Deficiency of ribosomal proteins reshapes the transcriptional and translational landscape in human cells

Abstract

Human ribosomes have long been thought to be uniform factories with little regulatory function. Accumulating evidence emphasizes the heterogeneity of ribosomal protein (RP) expression in specific cellular functions and development. However, a systematic understanding of functional relevance of RPs is lacking. Here, we surveyed translational and transcriptional changes after individual knockdown of 75 RPs, 44 from the large subunit (60S) and 31 from the small subunit (40S), by Ribo-seq and RNA-seq analyses. Deficiency of individual RPs altered specific subsets of genes transcriptionally and translationally. RP genes were under cotranslational regulation upon ribosomal stress, and deficiency of the 60S RPs and the 40S RPs had opposite effects. RP deficiency altered the expression of genes related to eight major functional classes, including the cell cycle, cellular metabolism, signal transduction and development. 60S RP deficiency led to greater inhibitory effects on cell growth than did 40S RP deficiency, through P53 signaling. Particularly, we showed that eS8/RPS8 deficiency stimulated apoptosis while eL13/RPL13 or eL18/RPL18 deficiency promoted senescence. We also validated the phenotypic impacts of uL5/RPL11 and eL15/RPL15 deficiency on retina development and angiogenesis, respectively. Overall, our study provides a valuable resource for and novel insights into ribosome regulation in cellular activities, development and diseases.

Figure 1.

Overview of the transcriptome and translatome datasets. (A) Length of RPFs in Ribo-seq libraries for all samples. Each point indicates one library. (B, C) The relative fraction of reads mapped to the CDS, 5′ UTR and 3′ UTR of annotated transcripts in RNA-seq (B) and Ribo-seq (C) libraries. When genome features were overlapped, CDS exons were prioritized over UTR exons. (D) Percentages of RPFs mapped to reading frames by combining all Ribo-seq libraries. Each point indicates one library. (E, F) Fractions of reads assigned to each nucleotide around the start codons (E) or stop codons (F) for all libraries in Ribo-seq data. Each bar indicates one sample. P-site was used to assign the short read to transcript location. (G) Pearson correlation coefficients between replicates in RNA-seq and Ribo-seq datasets. Log₂ RPKM values were used in the correlation analysis.

Figure 2.

Diversity of gene expression changes upon RP knockdown. (A) Numbers of up- and down-regulated differential genes in RNA-seq and Ribo-seq for all RPs. RPs are ranked by DTG numbers in Ribo-seq. (B) Frequency of common DEGs or DTGs between RPs in RNA-seq and Ribo-seq. (C) Distribution of DEG or DTG numbers for all the RPs. The distributions in RNA-seq and Ribo-seq were compared by Wilcoxon sign-rank tests. (D) Ranges of global gene fold changes for all RPs. The ranges in RNA-seq and Ribo-seq were compared by Wilcoxon sign-rank tests. (E) Percentages of genes under different gene expression regulation modes after knockdown of each RP. RPs were grouped by hierarchical clustering analysis of the regulation profiles. Rectangles indicates the primary regulation mode for RP groups.

Figure 3.

Co-regulation of 60S and 40S RPs. (A) The expression changes of RPs (rows) after knockdown of individual RPs (columns) in Ribo-seq dataset. Log₂ fold change values are shown. RPs for rows and columns according to ribosomal subunits as indicated by colored sidebars: cyan color indicates 60S RPs, orange color indicates 40S RPs. (B) Comparison of global changes in TE of 40S and 60S RPs after knockdown of RPs from 40S (left) or 60S (right). Median TE values of the remaining RPs were used for each RP targeted by siRNAs. (C) Numbers of RPs annotated to cellular locations where ribosomal assembly occurs. (D) Comparison of expression changes at the translatome level (left) and TE changes (right) of the remaining RPs after knockdown of individual RPs within different stages of ribosomal subunit assembly. Wilcoxon tests were performed to compare the difference between groups for each subunit. (E) Polysome profiling by sucrose-gradient-based centrifugation showing the changes in abundance of ribosomal components (40S, 60S, 80S monosome and polysomes) in A549 cells treated by specific siRNAs targeting eS8/RPS8, eS25/RPS25, uL5/RPL11 or eL15/RPL15.

Figure 4.

Functional characterization of molecular landscape of RP-deficient cells. (A) 64 RPs are ranked according to the number of significantly enriched GO BP terms. Annotation bar at the bottom indicates the enrichment of p53 signaling pathway after RP knockdown; Middle heatmap shows the fraction of GO classes; Bar plot at the top shows total number of enriched GO BP terms for each RP. (B) Summary of the recovered known functional associations of RPs. Details can be found in Supplementary Table S5. (C) Percentages of overlapped DEGs or DTGs after knockdown of 40S or 60S RPs. T-tests were performed. (D) Comparison of P53 protein levels after knockdown of Group1 and Group2 RPs. P53 protein levels were extracted from (42). (E) mRNA changes of p53 target genes in our RNA-seq datasets for the RPs in group1 and group2. The p53 target genes were extracted from M Fischer’s review (53). (F) Comparison of sgRNA expression levels at Day 2 and Day 7 after transfection. Linear regression models were estimated and tested for sgRNAs with RP targets and without targets. The estimated slope values indicate cell growth rate over time: slope >1 indicates positive cell growth, slope <1 indicates repressed cell growth. (G) Comparison of expression changes over time between sgRNAs targeting RPs from Group1 and that from Group2. T-test was performed.

Figure 5.

RP deficiency leads to divergent cell fates after cell cycle arrest. (A) Percentages of cells within different cell cycle stages (G1, S, and G2/M) by flow cytometry experiments on A549 cells at 24 h after knockdown of indicated RPs. Three replicates were used in t-tests. (**), P < 0.01; (***), P < 0.001. (B) Changes of cell viability by MTT assays on A549 cells at 24 h after knockdown of indicated RPs. Three replicates were used in t-tests. (**), P < 0.01; (***), P < 0.001. (C) Representative of western blotting assays (left panel) and quantification of p53 (middle panel) or p21 (right panel) protein levels at 24 h after knockdown of indicated RPs (n = 2 for p53 or 3 for p21 tests). T-tests were used. (*), P < 0.05; (**), P < 0.01; (***), P < 0.001. (D) Bar plots (left panel) showing the percentage of TUNEL+ cells at 72 h after knockdown of eS8/RPS8. Representative of TUNEL staining assays (right panel) for testing apoptosis in situ in A549 cells at 72 h after knockdown of eS8/RPS8. Three replicates were used in t-test. (*), P < 0.05. (E) Bar plots (left panel) showing the percentages of β-gal-positive cells at 24, 48 and 72 h after knockdown of eL13/RPL13 or eL18/RPL18. Representative of β-gal staining assays (right panel) for testing senescence in A549 cells at 24, 48 and 72 h after knockdown of eL13/RPL13or eL18/RPL18. Three replicates were used in t-tests. (**), P < 0.01.

Figure 6.

RP deficiency conferred different functional preference in development. (A) The ranked tissue development terms affected by RP knockdown according to the significance. Tissue development terms were defined according to GO structure. The p-values for all subentries belonging to the term of interest were combined with Fisher's method. Grey bars indicate GO terms enriched by down-DTGs; Red bars indicate GO terms enriched by up-DTGs. (B, C) Analysis of developing retinal cells after in vivo conditional knockdown of mouse uL5/Rpl11. (B) P0 retinas co-electroporated RP-targeting shRNAs or pU6 plasmid with the pCIG vector were collected at P12, and their sections were double-immunostained with an anti-GFP antibody and antibodies against Chx10, Pax6, Sox9 or Recoverine. Arrows point to representative colocalized cells. (C) The numbers of specific cell types and statistical testing results between groups. Two or three replicates were used in t-tests. (*), P < 0.05; (**), P < 0.01. (D) The relative mRNA levels by qPCR of representative genes in eL15/RPL15-deficient A549 cells. Three replicates were used in t-tests. (*), P < 0.05; (**), P < 0.01; (***), P < 0.001. (E) The relative mRNA levels of VEGFA in the control HUVEC and the HUVEC upon knockdown of indicated RPs. Three replicates were used in t-tests. (*), P < 0.05. (F) ELISA analysis showing the concentration of VEGF proteins in conditioned media from the control A549 cells and eL15/RPL15-deficient A549 cells. T-test was used. (**), P < 0.01. (G) Images of the control HUVEC and eL15/RPL15-deficient HUVEC 0hr and 6hr after a scratch was introduced in the monolayer with a pipette tip (left panel). The percentage of covered wound area for each replicate at 6 h over that at 0hr was estimated and compared between eL15/RPL15-deficient cells and the control cells (right panel). Four replicates were used in t-test. (*), P < 0.05.