Dual Inhibition of DKC1 and MEK1/2 Synergistically Restrains the Growth of Colorectal Cancer Cells

Abstract

Colorectal cancer, one of the most commonly diagnosed cancers worldwide, is often accompanied by uncontrolled proliferation of tumor cells. Dyskerin pseudouridine synthase 1 (DKC1), screened using the genome-wide RNAi strategy, is a previously unidentified key regulator that promotes colorectal cancer cell proliferation. Enforced expression of DKC1, but not its catalytically inactive mutant D125A, accelerates cell growth in vitro and in vivo. DKC1 knockdown or its inhibitor pyrazofurin attenuates cell proliferation. Proteomics, RNA immunoprecipitation (RIP)-seq, and RNA decay analyses reveal that DKC1 binds to and stabilizes the mRNA of several ribosomal proteins (RPs), including RPL10A, RPL22L1, RPL34, and RPS3. DKC1 depletion significantly accelerates mRNA decay of these RPs, which mediates the oncogenic function of DKC1. Interestingly, these DKC1-regulated RPs also interact with HRAS and suppress the RAS/RAF/MEK/ERK pathway. Pyrazofurin and trametinib combination synergistically restrains colorectal cancer cell growth in vitro and in vivo. Furthermore, DKC1 is markedly upregulated in colorectal cancer tissues compared to adjacent normal tissues. Colorectal cancer patients with higher DKC1 expression has consistently poorer overall survival and progression-free survival outcomes. Taken together, these data suggest that DKC1 is an essential gene and candidate therapeutic target for colorectal cancer.

Keywords: MEK1/2; dyskerin pseudouridine synthase 1; pyrazofurin; ribosomal protein; trametinib.

Figure 1

shRNA library screening identifies DKC1 promoting the proliferation of colorectal cancer cells. A) Negative selection of essential genes via NEST. B) Expression of DKC1 in colorectal cancer cell lines after lentiviral transduction of DKC1 shRNA or nonspecific shRNA (shCTL) by western blot (top) or qPCR (bottom). C) The cell growth curves of DKC1 knockdown or control colorectal cancer cells represent the O.D. values obtained from three independent CCK‐8 assays (mean ± standard deviation (SD)). D) 3D Matrigel colony formation assays of DKC1 knockdown and control cells. The top panel shows representative 3D Matrigel colony formation images (100×; the scale bar indicates 100 µm); the bottom panel shows bar charts indicating the number of colonies per well (mean ± SD). E) Colony formation assays of DKC1 knockdown and control cells. The top panel shows representative colony formation images; the bottom panel shows a bar graph of the number of colonies per well (mean ± SD). F) The left panel shows representative images of DKC1 knockdown or control human colorectal cancer (hCRC) organoids (scale bar: 200 µm). The right panel shows bar charts of DKC1 expression and organoid number and size (mean ± SD). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 ((B, D, E) one‐way ANOVA with Bonferroni correction, (C) two‐way ANOVA with Bonferroni correction, (F) two‐sided Student's t‐test).

Figure 2

DKC1 promotes colorectal cancer cell growth via its pseudouridine synthase activity. A) Immunoblot detecting DKC1 expression and dot blots monitoring cellular pseudouridine levels in control and DKC1 knockdown DLD‐1 and HCT116 cells with rescued expression of wild‐type DKC1 or mutant DKC1 (D125A). B–E) Results of cell growth curve (B), 3D Matrigel assays (C), and colony formation (D–E) of DLD‐1 or HCT116 cells with DKC1 knockdown and rescued expression of wild‐type DKC1 or D125A from three independent experiments (mean ± SD). F–G) Subcutaneous tumor models were established with control or DKC1 knockdown DLD‐1 cells with or without rescued expression of wild‐type DKC1 or D125A. F) The top panel shows tumor images after 4 weeks of injection. The bottom panel shows tumor volumes recorded at the indicated times. G) Immunostaining of DKC1 and Ki67 in the indicated tumors. Scale bar: 20 µm. H) The effects of the DKC1 inhibitor pyrazofurin (PF) on overall pseudouridine levels in DLD‐1 and HCT116 cells by dot blots (left) and cell growth (right). I) The effects of the DKC1 inhibitor PF on human colorectal cancer (hCRC) organoids. The right panel shows representative images of hCRC organoid treated with the indicated concentration of PF. Scale bar: 200 µm. The right panel shows bar charts indicating organoid number (mean ± SD). E.V: empty vector. *P < 0.05, **P < 0.01, ****P < 0.0001, ns: no significance ((B, F, H) two‐way ANOVA with Bonferroni correction, (C, D, E, I) one‐way ANOVA with Bonferroni correction).

Figure 3

Proteomic analysis reveals that DKC1 elicits ribosomal protein expression. A) Volcano plot of differentially expressed proteins obtained from proteomic analysis of triplicate samples of DKC1 knockdown DLD‐1 cells and control cells. A total of 114 downregulated proteins and 58 upregulated proteins were included. B,C) GO enrichment analysis for the cellular component category (B) and KEGG enrichment analysis (C) of differentially expressed proteins. D) Comparison of the mRNA abundance of 28 ribosomal proteins measured by qRT‐PCR and their protein levels measured by proteomic analysis of DKC1 knockdown DLD‐1 cells and control cells (All proteins with P < 0.05). E) qRT‐PCR assays evaluating the mRNA levels of the indicated genes in control and DKC1 knockdown DLD‐1 cells with or without enforced expression of wild‐type DKC1 or the DKC1 mutant (D125A) from three independent experiments (mean ± SD). F) Immunoblots assessing the abundance of the indicated ribosomal proteins in control and DKC1 knockdown DLD‐1 and HCT116 cells with or without enforced expression of wild‐type DKC1 or D125A and PF‐treated DLD‐1 and HCT116 cells (48 h). E.V: empty vector. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns: no significance ((D) two‐sided Student's t‐test, (E) one‐way ANOVA with Bonferroni correction).

Figure 4

DKC1 binds to and stabilizes the mRNA of downstream ribosomal proteins. A) Distribution of peaks enriched by DKC1 on the genome in DLD‐1 cells. B) Three new DKC1 binding motifs were discovered in DLD‐1 shCTL cells. C) Integrative Genomics Viewer (IGV) was used to visualize the distribution of reads on the indicated genes. D) Validation of the genome‐wide DKC1 enrichment results in by RIP‐qPCR from three independent experiments (means ± SD). E) Evaluation of the function of the reported pseudouridylation site (Ψ1182) in the RPS3 3′UTR. Top panel: Schematic diagram of RPS3 mRNA. Middle panel: Schematic diagram of constructs used for the dual luciferase assay. At position Ψ1182, a substitution of T with G was used to abolish the potentially modified sites in DKC1. Bottom panel: The effects of the wild‐type RPS3 3′UTR and mutant RPS3 3′UTR (RPS3‐3′UTR‐M) on luciferase activity in DKC1 knockdown HCT116 cells with or without enforced expression of wild‐type DKC1 or D125A were evaluated with a dual luciferase assay system. F,G) HCT116 or DLD‐1 cells were treated with actinomycin D (5 µg mL⁻¹) for the indicated time. The abundance of the indicated mRNAs in control and DKC1 knockdown cells with or without enforced expression of wild‐type DKC1 or D125A was monitored by qRT‐PCR at different time points, and mRNA decay curves were constructed and fit. Data in the graph are shown as the means ± SD from three independent experiments. E.V: empty vector. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns: no significance ((D) two‐sided Student's t‐test, (E) two‐way ANOVA with Bonferroni correction).

Figure 5

DKC1 promotes colorectal cancer cell growth by regulating ribosomal proteins. A,B) DLD‐1 cells with DKC1 knockdown were rescued with or without ectopic expression of DKC1 and were further transduced with lentiviral vectors carrying shRNA individually targeting the indicated ribosomal proteins. The cell growth curve (A) and colony formation (B) of these cells were assessed. (C‐D) DLD‐1 cells with DKC1 knockdown were transduced with lentiviral vectors encoding the indicated ribosomal proteins individually. C) The cell growth curve and D) colony formation of these cells and control cells were monitored. E) DLD‐1 cells with DKC1 knockdown were transduced with lentiviral vectors individually encoding the indicated ribosomal proteins. A total of 1.5 × 10⁶ of these cells and control DLD‐1 cells were subcutaneously injected into BALB/c nude mice. Tumor images after 4 weeks of injection (left). Tumor volumes recorded at the indicated times (right). E.V: empty vector. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns: no significance ((A, C, E) two‐way ANOVA with Bonferroni correction, (B, D) one‐way ANOVA with Bonferroni correction).

Figure 6

Combination of pyrazofurin and trametinib synergistically suppressed colorectal cancer. A) Immunoblot indicating the total expression levels and phosphorylation levels of ERK1/2 in DLD‐1 cells with knockdown of DKC1 or the indicated ribosomal proteins (left) or in DKC1‐silenced DLD‐1 cells with ectopic expression of DKC1 or the indicated ribosomal proteins (right). B) The effects of PF (48 h) on the phosphorylation levels of ERK1/2 in DLD‐1 and HCT116 cells by immunoblot. C) Lysates from HEK293T cells co‐transfected with plasmids expressing Myc‐tagged E.V, HRAS, KRAS, NRAS, ARAF, BRAF, CRAF, MEK1, and MEK2, along with RPS3 were immunoprecipitated with an anti‐Myc antibody and subjected to immunoblot analysis with an anti‐RPS3 antibody and anti‐Myc antibody. D) Lysates from HEK293T cells cotransfected with plasmids expressing Flag‐tagged E.V, RPL10A, RPL22L1, and RPL34, along with Myc‐HRAS were immunoprecipitated with an anti‐Flag antibody and subjected to immunoblot analysis with an anti‐Flag antibody and anti‐Myc antibody. E) (top) Schematic diagram showing the truncated RPS3. (bottom) Lysates from HEK293T cells cotransfected with plasmids expressing Flag‐tagged E.V, and the truncated RPS3, along with Myc‐HRAS were immunoprecipitated with an anti‐Flag antibody and subjected to immunoblot analysis with an anti‐Flag antibody and anti‐Myc antibody. F,G) The combination of PF and trametinib treatment on DLD‐1 (F) and HCT116 cells (G). Cell viability was assessed by ATP assay and the combination index was examined by Calcusyn. H,I) The combination of PF and trametinib treatment on subcutaneous tumor models established with HCT116 cells. (H) Tumor images at day 28 post‐injection. (I) Tumor volumes were recorded at the indicated times (left), and tumor weights were measured after dissection(right). E.V: empty vector. *P < 0.05, **P < 0.01, ****P < 0.0001, ((I) left, two‐way ANOVA with Bonferroni correction; right, one‐way ANOVA with Holm–Šídák's correction)

Figure 7

High expression of DKC1 in colorectal cancer predicts poor prognosis. A) Immunoblot indicating the expression of DKC1 and its downstream ribosomal proteins in colorectal cancer cell lines and normal mucosa. B) qRT‐PCR analysis evaluating the relative expression levels of DKC1 mRNA in 18 pairs of human primary colorectal cancer tissues and normal mucosa. C) The expression of DKC1 and the four indicated ribosomal proteins in 4 pairs of human primary colorectal cancer tissues (T) and normal mucosa (N) was assessed by immunoblotting. D) Correlation analysis between the DKC1 level and the mRNA abundance of the indicated ribosomal proteins in the TCGA colorectal cancer database. E) IHC analysis of 108 pairs of colorectal cancer and adjacent normal tissues. (Left) Representative IHC staining of DKC1 in the normal colon, high DKC1 expression and low DKC1 expression groups. Scale bar: 50 µm. (Right) Kaplan–Meier analysis of OS and PFS according to DKC1 expression in 108 colorectal cancer patients. F) A proposed model of the mechanism of pseudouridine synthase DKC1 to promote colorectal cancer cell proliferation and the rational of combination therapy using PF (DKC1 inhibitor) and trametinib (MEK1/2 inhibitor) to treat colorectal cancer. *P < 0.05, ((B) two‐sided Student's t‐test).