General considerations:
The ability to determine de novo synthesized (s4U labeled) transcripts by SLAMseq will depend on (1) the cellular s4U uptake kinetics, labeling time and s4U concentration, (2) the overall transcriptional activity of the cell type and (3) the library sequencing depth. These parameters need to be taken into account when designing a SLAMseq experiment, particularly when employing short s4U pulse labeling, where sequencing depth demands adjustments to the given cellular parameters.
We highly recommend testing optimal concentrations of s4U for the cell line of interest by assessing cellular toxicity (e.g. using CellTiterGlo® Luminescent Cell Viability assay, Promega after manufacturer’s instructions) and s4U incorporation rates (e.g. by HPLC, see below) prior to the sequencing experiment in order to meet s4U -labeling conditions that do not affect gene expression or cellular viability.
s4U Labeling of the cells (pulse/chase)
Protect s4U from light.
Pulse labeling:
Seed cells the day before the labelling experiment in a density that allows exponential growth for the duration of the experiment (50-80% confluency).
Prepare cell culture growth medium supplemented with non-toxic concentrations of s4U (i.e. 100 µM for mESCs; Note, this concentration can vary for different cell types).
Remove media from the cells and replace with s4U -containing media to start the pulse labeling.
Note, that regular exchange of fresh s4U-containing media (i.e. every three hours) can significantly enhance s4U incorporation.
Pulse/chase labeling:
After pulse labeling (see above), remove s4U -containing media from the cells.
Wash cells twice with 1 X PBS.
Add cell culture growth medium containing 100 X excess of uridine (compared to s4U concentration in the pulse) to cells.
Take off media at the time-points of interest and lyse the cells directly in TRIzol®.
Store samples at -80°C or directly proceed with RNA isolation.
RNA isolation
DTT is added during RNA isolation to keep the sample under reducing conditions.
Thaw lysate and incubate 5 min at room temperature
Add 200 µl chloroform per 1 ml of TRIzol®
Shake tube vigorously for 15 sec
Incubate at room temperature for 2-3 min
Spin down at 16,000 x g for 15 min at 4°C
Transfer aqueous phase to new tube
Add 1/100 volume of 10 mM DTT (0.1 mM final concentration), 1 volume 2-propanol and optionally 1 µl glycogen (20 mg/ml)
Vortex well
Incubate 10 min at room temperature
Spin down at 16,000 x g for 20 min at 4°C
Discard supernatant
Add 500 µl 75% EtOH and 5 µl of 10 mM DTT, vortex well
Spin down at 7,500 x g for 5 min at room temperature
Remove supernatant, let the pellet dry for ~5-10 min and resuspend in appropriate amount of water supplemented with 1 mM DTT (final concentration)
Incubate for 10 min at 55°C
Measure concentration by Nanodrop and freeze RNA at -80°C or proceed to the next step
Proceed with “Digestion to single Nucleosides” in order to prepare the samples for HPLC analysis and control s4U incorporation rates or to “Thiol modification” in order to alkylate s4U for subsequent library preparation and sequencing.
Optional: Digestion to single nucleosides and HPLC analysis
Note: Single nucleoside digestion and HPLC analysis is optional but highly recommended to confirm adequate RNA labeling conditions, a necessary requirement for a successful SLAMseq experiment. Because HPLC-analysis is less sensitive compared to SLAMseq, s4U -incorporation is only detectable at later labeling timepoints (i.e. ≥3 h after labeling with 100 µM s4U in mESC).
Prepare the reaction to digest and dephosphorylate total RNA to single nucleosides as described in Table 1.
Incubate overnight (≥16h) at 37°C
Add 6 µl 3 M NaOAc (pH 5.2), 150 µl ice-cold 100% EtOH and 30 µl 10 mM DTT (1 mM final concentration), vortex
Incubate 10 min at -80°C
Spin down at 12,500 x g for 5 min at room temperature
Transfer supernatant to a new tube
Add 30 µl 10 mM DTT and 270 µl ice-cold 100% EtOH, incubate for 10 min at -80°C
Spin down at 12,500 x g for 5 min at room temperature
Transfer supernatant to a new tube
Evaporate supernatant in speed-vac to complete dryness
Dissolve sample in 50 µl H2O. Store at -20°C if necessary.
Take 25 µl of the digested RNA sample and add 75 µl H2O
For the uridine and s4U standards, prepare the stock solutions from 2 mg/ml uridine and 0.1 mg/ml s4U stock solutions as described in Table 2.
Prepare the standard dilutions as described in Table 3.
- Load the samples and the standards on HPLC. HPLC is performed on a Supelco Discovery C18 (bonded phase silica 5 µM particle, 250 x 4.6 mm) reverse phase column. For running conditions refer to (Spitzer et al., 2014) and (Andrus and Kuimelis, 2001).
Thiol modification (Iodoacetamide treatment)
Prepare the reaction mix for each Iodoacetamide (IAA) treatment reaction as described in Table 4.
Incubate reaction at 50°C for 15 min
Stop reaction by quenching the reaction with 1 µl 1M DTT
Add 1 µl glycogen (20 mg/ml), 5 µl NaOAc (3M, pH 5.2), 125 µl EtOH 100%, vortex and precipitate for 30 min at -80°C
Spin down at 16,000 x g for 30 min
Wash with 1 ml 75% EtOH, vortex
Spin down at 16,000 x g for 10 min
Remove supernatant and let the pellet dry for 5-10 min
Resuspend in appropriate volume (5-10 µl) H2O
Proceed with RNA quality control and library preparation
Library preparation for High-Throughput Sequencing
For mRNA SLAMseq, we recommend 3´ end sequencing using the QuantSeq 3′ mRNA-Seq Library Prep Kit for Illumina (Lexogen) according to the manufacturer’s instructions (QuantSeq FWD kit).
Considerations for sample sequencing: As standard approach, we performed SLAMseq on an Illumina HiSeq 2500 machine. Sequencing depth needs to be adjusted to s4U incorporation, which is dependent on the experimental s4U labeling time, the cellular s4U uptake kinetics and the overall transcriptional activity of the cell type. As a rough estimate (based on experiments using mouse embryonic stem cells), we recommend for longer labeling times (i.e. >3h) 10-20 million and for short pulse labelling (i.e. <3h) 30-50 million reads per library for efficient quantification of s4U-labeled transcripts. We recommend sequencing reactions in single read 100 (SR100) mode, which recovers ~70% of all labeled transcripts in mESCs under steady-state labeling conditions (i.e. 100 µM s4U-labeling for 24 h).
Data analysis
In principle, any self-assembled bioinformatics analysis pipeline can be employed to recover and quantify T>C-conversion containing reads. Bioinformatics analysis in the publication associated with this protocol has been performed using Digital Unmasking of Nucleotide conversion-containing K-mers (DUNK) (see full description and availability from http://t-neumann.github.io/slamdunk). SLAM-DUNK (Neumann et al., manuscript in preparation) is a fully automated, modular and T>C-conversion-aware alignment software package streamlining SLAMseq data analysis. Briefly, SLAM-DUNK comes with several statistics and diagnostic plotting functions as well as a MultiQC (http://multiqc.info) plugin to make SLAMseq data analysis and integration available to bench-scientists.
SLAM-DUNK executes the following four main steps:
(1) SLAM-DUNK uses NextGenMap (Sedlazeck et al., 2013) to map the SLAMseq reads to a reference genome.
(2) Reads with a high number of mismatches and reads that map to more than one annotated 3′ UTR are discarded.
(3) SLAM-DUNK uses VarScan2 (Koboldt et al., 2012) to identify genuine SNPs between the reference and the sequenced genomes.
(4) SLAM-DUNK counts the number of T>C containing reads and estimates the fraction of labeled transcripts for each gene individually. SNP positions and read positions with a low base quality value are automatically excluded.
The final output of SLAM-DUNK reports for each transcript the genomic coordinates and the UTR region, the T content of the UTR region, the number of mapped reads (raw counts and norm. counts), the number of T>C conversion-containing reads, the coverage over T positions, the T>C conversion rate, and the number of multimapping reads.