eIF4A2 drives repression of translation at initiation by Ccr4-Not through purine-rich motifs in the 5’UTR.
Regulation of the mRNA life-cycle is central to gene expression control and determination of cell fate. miRNAs represent a critical mRNA regulatory mechanism, but despite decades of research, their mode of action is still not fully understood. Here we show eIF4A2 is a major effector of the repressive miRNA pathway functioning via the Ccr4-Not complex. We demonstrate that while DDX6 interacts with Ccr4-Not, its effects in the mechanism are not as pronounced. Through its interaction with the Ccr4-Not complex eIF4A2 represses mRNAs at translation initiation. We show evidence that native eIF4A2 has similar properties to chemically inhibited eIF4A1. We demonstrate that eIF4A2 exerts it repressive effect by binding purine-rich motifs which are enriched in the 5’UTR directly upstream of the AUG start codon. The data supports a model whereby purine motifs towards the 3’ end of the 5’UTR are associated with increased ribosome occupancy and possible uORF activation similar to that observed for mRNAs affected by inhibited eIF4A1.
Sample Processing Protocol
SILAC-labelled HEK293 cells were obtained by culturing in SILAC-DMEM lacking arginine and lysine (Life Technologies) supplemented with[13C6] L-arginine and [13C6] [15N2] L-lysine(SILAC medium - M) (Sigma-Aldrich) or [13C6][15N4] L-arginine and [2H4] L-lysine (SILAC heavy - H; Cambridge Isotope Laboratories, Tewksbury, MA) for 14h. Each comparison was done in the forward (H/M) and reverse (M/H) directions. After this, cells were harvested into SDS-free RIPA buffer. 150 µg of each quantified SILAC labelled lysates was mixed in a 1:1 ratio, total protein amount of 300µg. Samples were then reduced with DTT, to a final concentration of 5mM, and alkylated with IAA, final concentration of 50 mM. Samples were then subject to a two-step digestion, firstly with Endoproteinase Lys-C (ratio 1:33 enzyme:lysate) for 1 hour at room temperature then with trypsin (ratio 1:33 enzyme:lysate) overnight at 37°C. The digested SILAC samples were fractionated using reverse phase chromatography at pH 10. Solvents A (98% water, 2% ACN) and B (90% ACN, 10% water) were adjusted to pH 10 using ammonium hydroxide. 300 µg of digested peptides were loaded onto a Kinetex C18 column (150 x 2.1 mm) coupled with a Dionex Ultimate 3000 HPLC system, software version 6.7 (Chromeleon). Injected peptides were subject to a two-step gradient, 4-27% Solvent B in 36 mins then 27-48% Solvent B in 8 mins. The flow rate was set to 200 µL/min. The samples was collected into 21 fractions. Peptide samples were run on the Q-Exactive HF mass spectrometer coupled to an EASY-nLC II 1200 chromatography system (Thermo Scientific). Samples were loaded into a 20 cm fused silica emitter, packed in-house with ReproSIL-Pur C18-AQ, 1.9µm resin, which was heated to 35°C using a column oven (Sonation). Peptides were eluted at a flow rate of 300 nl/min over three optimised two-step gradient methods for fractions 1-7, 8-15 and 16-21. Step one was commenced for 20 mins and step two for 7 mins. For fractions 1-7 the % of solvent B was 2-20% at step one and 39% at step two; For fractions 8-15 was 4-23% at step one and 43% at step two; for fractions 16-21 was 6-28% at step one and 48% at step two. Peptides were electrosprayed into the mass spectrometer using a nanoelectropsray ion source (Thermo Scientific). An Active Background Ion Reduction Device (ABIRD, ESI Source Solutions) was used to decrease air contaminants. Data was acquired with the Xcalibur software (Thermo Scientific) in positive mode utilising data-dependent acquisition. The full scan mass range was set to 375-1400m/z at 60,000 resolution. Injection time was set to 20 ms with a target value of 3E6 ions. HCD fragmentation was triggered on the 15 most intense ions for MS/MS analysis. MS/MS injection time was set to 50 ms with a target of 5E2 ions. Ions that have already been selected for MS/MS were dynamically excluded for 25 s.
Data Processing Protocol
MS raw data was processed using MaxQuant software version 220.127.116.11 and searched with the Andromeda search engine against the Uniprot Homo sapiens database (95146 entries). First and main searches were done with a precursor mass tolerance of 20 ppm and 4.5 ppm, respectively. MS/MS mass tolerance was set to 20 ppm. Minimum peptide length was set to 6 amino acids and trypsin cleavage was selected allowing up to 2 missed cleavages. Methionine oxidation and N-terminal acetylation were selected as variable modifications and Carbamidomethylation as fixed modification. False discovery rate for peptide and protein identification was set to 1%. SILAC multiplicity was set to 3 and the medium (Arginine 6 and Lysine 4) and heavy (Arginine 10 and Lysine 8) labels were selected. MaxQuant output was processed using Perseus software version 18.104.22.168. Reverse and Potential contaminant proteins were removed as well as proteins identified only by site and those that did not have at least one uniquely assigned peptide. For protein amounts in control conditions, intensities were corrected for molecular weight. For relative protein amounts, H/M and M/H ratios from MaxQuant were used. Two replicates – forward and reverse labelled - were analysed.