Ribosome rescue factor PELOTA modulates translation start site choice and protein isoform levels of transcription factor C/EBPα

Figures

• 

  Figure 1:. A quantitative, two color fluorescent reporter for measuring start site selection on CEBPA (A) Schematic representation of the CEBPA transcript (top) and long and short isoforms (bottom). Length of the 5′ UTR, uORF and intercistronic region are indicated in nucleotides (nt). Amino acid positions of primary start codons and main stop codon are also indicated. (B) Model of long and short isoform translation. Expression of the long isoform relies on leaky scanning past the uORF start codon whereas the short isoform seems to primarily depend on reinitiation after uORF translation. (C) Schematic of the two color reporter assay. Long and short start sites are indicated. The long isoform encodes both green (mNeonGreen2) and red fluorescence (mScarlet) while the short isoform encodes only red fluorescence (mScarlet). Δ2: mutation of a second, in-frame start codon to ACA; ΔDBD: deletion of the CEBPA basic, DNA-binding domain. (D) Western blot of K562 cells stably expressing CEBPA reporters relative to β-actin loading control. Δ uORF start: AUG→ACA; Δ Long start: AUG→ACA. (E) Flow cytometry measurements of the green/red fluorescent ratio distribution in K562 cells stably expressing either…

• 

  Figure 2:. CRISPRi sublibrary screens identify regulators of CEBPA isoform expression (A) Schematic representation of FACS-based CRISPRi screening strategy. Bin labels reflect green/red distribution. FR: far red; NR: near red; NG: near green; FG: far green. (B) Distribution of sgRNA read counts across FACS bins for sgRNAs (in C and D) against DENR, PELO and CALR. (C) Gene Expression sgRNA sublibrary profile representing the relative shift in long and short isoform usage. Each point represents a single sgRNA with sgRNAs against DENR highlighted. Colors indicate cutoffs for significance (false discovery rate, FDR < 0.05), genes with at least 2 sgRNAs and nontargeting sgRNAs. (D) Cancer and Apoptosis sgRNA sublibrary profile, as in (C). sgRNAs against MCTS1 and PELO are highlighted. (E) Gene ontology (GO) terms associated with sgRNAs that were enriched in cells expressing more long isoform (a higher green/red ratio) in the Gene Expression sublibrary screen. Only sgRNAs with a FDR < 0.05 were included in the analysis. GO term analysis was performed using Fisher’s exact test using the Bonferroni correction for multiple testing. Categories chosen represent the most statistically significant terms…

• 

  Figure 3:. Validation of top CRISPRi sublibrary screen targets (A) Flow cytometry measurements of the CEBPA two color reporter cell line transduced with either a nontargeting sgRNA or the top scoring sgRNAs against DENR, DAP5 and PELO. (B) Comparison of the difference in green/red fluorescence ratio by flow cytometry between each target sgRNA and a nontargeting sgRNA and their corresponding read count enrichment in the CRISPRi sublibrary screens, ρ=0.96. (C) Western blot of PELO levels relative to β-actin in two color reporter lines transduced with either a nontargeting sgRNA or the top scoring PELO sgRNA. Clones represent separate sgRNA transductions. (D) Median green/red fluorescence measurements of the two color reporter by flow cytometry in the PELO rescue assay, n = 3 or 4. Re-expression of constructs was induced by 1 μM Shield1 (Shld1) and harvested 72h later. Measurements were normalized to uninduced, nontargeting sgRNA conditions. (E) Median cell count measurements of cell lines in the PELO rescue assay 72h post-Shld1 treament, n = 3.

• 

  Figure 4:. The effect of PELO on CEBPA translation depends on the uORF length (A) Distribution of green/red ratio by flow cytometry of cell lines stably expressing two color reporters with either a Δ uORF start codon mutant or a short (22 nucleotides), unstructured 5′UTR with no uORFs. (B) Difference in median green/red fluorescence ratio by flow cytometry between each target sgRNA and a nontargeting sgRNA in the indicated stable cell lines. Statistical significance was calculated by two-sided t-tests with p-values denoted by: * : < 0.05 and ** : < 0.01. (C) Schematic of CEBPA uORF with length of uORF and intercistronic region indicated in nucleotides (nt). (D) Distribution of green/red ratio by flow cytometry of cell lines stably expressing two color reporters with varying uORF lengths. Length of uORFs are indicated in nucleotides (nt). The distance between the uORF stop codon and main start codon is 7 nt in all reporters. (E) Median difference in green/red fluorescence ratio by flow cytometry between each target sgRNA and a nontargeting sgRNA in the indicated stable cell lines. Statistical significance was calculated as in (B).

• 

  Figure 5:. Three color reporter reveals that PELO knockdown decreases long isoform expression (A) Schematic of three color reporter assay (top). Flow cytometry measurements of median green and red fluorescence normalized to IRFP670 in stable cell lines expressing the three color reporter, the Δ uORF start codon mutant reporter or the Δ Long start codon mutant reporter, n = 3 (bottom). (B) Flow cytometry measurements of median green and red fluorescence normalized to IRFP670 in stable cell lines expressing the wild type three color reporter transduced with indicated sgRNAs. (C) Median green and red fluorescence normalized to IRFP670 in stable cell lines expressing the wild type three color reporter transduced with indicated dual sgRNAs.

• 

  Figure 6:. PELO knockdown increases uORF translation and activates mTOR (A) Ribosome occupancy profile of CEBPA 5′ UTR. Start and stop codons belonging to the uORF and long isoform start codon are indicated. Read counts were normalized by CDS occupancy and the median read count was determined across replicates, n=3. (B) Scatterplot of log2 fold changes (PELO KD/control) in ribosome footprint and RNA-seq counts. Each point represents a single gene. Genes with a statistically significant (FDR < 0.05) log2 fold change in translation efficiency (TE) are indicated in blue. (C) Cumulative distribution of TE log2 fold changes of TOP-containing mRNAs. (D) Median green and red fluorescence normalized to IRFP670 in stable cell lines expressing the wild type three color reporter transduced with indicated dual sgRNAs. (E) Median green/red fluorescence measurements in reporter cell lines transduced with the indicated sgRNAs. Cells were treated with either DMSO or 1 μM PP242 (Sigma-Aldrich) 6 days post-transduction. Cells were harvested and assayed by flow cytometry 48h post-drug treatment. (F) Flow cytometry measurements of median green and red fluorescence from (E), n = 3.