Alterations in nucleolar structure and gene expression programs in prostatic neoplasia are driven by the MYC oncogene

Abstract

Increased nucleolar size and number are hallmark features of many cancers. In prostate cancer, nucleolar enlargement and increased numbers are some of the earliest morphological changes associated with development of premalignant prostate intraepithelial neoplasia (PIN) lesions and invasive adenocarcinomas. However, the molecular mechanisms that induce nucleolar alterations in PIN and prostate cancer remain largely unknown. We verify that activation of the MYC oncogene, which is overexpressed in most human PIN and prostatic adenocarcinomas, leads to formation of enlarged nucleoli and increased nucleolar number in prostate luminal epithelial cells in vivo. In prostate cancer cells in vitro, MYC expression is needed for maintenance of nucleolar number, and a nucleolar program of gene expression. To begin to decipher the functional relevance of this transcriptional program in prostate cancer, we examined FBL (encoding fibrillarin), a MYC target gene, and report that fibrillarin is required for proliferation, clonogenic survival, and proper ribosomal RNA accumulation/processing in human prostate cancer cells. Further, fibrillarin is overexpressed in PIN lesions induced by MYC overexpression in the mouse prostate, and in human clinical prostate adenocarcinoma and PIN lesions, where its expression correlates with MYC levels. These studies demonstrate that overexpression of the MYC oncogene increases nucleolar number and size and a nucleolar program of gene expression in prostate epithelial cells, thus providing a molecular mechanism responsible for hallmark nucleolar alterations in prostatic neoplasia.

Figure 1

MYC controls nucleolar size and number in prostate cancer cells. A: Nucleolar enlargement in a high-grade PIN lesion of Lo-MYC mouse. Electron micrographs showing markedly enlarged nucleolus in a PIN cell from a Lo-MYC mouse (right) as compared with wild-type mouse (left). Arrows indicate nucleoli. Original magnification ×8000. B: Nucleoli were visualized by AgNOR staining and enumerated in 50 to 80 cells after MYC knockdown. Graphs showing the distribution of the number of nucleoli per cell after MYC knockdown (black bars), as compared to control cells (gray bars).

Figure 2

MYC regulates the expression of nucleolar genes, and nucleolar function in prostate cancer cells. A: Genome-wide mRNA expression changes were assessed 48 hours after MYC knockdown in PC3, DU145, and LNCaP cells using Agilent's 44k whole-genome array platform. GSEA was applied based on pathways and biological categories defined by GO, with a false discovery rate cutoff of 1% for significance. Selected categories (eg, all three cell types affected, two different types affected but the other not affected, etc.) of enriched gene sets after MYC knockdown are shown. B: Western blots showing decreased expression of several nucleolar proteins after transfection with siRNA targeting MYC, or control scrambled siRNA (Scr). Tubulin, whose expression is not MYC responsive, was used as a loading control. C: The relative decrease in the levels of pre-rRNA (black bars), 5.8s rRNA (dark grey bars), 18s rRNA (grey bars), and 28s (white bars) rRNA were quantified by real-time PCR after siRNA-mediated MYC knockdown. Data were log₂ transformed, and error bars show the SD of three independent experiments. *P < 0.03, **P < 0.01. D: Elevated expression of 5.8s rRNA, 18s rRNA, and 28s rRNA in Lo-MYC mice, as compared to wild-type controls.

Figure 3

MYC regulates fibrillarin expression in prostate cancer cells. A: Fibrillarin protein overexpression correlates with MYC protein overexpression in Lo-MYC mice. Upper left panel, medium power (original magnification, ×100) view of Lo-MYC mouse showing a region containing mixed PIN and normal glands. MYC staining occurs only in PIN cells. Upper right panel, medium power view of an adjacent section of that in upper left panel, showing heterogeneous staining for fibrillarin that corresponds well with regions of MYC staining in upper left panel. Lower panels are higher power views (original magnification, ×400) corresponding to boxed areas in upper panels. Note the correspondence between fibrillarin overexpression and MYC staining. Note also that, as expected, fibrillarin staining occurs predominantly in subnuclear structures consistent with nucleoli, which are markedly enlarged in cells staining positive for MYC. B: Elevated fibrillarin mRNA expression Lo-MYC mice, as compared to wild-type controls. C: Decrease in fibrillarin mRNA following MYC knockdown. Data were log₂ transformed, and error bars show the SD of three independent experiments. **P < 0.005. D: ChIP was performed on PC3, LNCaP, P493, and H9 embryonic stem cells using an anti-MYC antibody (black bars) or an IgG control (white bars). A region upstream of the FBL gene, as well as a known negative control region were amplified by quantitative real-time PCR.

Figure 4

Fibrillarin is required for prostate cancer cells proliferation, and proper rRNA processing. A: Efficacy of siRNA-mediated fibrillarin knockdown (+) in CWR22rv1, LNCaP, DU145, and PC3 prostate cancer cell lines was verified by Western blotting. Cells were transfected with a nontargeting siRNA pool as a control (−). B: In clonogenic survival assays, colony formation was measured in cells transfected with fibrillarin siRNA (black bars), or cells transfected with control siRNA. Error bars represent the SD of three independent experiments. C: The relative decrease in the levels of pre-rRNA (black bars), 5.8s rRNA (dark grey bars), 18s rRNA (grey bars), and 28s (white bars) rRNA were quantified by real-time PCR after fibrillarin knockdown with siRNA. Data were log₂ transformed, and error bars show the SD of three independent experiments. *P < 0.04, **P < 0.01.

Figure 5

Fibrillarin is elevated in prostate adenocarcinoma, PIN, and proliferative inflammatory atrophy. A: Elevated fibrillarin mRNA expression in primary prostate cancer tissues, as compared to matched normal prostate tissue. FBL expression in prostate cancer tissues was measured by quantitative real-time PCR and normalized to TBP, and matched normal tissue. B: Western blot showing total fibrillarin expression in matched normal (N) and tumor (T) prostate tissues. C: Fibrillarin protein expression is elevated in PIN. D: Fibrillarin protein expression is elevated in a primary prostate cancer lesion. E: Combined dot plot and box-and-whisker graph of the fibrillarin nuclear area ratio of normal prostate glands, atrophy, PIN, and primary and metastatic carcinoma. Each dot represents the average score from all TMA cores for a given diagnosis from a single patient. *P < 0.005, **P < 0.001.