| Experimental protocol | Total RNA was isolated from either cell pellets or dissected tissue by extraction with TRI Reagent (Sigma-Aldrich). The cell pellet (∼2 000 000 cells) or tissue (∼50 mg) was suspended in 2 ml TRI Reagent and thoroughly mixed, in case of tissue also crushed with a pestle. Following addition of 400 μl chloroform (Carl Roth) and thorough mixing, samples were centrifuged at 18 000 g for 15 minutes, the aqueous phase transferred to a new tube and 1 ml 2-propanol (Carl Roth), as well as 1 μl Glycogen (Thermo Fisher Scientific), were added. Following another round of centrifugation at 18 000 g for 15 min, supernatant was removed and 1 ml ethanol (75%) (Thermo Fisher Scientific) was added for a final centrifugation step. Finally, supernatant was carefully removed, the RNA pellet briefly airdried and resuspended in RNase free water. The required amount of TRI Reagent varied depending on the number of cells or tissue size used, with respective changes to the other Reagents that were added. The E. coli wildtype strain Keio parent (BW25113) and the knockout strains ΔdusA and ΔdusB were purchased from the E. coli Keio knockout collection (GE Healthcare (DharmaconTM, England). E. coli cultures were grown in 100 ml LB medium (Carl Roth) at 37°C and 190 rpm until an optical density of 0.4 at 600 nm. Paraquat dichloride hydrate (Sigma-Aldrich) was added to final concentrations of 0, 0.1 and 0.3 mM and bacterial growth was continued until an OD600 of 0.7. Chloramphenicol (100 μg/ml, Carl Roth) was added to the culture which was incubated for 3 min and then harvested by centrifugation (10 min, 10 000 g, 4°C). The pelleted cells were resuspended in buffer (100 mM NH4Cl (Merck, Darmstadt, Germany), 10 mM MgCl2 (Carl Roth), 20 mM Tris, pH 7.5 (Carl Roth)), lysozyme (Carl Roth) was added and freeze-thaw cycles in liquid nitrogen were performed. Lysis was completed by adding 10% deoxycholate (Sigma-Aldrich) and cell wall debris was removed by centrifugation (12 000 g, 10 min, 4°C). Cell lysate was loaded on top of sucrose gradients from 5 to 40% using Biocomp (BioComp, Fredericton, Canada) gradient station model 108 (settings: time 1.23 min, angle 81.5 °, speed 21 rpm). Gradients were ultracentrifuged (150 000 g, 4°C, 2.5 h, Beckman Optima MAX-XP Ultracentrifuge, SW40 Ti rotor from Beckman Coulter, CA, USA) and fractionated collecting free RNA fraction (F0) and polysomal fraction (F3). Total RNA was extracted using TRI Reagent (Sigma-Aldrich) and separated subsequently on a 10% denaturing PAGE gel stained with GelRed (Biotium/BIOTREND, Köln, Germany). tRNA bands were excised from the gel after visualization of RNA bands using Typhoon 9400 (excitation wavelength of 532 nm, Amersham Bioscience/GE Healthcare, Chicago, IL, USA) and mashed with a scalpel. tRNA was extracted from the gel adding 300 μl 0.5 M ammonium acetate (Merck) before overnight incubation at 25°C and 750 rpm. After filtration through NanoSep 0.45 μM spin columns (VWR, Darmstadt, Germany), RNA was precipitated using three volumes of 100% ethanol (Carl Roth). cDNA oligonucleotides for quantification (cDOQs) were designed from tRNA references retrieved from tRNAdb, MODOMICS, and gtRNAdb. The last 40 nucleotides towards the tRNA 3′ end were compiled to non-redundant references utilizing a R-based construction pipeline by Pichot et al., collapsing sequences with a levenshtein distance below six nucleotides into one species. Next, the reverse-complement of these retrieved, non-redundant references was merged with the respective P5 (AGACGTGTGCTCTTCCGATCT) and P3 (GATCGTCGGACTGTAGAACTCTGAAC) sequence on the 5′ and 3′ end. The full-length constructs of 87 nucleotides were finalized with either a 6-FAM (cytosolic tRNAs) or Cyanine-5 (mitochondrial tRNAs) attached to the 5′ end. All cDOQs were ordered and synthesized at biomers.net GmbH (Ulm, Germany) or Integrated DNA Technologies (Coralville, IA, USA). For indexing PCR, twelve i7 and eight i5 primers were designed to allow multiplexing up to 96 samples. To reach full compatibility with Illumina platforms, a custom primer for the i5 index read on Illumina NextSeq platforms was designed and utilized on the sequencing platform. All Primers were ordered and synthesized at biomers.net GmbH. Starting from 50–200 ng of total RNA or 5–50 ng of total tRNA, 5x hybridization buffer (150 mM HEPES pH 7.5 (Carl Roth), 500 mM potassium acetate (Carl Roth)) was added. Following addition of a 5-fold excess of an equimolar mixture of cDOQs for the selected targets (e.g. a single tRNA or 43 cytosolic + 22 mitochondrial tRNAs in human), the solution was denatured at 94°C for 2 min and afterwards gradually cooled down to 25°C within 20 min. The sample was then analyzed on a 10% native polyacrylamide-gel in 1x TBE buffer, running at constant power of 60 mA for 45 min. Depending on the experimental context, optional GelRed-staining (Biotium, Fremont, CA, USA) was carried out for 15 min to visualize nucleic acids. cDOQ-attached fluorescent dyes used for stainless visualization comprised 6-FAM or Cy5. Bands of interest (e.g. cDOQ:tRNA hybrid) were excised, crushed and eluted overnight in 300 μl 0.5 M ammonium acetate (Merck) at 750 rpm and 15°C. Afterwards, gel particles were removed with Nanosep 0.45 μM spin columns (VWR), centrifuging at 2000 g for 1 min. Finally, 1 μl glycogen was added to the aqueous flow-through, together with 2.5 volumes of ethanol (99,5%, Carl Roth), storing the mixture at −80°C for 1 h or −20°C overnight. Afterwards, samples were centrifuged at 18 000 g for 1 h, supernatant removed, the pellet washed with 500 μl ethanol (70%) and again centrifugated at 18 000 g for 30 min. Supernatant was carefully removed, the pellet left to air-dry for 10 min at room temperature and then resuspended in 10 μl of RNAse-free water. Nucleic acids from excised cDOQ:tRNA hybrids were subjected to index PCR, using half (5 μl) of the previously purified product. For samples with low tRNA input (5 ng or less), complete use of retrieved product was necessary. From a master mix, Taq-Buffer (NEB, Frankfurt am Main, Germany), dNTPs (NEB) and MgCl2 (Carl Roth) were added to final concentrations of 1× (Buffer), 0.5 mM (dNTP) and 3 mM (MgCl2). Individual i5 and i7 primer combinations were added to a final concentration of 0.25 μM. Reaction was carried out with Taq-Polymerase (NEB), initially denaturing at 94°C for 30 s, followed by normally 3–6 cycles of denaturation (94°C, 15 s), annealing (62°C, 30 s) and extension (72°C, 15 s), as well as a final extension step at 70°C for 5 min. Next, the reaction mixture was purified on a denaturing 10% PAGE in 1× TBE buffer, running at constant power of 60 mA for 1 hour, excising the product band around 169 nt length. Following overnight elution, ethanol precipitation and resuspension as described previously, samples were sequenced on an Illumina NextSeq 2000 or MiSeq platform. For sequencing on the NextSeq 2000 platform, a custom i5 index read primer was used. The achieved read output proved to be dependent on the number of different samples pooled on a flow cell, as well as the amount of sequencing-ready library loaded for each sample. Thus, for samples with low tRNA input (5 ng or less), yielding less DNA in the low-cycle index PCR, adjustments in library amount used and the overall sample number loaded on the flow-cell had to be considered. Library preparation of tRNA fractions was based on the NEBnext Small RNA Library Prep Set for Illumina (NEB) and is briefly described below. First, tRNA was dephosphorylated with Antarctic Phosphatase (NEB) and afterwards phosphorylated with Polynucleotide Kinase (NEB). Following purification via RNeasy MinElute Cleanup kit (Qiagen, Hilden, Germany), a preadenylated 3′ adaptor was ligated and the reverse transcription primer hybridized to minimize adaptor-dimer formation. Next, the 5′ adaptor was ligated and reverse transcription was performed using SuperScript IV (Thermo Fisher Scientific) for its high sequence fidelity and ability to process sites comprising RNA modifications. Finally, polymerase chain reaction was performed using custom i5 and i7 indexes and the PCR product purified via gel excision on a denaturing polyacrylamide gel. Full length product was then subjected to sequencing on an Illumina NextSeq 2000 platform using a custom i5 index read primer (GATCGTCGGACTGTAGAACTCTGAAC). |
