Nuclear pore component Nup98 is a potential tumor suppressor and regulates posttranscriptional expression of select p53 target genes

Abstract

The p53 tumor suppressor utilizes multiple mechanisms to selectively regulate its myriad target genes, which in turn mediate diverse cellular processes. Here, using conventional and single-molecule mRNA analyses, we demonstrate that the nucleoporin Nup98 is required for full expression of p21, a key effector of the p53 pathway, but not several other p53 target genes. Nup98 regulates p21 mRNA levels by a posttranscriptional mechanism in which a complex containing Nup98 and the p21 mRNA 3'UTR protects p21 mRNA from degradation by the exosome. An in silico approach revealed another p53 target (14-3-3σ) to be similarly regulated by Nup98. The expression of Nup98 is reduced in murine and human hepatocellular carcinomas (HCCs) and correlates with p21 expression in HCC patients. Our study elucidates a previously unrecognized function of wild-type Nup98 in regulating select p53 target genes that is distinct from the well-characterized oncogenic properties of Nup98 fusion proteins.

Figure 1. Nup98 Depletion Selectively Reduces p21 mRNA and Protein Accumulation upon p53 Activation

(A) HepG2 cells were treated either with control (ctrl) or two Nup98 siRNAs (Nup9898#1 and Nup9898#2) for 72 h and CPT (300 nM) was added for the final 24 h before lysis of cells for mRNA and protein analysis. Relative expression of p53 target genes: p21, PUMA, GADD45, NOXA, ATF3 and Nup98 was measured by qRT-PCR and normalized to 1 for each ctrl siRNA and drug treatment condition. Data are presented as mean ± standard deviation (SD) of three independent experiments. (B) HepG2 cell were treated as in (A) and cell extracts were examined by immunoblotting with indicated antibodies. (C) HepG2 cells were treated either with control (ctrl) or Nup9898#1 and Nup9898#2 siRNA for 72 h and Nutlin-3 (20 µM) was added 24 h prior to extraction of cells for protein analysis. Cell extracts were analyzed by immunoblotting with the indicated antibodies.

Figure 2. Post-transcriptional Regulation of p21 by Nup98

(A) Nup98 regulates the levels of mature but not nascent p21 mRNA. Upper left panel: schematic depicts primers used to amplify different regions of nascent and mature p21 mRNA. The numbers indicate the position of the amplicon within the p21 gene locus. Rectangles represent exons 1 to 3. HepG2 cells were treated as described in Figure 1A. Nascent p21 mRNA induction was measured by qRT-PCR with primer pairs annealing to intronic regions of the unspliced transcript: first intron (+4001) (upper right panel) and second intron (+7011) (lower left panel), respectively. Lower right panel: spliced p21 mRNA was measured by using the exon-spanning primer pair (+5817/+7223) indicated in the schematic. Data are presented as mean ± SD of two independent (experiments. (B) Nup98 regulates exogenous p21 mRNA. Left panel: H24-p21 cells containing a tetracycline (Tet-off)-regulatable p21 expression construct were treated either with control (ctrl) or Nup98 siRNAs (Nup9898#1 and Nup9898#2) for 72 h. as in Figure 1. p21 was induced by removal of Tet 24 h before harvesting. Left panel shows relative p21 and Nup98 mRNA expression measured by qRT-PCR and normalized to 1 for the “ctrl siRNA tet-“ condition. Data are presented as mean ± SD derived from two independent experiments. Right panel: protein expression from the same experimental condition analyzed by immunoblotting with indicated antibodies. (C) PUMA levels are not affected by Nup98 knock-down H1299-derivative cell line containing a tetracycline (Tet-on)-regulatable PUMA expression construct (H24-PUMA) was treated either with control (ctrl) or two Nup98 siRNAs (Nup9898#1 and Nup9898#2) for 72 h. PUMA was induced by adding Doxycycline 24 h before harvesting. Cell extracts were analyzed by immunoblotting with indicated antibodies. (D) Nup98 stabilizes mature p21 mRNA HepG2 cells were treated either with control (ctrl) or Nup9898#2 siRNA (Nup9898#2) for 72 h and Nutlin-3 (20 µM) was added 24 h before blocking mRNA synthesis with 0.4 µM Actinomycin-D (Act-D). Cells were harvested at indicated time points and mature p21 mRNA decay was measured by qRT-PCR with the exon-spanning primer pairs described in Figure 2A. The solid and dotted lines indicate the half-life of p21 in ctrl-siRNA or Nup9898#2 conditions, respectively.

Figure 3. Nup98 Affects Nuclear and Cytoplasmic p21 mRNA level and Protects p21 mRNA from Degradation by the Exosome

Nup98 knock-down affects the levels of nuclear and cytoplasmic p21 mRNA. (A-E). Left panels show images of HepG2 cells that were treated either with ctrl (A,B) or Nup9898#1 (C) Nup9898#3 (D) or p21 (E) siRNAs for 72 h and then treated with either DMSO (A) or Nutlin-3 (20 µM; B,C,D and E) for 24 h prior to single-mRNA FISH analysis as described in Methods. FISH images were detected in the Cy5 channels and color-coded as red. The blue channel contains a DAPI stain to demarcate the nuclei. For display 5 to 15 images, depending on cell shape, from the cell mid-section were averaged in both channels. Scale bar shown in panel E equals 2 µm in all images. Right panels are histograms of p21 mRNA molecule counts per cell corresponding to panels A (48 cells counted); B (58 cells counted), C (29 cells counted) D (55 cells counted) and E (59 cells counted). Data are pooled from three repeats of the experiments with errors and statistical analysis shown in Table S1. (F) Exosc3 knockdown rescues lowered p21 mRNA levels upon Nup98 depletion. H24 p21 tet-off cells were treated individually with either control (ctrl) or Nup9898#1 siRNA, or co-transfected with Exosc3 siRNA (Ex398#1 and Ex398#2) for 72 h as indicated. p21 was induced by removal of Tet 48 h before harvesting. Relative p21 mRNA levels were measured with qRT-PCR. Data are presented as mean ± SD (left panel). Western blot (inset) shows Exosc3 levels in p21 H24 cells treated either with ctrl or Ex398#1 and Ex398#2 siRNA. Actin serves as loading control.

Figure 4. The 3’-UTR of p21 mRNA Co-associates with Nup98 and Is Required for Nup98 Mediated Increased Levels of p21 mRNA

(A) Nup98 associates with the 3’-UTR of p21. RNA-IP was performed as described in Methods. Briefly, Nup98 was immunoprecipitated from 0.1% formaldehyde cross-linked samples of CPT (300 nM for 24 h) treated or untreated HepG2 cells. Co-immunoprecipitated mRNAs cross-linked to Nup98 were reverse-transcribed and amplified by qRT-PCR. Primer pairs specific for different regions of p21 mRNA or PUMA mRNA were used. In the schematic below the graph rectangles represent exonic regions of the respective genes and shaded rectangles indicate the 3’-UTRs of p21 and PUMA mRNA as indicated. The numbers below the lines indicate the position of each amplicon within the p21 or PUMA gene locus, respectively. Data are presented as mean ± SD. (B) The 3’-UTR of p21 mRNA is required for increased p21 expression levels mediated by ectopically expressed Nup98. Hep3B cells were co-transfected with HA-tagged Nup98 (0.75 or 1.25 µg) and p21 cDNA containing the 3’-UTR (p21 FL, 200 ng) or p21 cDNA without the 3’-UTR (p21 del 3’-UTR, 20ng) 24h prior to RNA extraction. p21 mRNA levels transcribed from the respective constructs were measured with qRT-PCR. Data are presented as mean ± SD. (C) The 3’-UTR of p21mRNA is sufficient for stabilization by Nup98. H1299 cells were co-transfected with HA-tagged Nup98 (1.75 µg) and a Luciferase construct (Luc/p21 3’UTR, 200 ng) containing the 3’UTR of p21 mRNA or the Luciferase construct lacking the p21 mRNA 3’UTR (Luc ctrl, 20ng) 48 h prior to RNA extraction. The mRNA levels transcribed from the respective constructs were measured with qRT-PCR. Data are presented as mean ± SD of two independent experiments. (D) The 3’-UTR of p21mRNA is sufficient for stabilization by Nup98. H1299 cells were co-transfected with HA-tagged Nup98 (1.75 µg) and a construct containing the coding DNA sequence (CDS) and the 3’UTR of PUMA (Pu FL, 200ng) or a hybrid construct containing the PUMA CDS fused to the 3’UTR of p21 (Pu/p21, 200ng) The mRNA levels transcribed from the respective constructs were measured with qRT-PCR. Data are presented as mean ± SD. (E) The GLFG repeat domain of Nup98 is required for maximal increase of p21 expression H1299 cells were co-transfected with GFP tagged full length (98#1) or truncated (98#2–4) Nup98 constructs with indicated concentrations and 1 µg untagged full length p21 cDNA 24 h before protein extraction. Immunoblot shows protein expression of transfected Nup98 and p21 detected with the indicated antibodies. Actin serves as loading control.

Figure 5. An In Silico Approach Identifies 14-3-3σ as another p53 Target Gene mRNA that Associates with and is Regulated by Nup98

(A) Flow-chart describing the unbiased genomewide in silico cis-regulatory analysis that yielded the CCc/tC motif. The MatrixREDUCE algorithm was used to discover a nucleotide sequence motif (in the form of a weight matrix) whose occurrences in the 3’-UTR optimally predict p53-dependent changes in steady-state mRNA expression 72 h after doxorubicin treatment. (B) Motif occurrences in a 400 bp window immediately upstream of the polyadenylation site are significantly more predictive of the expression changes than those downstream, suggesting that the C-rich motifs represents the specificity of an RNA-binding factor. (C) HepG2 cells were treated either with ctrl or Nup9898#1 and Nup9898#2 siRNA for 72 h and 300 nM (CPT) was added for the final 24 h before RNA extraction. 14-3-3σ mRNA was measured by qRT-PCR. The ctrl siRNA and no CPT condition was taken as 1. Data are presented as mean ± SD of two independent experiments. (D) RNA-IP experiment in HepG2 cells was performed as in Figure 4A. Primer pairs specific for different regions of 14-3-3σ mRNA were used to localize the potential regions of interaction with Nup98. The rectangle represents the single 14-3-3σ exon and the shaded area indicates the 3’-UTR. The numbers indicate the position of each amplicon within the 14-3-3σ gene locus. Data are presented as mean ± SD.

Figure 6. Nup98 Depletion Increases Camptothecin (CPT) Induced Cell Death and Decreases Nutlin Induced Senescence

(A) Nup98 knock-down sensitizes cells to CPT induced apoptosis. Upper panels: FACS analysis was performed with propidium iodide stained samples of HepG2 cells following transfection with control (ctrl) or Nup98 (9898#1 and 9898#2) siRNAs and drug treatment as described in Figure 1A. Data was analyzed using the ModFit program. Lower left panel: bar diagram shows relative % of subG1 cells taken as a measurement of cell death for each indicated condition. Data are presented as mean ± SD derived from two independent experiments. Lower right panel: protein levels in HepG2 cells treated as above were analyzed by immunoblotting with indicated antibodies. (B) p21 knockdown recapitulates the effects of Nup98 knockdown in CPT treated cells. Left panel: FACS analysis was performed in HepG2 cells as described in (A) following transfection with p21 siRNA (2198#1 and 2198#2) and drug treatment. Bar diagram shows relative percentage (%) of subG1 cells as a measurement of cell death for each indicated condition. Data are presented as mean ± SD derived from two independent experiments. Right panel: protein levels in cells treated as above detected by Western blotting with indicated antibodies. (C) Nup98 knock-down decreases Nutlin induced senescence. Sk-Hep1 cells were treated either with ctrl or Nup9898#1 and Nup9898#2 siRNA for 24h and Nutlin-3 (10 µM) was added as indicated 5 days prior to staining senescence associated β galactosidase (SA- β -Gal). Shown are representative images of cells with SA-β -Gal staining for the indicated conditions.

Figure 7. Nup98 Expression Is Downregulated in Hepatocellular Carcinomas (HCCs) in MDR-2 −/− Mice and in Patients with HCC

(A-B) Lowered hepatocellular Nup98 immunoreactivity in murine tumor tissue compared to non-tumorous tissue. (A) Corresponding representative liver sections of MDR-2 −/− mice (15 months) either stained with H&E (upper left and upper right panel) or immunohistochemically stained with a polyclonal Nup98 antibody (lower left and lower right panel). Non-tumorous (NT) and tumorous (T) liver tissue is shown in the left and right column respectively. Black arrows indicate nuclei of hepatocytes and hepatocellular tumor cells and blue arrows indicate nuclei of sinusoidal cells and inflammatory cells. Scale bar represents 100 µm. (B) Diagram shows median values of semiquantiative hepatocellular Nup98 immunohistochemical staining intensities of non-tumorous (NT) and tumorous (T) liver tissue samples of individual MDR-2 −/− mice. Statistical significance was determined from non-parametric testing (Mann-Whitney-U-test, **p<0.01). (C) Nup98 expression is downregulated in human hepatocellular carcinoma. The relative Nup98 mRNA expression in 27 human hepatocellular carcinoma (HCC) tissue samples (black bars) was assessed by qRT-PCR and compared to the similarly assessed averaged expression (horizontal line) of 2 non-tumorous (NT) liver tissues normalized to 1 (white bars). Red dotted lines indicate thresholds of 2-fold higher or lower expression than the averaged expression in non-tumorous liver tissue. Relative expression in primary hepatocytes (PHH) is also shown (grey bar). 18s-rRNA was used as a housekeeping gene for internal normalization. Data are presented as mean ± SD. (D) Nup98 and p21 expression are correlated in human hepatocellular carcinoma (HCC). Scatterplot depicts relative Nup98 (x-axis) and p21 (y-axis) mRNA levels measured by qRT-PCR in tumors of the same cohort of HCC patients as in (C). Pearson’s correlation coefficient was used as a measure of association.