Dissecting the transcriptional phenotype of ribosomal protein deficiency: implications for Diamond-Blackfan Anemia

Abstract

Defects in genes encoding ribosomal proteins cause Diamond Blackfan Anemia (DBA), a red cell aplasia often associated with physical abnormalities. Other bone marrow failure syndromes have been attributed to defects in ribosomal components but the link between erythropoiesis and the ribosome remains to be fully defined. Several lines of evidence suggest that defects in ribosome synthesis lead to "ribosomal stress" with p53 activation and either cell cycle arrest or induction of apoptosis. Pathways independent of p53 have also been proposed to play a role in DBA pathogenesis. We took an unbiased approach to identify p53-independent pathways activated by defects in ribosome synthesis by analyzing global gene expression in various cellular models of DBA. Ranking-Principal Component Analysis (Ranking-PCA) was applied to the identified datasets to determine whether there are common sets of genes whose expression is altered in these different cellular models. We observed consistent changes in the expression of genes involved in cellular amino acid metabolic process, negative regulation of cell proliferation and cell redox homeostasis. These data indicate that cells respond to defects in ribosome synthesis by changing the level of expression of a limited subset of genes involved in critical cellular processes. Moreover, our data support a role for p53-independent pathways in the pathophysiology of DBA.

Keywords: Bone marrow failure; Diamond Blackfan Anemia; Ribosomal protein; Ribosomopathy.

Fig. 1

p53 in TF1 cells. A. TF1 cells do not present the wild type form of p53. Sequencing of genomic DNA showed two mutations in trans: one leads to the skipping of exon 7 and nonsense mediated mRNA decay (NMD), the other induces frameshift without NMD and was also detected by cDNA sequencing, as shown in the electropherogram. The aberrant transcript gives rise to a mutant protein that carries 93 incorrect amino acids at the C-terminus. Electropherogram of p53 from CD34⁺ primary cells and a schematic representation of p53 protein domains are shown as a wild type control. B. Immunoblotting performed with an antibody against the N-terminal region of p53 reveals a smaller protein in TF1 cells than the full-length p53 expressed by CD34⁺ cells.

Fig. 2

RPS19 silencing in TF1 cells. A. Western blot showing the downregulation of RPS19 protein in TF1 cells, compared to scrambled controls, after four days of DOX treatment. The densitometry analysis, performed on three replicates, shows a statistically significant downregulation of RPS19. *p value < 0.05. B. Growth curve of TF1 cells treated with DOX for four days. C. Cell cycle analysis by flow cytometry of TF1 cells treated with DOX for four days and stained with propidium iodide. The bar graphs show the percentage of cells in subG1 phase on total cells and the percentage of cells in G₀/G₁ and G2/M phase on viable cells, as the mean of three replicates. Standard deviation bars are shown. *p value < 0.05.

Fig. 3

PCA on RP deficient TF1 cells. Score plot of the first two PCs calculated on the dataset containing TF1 cell lines downregulated for RPS19, RPL5 and RPL11. Samples are separated along PC₁ in controls (positive scores; empty circles) and pathological (negative scores; full circles). Labels: S19 = TF1 downregulated for RPS19; CS = scrambled controls for RPS19; L5 and L11 = TF1 downregulated for RPL5 and RPL11; CL = scrambled controls for RPL5 and RPL11.

Fig. 4

Validation of microarray results by qRT-PCR. Fold change of the expression of eight altered genes in RP depleted TF1 cells compared to scrambled controls (set equal to 1). Data were obtained by qRT-PCR measurement and normalized to GAPDH or β-actin levels. *p value < 0.05, ^○p < 0.01, ^‡ p < 0.001.