 89139-636)

• Chloroform: Isoamyl alcohol (Fisher Scientific, cat. no. 3160-450ML)

• Phase-lock gel tube, heavy 1.5 mL (5-Prime, cat. no. FP2302820)

• Glycogen (AmericanBio, cat. no AB00670-00020)

• Dynabeads MyOne Streptavidin C1 magnetic beads (Life Technologies, cat. no. 65001)

• Tris•HCl (Fisher Scientific, cat. no. T5941-500G)

• Sodium chloride (Sigma Aldrich, cat. no. S9888-2.5KG)

• Tween-20 (Mp Biomedicals Inc., cat. no. ICN19484180)

• β-mercaptoethanol (Sigma Aldrich, cat. no. M3148-25ML)

• Qubit RNA HS assay kit (Thermo Fisher Scientific, cat. no. Q32855)

• Ambion century plus RNA marker (Thermo Fisher Scientific, cat. no. AM7145)

• MAXIscript T7 transcription kit (Thermo Fisher Scientific, cat. no. AM1312)

• Cy5-CTP (GE Healthcare, cat. no. 25-8010-87)

• 4-thiouridine-5’-triphosphate (s⁴UTP) (Trilink Bio, cat. no. N-1025)

• SequaGel UreaGel 29:1 Denaturing Gel System (National Diagnostics, cat. no. EC-829)

• Gel cassettes 1.0 mm (Life Sciences, cat. no. NC2010)

• Nanodrop 2000c Spectrophotometer (Thermo Fisher Scientific, cat. no. ND-2000c)

• MagRack 6 magnetic stand (GE Healthcare, cat. no. 28948964)

• DynaMag 96 side (Life Technologies, cat. no. 12331D)

• Qubit 3.0 Fluorometer (Thermo Fisher Scientific, cat. no. Q33216)

• Rotator (Glas-Col, cat. no. 099A MR1512)

• Typhoon FLA 9500 (GE Healthcare)

Metabolic labeling of cells and isolation of total cellular RNA

In this section we describe the metabolic labeling of HEK293T RNA with s⁴U. The protocol describes the labeling of cells in a 6-well plate, and includes two time points and one control, with each sample in duplicate (six total, Figure 1A). This protocol can be adapted to enrich s⁴U-RNA from several different types of input RNAs from various sources. Labeling experiments with s⁴U have been described in several organisms and cellular systems including bacterial (e.g., Favre et al. 1986), yeast (e.g., Miller et al. 2011, Sun et al. 2012), insect (e.g., Burow et al. 2015), and mammalian (e.g., Dolken et al. 2008, Rabani, et al. 2011) cells. The experimental design can also be adapted to different time points or treatment conditions. The details of the metabolic labeling described here offer a guide to one successful labeling regime that provides strong enrichment.

Figure 1.

A) Structure of s⁴U and schematic of s⁴U-RNA metabolic labeling and enrichment. B) Flow chart of experiments that are generally performed on the first day of the protocol, which includes s⁴U metabolic labeling in HEK293T cells and RNA isolation. Dotted lines indicated optional steps for RNA shearing. C) Flow chart of the remainder of the enrichment, which includes s⁴U-RNA biotinylation and enrichment on streptavidin beads. Pink and blue boxes indicate fractions that are saved as flow through and elution, respectively.

1. Plate HEK293T cells in a 6-well plate at 3 × 10⁵ cells/well and allow the cells to recover in RPE1 media (without s⁴U) overnight.

We generally plate cells at a low enough density so that they will not reach confluence during the experiment. In this protocol, we expect the cells to reach ~80% confluence by the end of the experiment. While we generally do not split cells during treatment, s⁴U treatments have been successful through splitting and re-plating cells (including trypsin treatment).

• 2.Freshly dissolve solid s⁴U in water to 1 M (260 mg/mL). Dilute 14 μL of this 1 M s⁴U stock into 20 mL of RPE1 media (to 700 μM s⁴U final) and mix by vortexing for 10 seconds. Aspirate media from cells in all six wells. For wells 1-2, add 3 mL/well of s⁴U containing media. For the samples in wells 3-6, add 3 mL/well of media without s⁴U.

While we recommend freshly dissolving s⁴U solid prior to the experiment, we have observed enrichment when using stocks that were stored at −20°C for more than a week.

Successful experiments have been performed using a range of concentration of s⁴U in the media. Because of the efficiency of MTS chemistry, we have found lower concentrations (as low as 100 μM) of s⁴U provide sufficient enrichment. In other cases, s⁴U can be used up to to 1 mM---these higher concentrations may be advantageous for especially short treatments (Fuchs et al. 2014, Schwalb et al. 2016). One concern that has been raised about s⁴U is its potential for toxicity (Burger et al. 2013). We have found that at low concentrations of s⁴U (100 μM), the total RNA levels in cells are very similar between treated and untreated cells, even after 22 days of treatment (Duffy et al. 2015), consistent with others’ findings (Gregersen et al. 2014, Hafner et al. 2010), suggesting that any perturbations to the steady state RNA levels caused by these s⁴U treatment conditions are minimal.

• 3.After 1.5h of s⁴U labeling, remove media from cells in wells 3-4 and add 3 mL of s⁴U-containing media (described in step 2).
• 4.After the samples in wells 1-2 have been exposed to s⁴U for 2h total (and the samples in wells 3-4 have been exposed for 30 min), remove the media from all wells and immediately add 1 mL TRIzol. Pipet up and down five times and transfer the TRIzol samples to labeled 1.7 mL microfuge tubes.

Ideally the RNA purification will be performed immediately, but we have observed successful enrichment for RNA samples that have been frozen at −80°C for up to three months.

RNA isolation

• 5.Add 200 μL chloroform to each of the 1 mL TRIzol samples from step 4. Shake the tubes vigorously for 15 s and let sit for 2 min.
• 6.Centrifuge the tubes for 5 min at 12,000 × g, 4°C. Transfer aqueous phase (~500 μL) to new labeled tubes.
• 7.To each aqueous phase from step 6, add 0.5 mL of 100% isopropanol, 1 μL 100 mM DTT (final concentration: 0.1 mM DTT) and 1 μL RNase-free glycogen (5-10 μg). Incubate samples at room temperature for 10 min.

The s⁴U-RNA is light sensitive and prone to oxidation. While these steps can be performed under standard laboratory lighting, try to minimize the time of light exposure. The DTT is included to help minimize oxidation of the s⁴U.

• 8.Centrifuge samples at 12,000 ×g for 20 min at 4°C. Carefully remove the isopropanol from the RNA/glycogen pellet.
• 9.Add 1mL of 75% ethanol to the pellet, shake for 10 s and centrifuge at 12,000 × g for 10 min at 4°C.
• 10.Remove the ethanol completely from the RNA/glycogen pellet. To do so, first remove the majority of the ethanol with a P1000, then spin the tubes again on a counter top centrifuge. Use a gel-loading tip to remove the remaining ethanol. Let the pellet air-dry for 2 min by leaving the tube open under a Kimwipe cover.
• 11.Resuspend each pellet in 20 μL of RNase-free water. Measure the RNA concentrations using a Nanodrop spectrophotometer. Adjust the volume to bring RNA concentration to 200 ng/μl with water.

Generally we retrieve 10-20 μg of material from 10⁶ HEK293T cells after this step, which includes RNA as well as any contamination from DNA and free nucleotides that were carried through the TRIzol extraction and precipitation.

• 12.Add 10x Turbo DNase buffer to a final 1x concentration. Add 1 μL Turbo DNase (2U) per 10 μg of RNA. Invert the tube several times and concentrate the liquid in the bottom by centrifugation for ~5 s. Incubate the reaction at 37°C for 30 min.
• 13.Prepare and label six (one per sample) phase lock gel tubes by centrifugation at 10,000 × g for 30 sec according to the manufacturer's directions.
• 14.Adjust the volume of the RNA samples from step 12 to 315 μL with water. Add 35 μL of sodium acetate (3M stock, pH 5.5) and add samples to phase lock tubes. Then add 350 μL phenol:chloroform:isoamyl alcohol (ratio 25:24:1 and pH 7.7-8.3) to each phase lock tube. Shake the tubes vigorously for 15 s and let incubate at room temperature for 2 min. Centrifuge at 12,000 × g for 10 min at 4°C.
• 15.Transfer the aqueous phase into in a new tube. Repeat steps 7-10 to precipitate the RNA with isopropanol.
• 16.Resuspend the RNA pellet in 30 μL of RNase-free water.
• 17.Assay RNA quality by Nanodrop to calculate RNA concentration and also perform a UV-Vis scan for s⁴U concentration.

Generally this protocol yields 10-12 μg of RNA from 10⁶ HEK293T cells. s⁴U absorbs light at 334 nm. In principle this absorbance can be used to calculate the amount of s⁴U in the sample, but we have found that conventional Nanodrop is not sufficiently sensitive to allow accurate quantitation. However, this scan does provide a qualitative check to see that there is a local maximum at 334 nm.

RNA shearing (optional)

While this protocol successfully enriches s⁴U-RNA with unsheared RNA samples, we have found that RNA shearing can increase yields and decrease background. As most short-read RNA-seq protocols require shearing of the RNA samples as part of a library preparation, it is convenient to perform the shearing prior to enrichment if it does not conflict with experimental design.

• 18.Adjust the RNA samples from step 16 to 40 μL with water. Add 40 μL 2x fragmentation buffer and place sample at 94°C for exactly 4 min.
• 19.Quickly spin the RNA sample on a countertop centrifuge and immediately place on ice.
• 20.Add 20 μL of 250 mM EDTA (final concentration: 50 mM EDTA) to each sample, mix by vortexing and incubate on ice for 2 min.

Modified RNeasy MinElute Cleanup

• 21.Add 350 μL buffer RLT and 250 μL 100% EtOH to each RNA sample from step 16 or 20, mix well by pipetting. Apply these samples to RNeasy columns.
• 22.Centrifuge the columns 15 s at 12,000 × g, 4°C. Discard the flow through.
• 23.To each column, add 500 μL RPE buffer supplemented with 35 μL of 1% β-me (final concentration: 10 mM β-me).

Note that the addition of a reducing agent in this step is important to reduce any disulfides that have formed with the s⁴U. [See figure 2]

Figure 2.

Example of yields from the s⁴U-RNA enrichment described in this protocol. RNA from cells treated for 2h or 30 min with s⁴U was enriched using the main text protocol and assayed by Qubit. Percent input was calculated for enriched samples relative to 10% input standards for each sample. Bars indicate the average of two biological replicates, with filled circles indicating the yield from individual experiments.

• 24.Centrifuge the samples 15 s at 12,000 × g, 4°C. Discard the flow through.
• 25.Add 500 μL freshly prepared 80% EtOH and centrifuge the samples for 2 min at 12,000 × g, 4°C. Discard the flow through.
• 26.Switch the columns to new 2 mL collection tubes. Centrifuge the samples 5 min at maximum speed, 4°C.
• 27.Transfer column to a 1.7 mL microfuge tube, and add 14 μL RNase-free water. Centrifuge the samples 1 min at ≥12,000 x g, 4°C.

Biotinylation of s⁴U-RNA with the activated disulfide methane thiosulfonate (MTS) bitoin

• 28.Dilute solid MTSEA-biotin-XX (MW 607.7 g/mol) in dry DMF to 1 mg/mL (1.64 mM) to make a concentrated stock.

The MTSEA-biotin-XX stocks are stable at −20°C for at least 3 months.

• 29.
  In order to biotinylate the RNA samples, mix the following reagents in a 1.7 mL microfuge tube:

  • 2 to 5 μg of RNA from step 27
  • 1 μL 1 M HEPES, pH 7.4 (final concentration: 20 mM HEPES, pH 7.4)
  • 1 μL 0.5 M EDTA (final concentration: 1 mM EDTA)
  • Nuclease-free water to 40 μL total volume
• 30.Dilute 4 μL of the 1 mg/mL MTSEA-biotin-XX stock into 76 μL DMF (50 μg/mL, 82 μM final concentration) and mix by vortexing.
• 31.Add 10 μL of the MTSEA-biotin-XX solution from step 30 into each reaction from step 29 (results in the addition of 500 ng MTSEA-biotin-XX to each reaction, final concentrations 16.4 μM MTSEA-biotin-XX, and 20% DMF).

These conditions include excess MTSEA-biotin-XX and should be sufficient for up to 5 μg of RNA. For larger scale reactions with more RNA, we recommend increasing the volume of the reaction but retaining the concentrations of each component.

• 32.Cover the reactions with foil and incubate these reactions at room temperature in the dark for 30 min with rotation.

Remove unreacted MTS-biotin from RNA samples

• 33.Adjust the volume of each sample to 100 μL by adding 50 μL of water. Add the solutions to labeled phase lock tubes and add 100 μL chloroform:isoamyl alcohol to each tube.
• 34.Shake the samples vigorously for 15 s and let sit 2 min.
• 35.Centrifuge the samples for 5 min at 12,000 × g, 4°C. For each sample, transfer the aqueous phase (100 μL) to a new labeled 1.7 mL microfuge tube.
• 36.To each aqueous phase from step 35, add 350 μL buffer RLT and 250 μL 100% EtOH, mix well by pipetting, and apply to an RNeasy column.
• 37.Centrifuge the columns for 15 s at 12,000 × g, 4°C. Discard the flow through.
• 38.Add 500 μL buffer RPE to each column and centrifuge the columns for 15 s at 12,000 ×g, 4°C. Discard the flow through.
• 39.Add 500 μL freshly prepared 80% EtOH to each column and centrifuge the columns for 2 min at 12,000 × g, 4°C. Discard the flow through.
• 40.Switch the columns to new 2 mL collection tubes. Centrifuge samples 1 min at maximum speed, 4°C.
• 41.Transfer the columns to new microfuge tubes. Add 50 μL RNase-free water to the center of each column. Elute the RNA by centrifuging these samples for 1 min at ≥12,000 xg, 4°C. Proceed immediately to isolation of the biotinylated s⁴U-RNA.

Block streptavidin beads

This portion of the protocol can be carried out in parallel with the RNA biotinylation step, as steps 28-41 require ~1h to complete.

• 42.For six samples, aliquot 66 μL Dynabeads MyOne Streptavidin C1 beads into one 1.7 mL microfuge tube. Place the tube in a magnetic rack for 2 min and remove supernatant with a pipet.

Prepare 10 μL of beads per sample and include 10% extra volume.

• 43.Wash the beads twice by resuspending them in 500 μL nuclease-free water and mixing by pipetting up and down five times. Place the tubes in a magnetic rack for 2 min and remove supernatant with a pipet after each rinse.
• 44.Wash beads twice with 500 μL high salt wash buffer, mixing and capturing the beads as described in step 43.
• 45.Add 330 μL freshly made bead blocking buffer, resuspend the bead mixture completely by pipet, and incubate for 1h at room temperature. Place the tubes in a magnetic rack for 2 min and remove the supernatant with a pipet.
• 46.Wash the beads twice with 500 μL high salt wash buffer. Place the tubes in a magnetic rack for 2 min and remove the supernatant with a pipet.
• 47.Resuspend the beads in 660 μL of high salt wash buffer. Pipet up and down and immediately aliquot 100 μL of bead suspension into each of six PCR tubes in a PCR-strip.
• 48.Immediately before applying the RNA samples (step 49), capture the beads in a 96-well magnetic rack and remove the buffer by pipet.

Isolate s⁴U-containing transcripts with streptavidin beads

• 49.Add 5 μL high salt wash buffer to the 50 μL RNA solutions from step 34 and mix well. Add each sample to beads from step 48. Cover with foil and incubate the samples at room temperature for 15 min on a rotator at 30 rpm (or similar) to ensure the beads are mixing during the incubation.
• 50.Place the tubes in a 96-well magnetic rack for 2 min and remove supernatant with a pipet. Save the supernatant on ice as “Flow through”.
• 51.Wash the beads three times by resuspending them in 100 μL high salt wash buffer and mix by inverting the tubes five times rapidly. Quickly centrifuge the tubes. Place the tubes in a magnetic rack for 2 min and remove the supernatant with a pipet after each rinse.
• 52.Add 25 μL freshly made elution buffer to the beads, mix the beads by inversion as in step 51, wrap the tubes in foil, and incubate them at room temperature in the dark for 15 min with rotation as in step 49.

This elution buffer will reduce the disulfide bond that formed between biotin and 4-thiouridine, thereby eluting the s⁴U-RNA and leaving biotin bound to the streptavidin beads.

• 53.Quickly centrifuge the tubes and capture the beads in a 96-well magnetic rack for 2 min. Carefully retrieve the supernatant with a pipet and save this sample as “elution”.
• 54.Add another 25 μL elution buffer and immediately place tubes in a 96-well magnetic rack for 2 min. Remove the supernatant with a pipet and combine this elution for each sample with the corresponding elution from step 53. Place samples on ice.

Quantify s⁴U-RNA enrichment

Successful enrichment can be tested immediately after step 54 by examining the total amount of s⁴U-RNA retrieved in s⁴U treated samples in comparison to the untreated controls. We have found this enrichment is easily observed for long s⁴U treatments using a Qubit HS assay kit according to the manufactures directions. This analysis is presented to provide assay conditions and expected yields from the protocol above.

• 55.Prepare a 10% Input sample by diluting 200-500 ng RNA from step 29 (10% of RNA used in biotinylation) into 50 μL of elution buffer. Vortex and place on ice with samples from steps 50 (Flow through) and 54 (Elution).
• 56.Make a master mix of 3582 μL of Qubit buffer and 18 μL dye (199 μL buffer and 1 μL dye per sample, 18 samples total) and vortex for 10 s to mix.
• 57.Aliquot 190 μL of master mix into two Qubit tubes. Vortex both Qubit RNA HS standards for 10 s, then add 10 μL of standard 1 and standard 2 to each tube, respectively.
• 58.Aliquot 194 μL of master mix into another Qubit tube. Add 5 μL of elution buffer to the tube.
• 59.Aliquot 194 μL of master mix into the remaining 15 tubes. Add 5 μL of the appropriate sample (Input, Flow through, or Elution) to each tube.
• 60.Vortex all Qubit tubes for 3 sec, incubate at room temperature for 2 min, and measure the fluorescence in each tube by Qubit, starting with standards 1 and 2.
• 61.To estimate the yield, first subtract RFU of elution buffer sample from all sample RFU values. Then calculate percent input using the equation (RFU_Sample/RFU_Input)*10.

Calculate the yield of the experiment by percent input rather than Qubit concentration because the elution buffer gives slightly different readings than RNA in water. Expected yields from this experiment are 3-5% Input.

Prepare RNA samples for RNA-Seq or RT-qPCR

• 62.Adjust the RNA samples from steps 50 (Flow through), 51 (Elution), and 55 (10% Input) to 100 μL with water.
• 63.Purify and concentrate the RNA samples using the RNeasy Minelute protocol described in steps 21-27.

The RNeasy Minelute cleanup kit removes excess salts and DTT that could complicate downstream analysis. Following cleanup, samples can be assayed via qPCR or RNASeq, using standard procedures.

Enrich an in vitro-transcribed RNA ladder

This in vitro transcribed ladder is a useful positive and negative control for MTS-biotin enrichment independent of metabolic labeling in cells. We suggest including this control in any MTS-biotin enrichment experiment to monitor the efficiency of enrichment. Under these conditions we expect nearly quantitative retrieval of the positive control and retrieval of undetectable amounts of the negative control.

RNA ladder synthesis

This protocol follows the MAXIscript T7 transcription kit instructions, with the addition of modified nucleoside triphosphates to incorporate a fluorescent label (Cy5) and 4-thiouridine.

• 64.
  Label two 0.2 mL PCR tubes “+s⁴U” and “−s⁴U”. To the “+s⁴U” tube, add the following:

  • 1 μL Ambion century plus RNA marker template (1 μg/μL)
  • 2 μL 10x transcription buffer
  • 1 μL 10 mM GTP, ATP, and Cy5-CTP
  • 0.5 μL 10 mM UTP
  • 0.5 μL 10 mM s⁴UTP
  • 9 μL nuclease-free water
  • 2 μL T7 enzyme mix
• 65.
  To the “−s⁴U” tube, add the following:

  • 1 μL Ambion century plus RNA marker template (1 μg/μL)
  • 2 μL 10x transcription buffer
  • 1 μL 10 mM GTP, ATP, UTP, and Cy5-CTP
  • 9 μL nuclease-free water
  • 2 μL T7 enzyme mix
• 66.Invert both tubes three times vigorously and quickly centrifuge. Repeat once more.
• 67.Cover the samples in foil and incubate at 37°C for 1h.
• 68.Add 1 μL of TURBO DNase to samples and mix as in step 66.
• 69.Cover the samples in foil and incubate at 37°C for 15 min.
• 70.Prepare and label two phase lock gel tubes by centrifugation at 10,000 × g for 30 s according to the manufactures directions.
• 71.Using material from step 69, adjust the volume of the RNA samples to 150 μL with water and add samples to phase lock tubes. Then add 200 μL phenol:chloroform:isoamyl alcohol (ratio 25:24:1 and pH 7.7-8.3) to each phase lock tube. Shake the tubes vigorously for 15 s and let incubate at room temperature for 2 min. Centrifuge at 12,000 × g for 10 min at 4°C.
• 72.Transfer the aqueous phase into in a new tube. Adjust the volume to 200 μL with nuclease-free water.
• 73.Add 10 μL of sodium acetate (3M stock, pH 5.5; final concentration: 150 mM sodium acetate), 1 μL 100 mM DTT (final concentration: 0.1 mM DTT) and 1 μL RNase-free glycogen (5-10 μg). Mix well by vortexing.
• 74.Add 600 μL of ice cold 100% ethanol, mix well by vortexing, and incubate at -80°C for 20 min.
• 75.Centrifuge samples at 12,000 × g for 30 min at 4°C. Carefully remove the ethanol from the RNA/glycogen pellet.
• 76.Add 1 mL of 75% ethanol to the pellet, shake for 10 s and centrifuge at 12,000 × g for 10 min at 4°C.
• 77.Remove the ethanol completely from the RNA/glycogen pellet. To do so, first remove the majority of the ethanol with a P1000, then spin the tubes again on a counter top centrifuge. Use a gel loading tip to remove the remaining ethanol. Let the pellet air-dry for 2 min by leaving the tube open under a Kimwipe cover.
• 78.Resuspend each pellet in 40 μL of RNase-free water.
• 79.Assay RNA quality by Nanodrop to calculate RNA concentration and also perform a UV-Vis scan to qualitatively analyze the s⁴U concentration using the absorbance at 334 nm.

RNA ladder enrichment

• 80.Biotinylate and enrich 500 ng RNA ladder using steps 28-54 of the main text protocol.
• 81.Adjust the RNA ladder from steps 50 (Flow through), 51 (Elution), and 55 (10% Input) to 100 μL with water.
• 82.Purify and concentrate RNA ladder samples using the RNeasy Minelute protocol in steps 21-27.

Visualize RNA products by urea-PAGE

• 83.
  Using the SequaGel UreaGel 29:1 Denaturing Gel System, make a 5% urea gel by combing the following in a 15 mL falcon tube:

  • 14 mL system diluent
  • 4 mL system concentrate
  • 2 mL system buffer
  • 80 μL 10% (w/v) ammonium persulfate
  • 4 μL TEMED
• 84.Mix well by inverting the tube and quickly pour into a 1.0 mM gel cassette and insert a 10-well comb. Let the gel polymerize for 30-60 min.
• 85.Load the gel into a gel electrophoresis apparatus containing 1x TBE buffer and pre-run the gel for at least 15 min at 200V.
• 86.While pre-running the gel, add 7 μL of each RNA ladder sample to 7 μL of urea gel loading buffer and mix well by pipetting.
• 87.Heat samples to 80°C for 5 min, then load 14 μL sample into each well using a gel loading tip. Wash the wells with a syringe prior to loading.
• 88.Run the gel for 45 min at 200V. Visualize RNA ladder on a Typhoon FLA 9500 instrument using Cy5 fluorescence filters (PMT: 1000).

REAGENTS AND SOLUTIONS

Use nuclease-free water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2

RPE1 medium

• 500 mL DMEM high glucose

• 50 mL fetal bovine serum

• 5 mL penicillin/streptomycin

RNA fragmentation buffer, 2x

• 150 mM Tris pH 8.3

• 225 mM KCl

• 9 mM MgCl₂

High salt wash buffer

• 10x stock solution contains:

• 100 mM Tris, pH 7.4

• 10 mM EDTA

• 1 M NaCl

• Add 0.05% Tween-20 fresh to 1x solution each time

• Store at room temperature up to 6 months

Bead blocking buffer

• 1x high salt wash buffer

• 40 ng/μL glycogen (20 μg/μL stock solution)

Elution buffer, 1x

• 100 mM DTT

• 20 mM HEPES, pH 7.4

• 1 mM EDTA

• 100 mM NaCl

• 0.05% Tween

• Prepare solution fresh each experiment

1x TBE

• 100 mM Tris Base

• 100 mM Boric Acid

• 2 mM EDTA

• Store at room temperature up to 6 months

Urea gel loading buffer

• 990 μL deionized formamide

• 10 μL 500 mM EDTA

• Store at room temperature up to 6 months

COMMENTARY

Background Information

Traditional RNA-Seq methods allow the quantification of RNA steady-state levels in cells. In contrast, the study of RNA dynamics is enabled through biochemical enrichment for different sub-populations of RNA prior to RNA-Seq. Metabolic labeling with non-canonical nucleosides is a widely used technique to separate distinct RNA populations. Ideally, the nucleoside should be absent from endogenous RNA, readily incorporated into cellular RNA, and provide a handle for biochemical purification following metabolic labeling.

To this end, numerous studies have employed 5-bromouridine (5-BrU; Tani et al., 2012), 5-ethynyluridine (5-EU; Jao and Salic, 2008), and 4-thiouridine (TU or s⁴U; Cleary et al., 2005 and Miller et al., 2009). 5-BrU-labeled transcripts can be purified by anti-BrdU antibodies, which also recognize BrU, but this approach is limited by the efficiency of antibody enrichment and nucleoside uptake. 5-EU can be used for isolation of nascent transcripts from whole cells, and click chemistry can be used to covalently enrich metabolically labeled RNA. However, 5-EU introduces a greater structural perturbation, and prolonged exposure to 5-EU causes cell toxicity, raising challenges for experiments that require longer treatments.

In contrast, 4-thiouridine can be introduced into cell culture media and is readily incorporated into cellular RNA. Under the conditions listed here, we have not detected significant toxicity even after prolonged exposure, making s⁴U suitable to measure acute changes in transcription and long-lived transcripts in the same experimental set-up. In addition, unlike antibody-based approaches, the biotin enrichment allows for stringent washes to decrease nonspecific background and the disulfide chemistry is reversible, allowing for elution of RNA from streptavidin beads followed by standard RNA-Seq library preparation methods. While MTS chemistry is specific to s⁴U when using isolated RNA, MTS is not specific to s⁴U in the context of a cell or cell extract because the activated disulfide reacts with other thiols, including cysteines on proteins. Therefore the MTS chemistry described here is only appropriate for isolated RNA. In order to purify metabolically labeled RNA in a more complex environment, alternative chemistry such as 5-EU, which displays orthogonality to other biomolecules, would be more suitable.

For studying RNA turnover, and other applications where different RNA populations need to be separated from each other (e.g., TU-tagging) MTS-biotin increases labeling yields and decreases enrichment bias. Therefore this MTS chemistry expands the utility of s⁴U metabolic labeling by decreasing the number of s⁴U's required to reliably capture a given RNA, the amount of starting material required, and the concentration of s⁴U required during cellular labeling.

Critical Parameters

s⁴U incorporation

The enrichment will only be successful when the RNA contains sufficient levels of s⁴U. In metabolic labeling experiments, incorporation of s⁴U into RNA can be controlled by two factors: concentration of s⁴U during cell treatment and time of s⁴U exposure. Insufficient s⁴U incorporation can lead to many transcripts without s⁴U, and therefore will lead to low yields and less enrichment over background. Insufficient labeling can also favor enrichment of longer transcripts that have more uridine residues (and therefore a greater probability of s⁴U incorporation).

RNA quality

4-thiouridine is light sensitive, so RNA should be kept in the dark as much as possible throughout the protocol. In addition, s⁴U oxidation occurs during RNA cleanup, which is why DTT or β-me should be included during these steps of the protocol. Oxidation decreases the amount of s⁴U available to react with MTS-biotin, which can decrease yields and introduce biases.

Control samples

RNA from cells that have not been labeled with s⁴U serves as a negative control to assess nonspecific background. We routinely reserve 10% of an input sample to quantify RNA yields and calculate fold enrichment of s⁴U-RNA relative to the negative control. A useful positive control is an in vitro transcribed RNA with s⁴U, which should be captured with nearly quantitative efficiency.

Next-generation RNA sequencing

If s⁴U-RNAs are to be analyzed by next-generation RNA sequencing, samples can be depleted of rRNA (preferred) or selected for polyA-RNA. In order to compare RNA enrichment between two samples, we recommend adding an exogenous spike to the enriched samples from step 63, using the same amount of spike per sample. Examples of normalization spikes include RNA from S. pombe or synthetic RNA standards. Normalization spikes can be used to quantify fold enrichment of entire samples or individual transcripts relative to input.

Troubleshooting

Suggestions for overcoming or avoiding commonly encountered problems

Step 32

Problem: Inefficient RNA biotinylation

• Possible reason: The MTS biotin has hydrolyzed.

  • ○ Solution: Store MTS-biotin stocks in dry DMF and replace stocks every 3-6 months.
• Possible reason: s⁴U has oxidized.

  • ○ Solution: Pre-aliquot s⁴U solid and dissolve fresh in water before each experiment.
• Possible reason: The biotin-MTS-DMF stock is not mixed well with the RNA.

  • ○ Solution: Vortex the solution well after adding MTS biotin and make sure that the solution is mixed in the tube rotator. Try changing the mode or speed of rotation until the solution is constantly mixing within the tube.
• Possible reason: s⁴U-RNA is oxidized.

  • ○ Solution: Use fresh DTT and β-me stocks for all RNA cleanup procedures in order to pre-reduce oxidized products (e.g., s⁴U-s⁴U disulfides). Use the IVT ladder in the supplemental protocol as a positive control by cleaning up RNA in parallel to cellular RNA samples in steps 7-16 and 21-27. Run cleaned up IVT ladder samples on a gel and look for slower migrating species that may be indicative of disulfide formation.
• Possible reason: s⁴U has been photocrosslinked.

  • ○ Solution: Protect RNA from light during all incubation steps and turn off all unnecessary benchtop lighting while working with RNA. Use the IVT ladder in the supplemental protocol as a positive control. Photocrosslinked IVT ladder may cause an upshifted smear when run on a gel.

Step 61

Problem: Low RNA recovery

• Possible reason: RNA starting amounts are too small.

  • ○ Solution: Measure the concentration by NanoDrop in step 41. The concentration should ideally be equal to the starting concentration in step 29. If the amount of starting material in step 29 is too low, there may still be RNA enrichment but it is below the limit of detection of the instrument. These samples can still be assayed by more sensitive assays including RNA-Seq.
• Possible reason: Inefficient elution.

  • ○ Solution: Dissolve DTT fresh to make elution buffer. If necessary, repeat elution step with an additional 25 μL of elution buffer and pool with elutions 1 and 2.
• Possible reason: Inaccurate measurement by Qubit.

  • ○ Solution: Repeat measurement with fresh regents or use alternative method (e.g., qPCR for expected transcripts).
• Possible reason: RNA has been exposed to RNase.

  • ○ Solution: Clean your work area, pipettes and gloves to remove RNases. Prepare new solutions if needed.

Step 61

Problem: High background in negative control

• Possible reason: Beads were insufficiently blocked in step 45.

  • ○ Solution: Make sure beads are fully resuspended in blocking buffer and make sure that the beads are constantly mixing in a rotator during the blocking step. Try changing the mode or speed of rotation until the solution is constantly mixing within the tube.
• Possible reason: Beads are not washed properly in step 51.

  • Solution: Make sure beads are fully resuspended in between washes and that supernatant is not trapped in tube cap or at the bottom of the tube. Make sure that the beads do not dry out between washes.

Anticipated Results

Before proceeding to downstream analysis, it is important to assay the yield relative to input and fold enrichment over background for each sample. Figure 2 represents the expected yield for sheared and unsheared RNA samples after 2h or 30 min of s⁴U treatment as well as a no s⁴U control. Sheared samples give slightly better yield and lower background compared to unsheared samples, which is why we prefer shearing if the downstream analysis permits. 2h s⁴U treatment is expected to give RNA yields of 4-5% input, whereas 30 min s⁴U treatment is expected to give RNA yields of 2-3% input. The yield of the negative control samples is limited by the Qubit assay, so it is difficult to accurately determine the fold enrichment, but greater than 10-fold enrichment is expected.

The in vitro transcribed ladder serves as a useful positive and negative control for s⁴U RNA enrichment. Following our supplemental protocol, we would expect near complete capture of the +s⁴U-RNA ladder, with undetectable levels of the −s⁴U-RNA ladder. These samples can be enriched in parallel to cellular RNA samples to ensure the robustness of the protocol.

Time Considerations

As presented, this protocol requires 2 days of experiments, with 8.5h on day 1 and 3.5h on day 2. Once the cells are plated, the s⁴U treatment requires 2.5 h. Following cell harvest, RNA isolation from cells and DNase treatment and optional shearing takes 6 h, Starting on day 2, RNA biotinylation takes 30 min. Removal of excess biotin takes 30 min and binding to streptavidin beads takes 15 min if bead blocking is performed in parallel. The streptavidin bead wash and RNA elution steps should take no more than an hour, at which point samples can be stored at -20°C or proceed to downstream analysis. Therefore, the total amount of time required for the RNA isolation from the start of the s⁴U treatment is 8.5h, and the enrichment steps from isolated RNA requires 3.5h.