Article Figures & Data
Figures
- FIG. 1.
Phosphorylation of Ser170/172 by CK2 is required for TIF-IA activity. (A) TIF-IA is phosphorylated by CK2 in vitro. Immunopurified Flag-tagged TIF-IA was in vitro phosphorylated by recombinant human CK2 in the absence or presence of the CK2 inhibitor TBB or DMAT as indicated. After SDS-polyacrylamide gel electrophoresis, 32P-labeled TIF-IA was visualized by autoradiography. The amino acid sequences of TIF-IA encompassing putative CK2 consensus sites (S/T-X-X-D/E) from human (H.s.), mouse (M.m.), Xenopus laevis (X.l.), and chicken (G.g.) are shown above. Phosphoacceptor sites in TIF-IA (serines 170 and 172) are highlighted. DMSO, dimethyl sulfoxide. (B) Serines 170 and 172 are phosphorylated in vivo. HEK293T cells expressing Flag-tagged TIF-IA or TIF-IAS170/172A were metabolically labeled with [32P]orthophosphate for 3 h. TIF-IA was immunoprecipitated and subjected to two-dimensional tryptic phosphopeptide mapping. Phospho-Ser170/172 was contained in three phosphopeptide spots (spots h, i, and j) because of partial cleavage at two adjacent nonoptimal trypsin cleavage sites (Arg145 and Lys163) in the vicinity of Ser170/172. (C) CK2 phosphorylates Ser170/172. Phosphorylation of immunopurified Flag-tagged TIF-IA and TIF-IAS170/172A from mock- or DMAT-treated (50 μM, 2 h) cells was monitored on immunoblots using anti-Flag antibodies or antibodies specific for phospho-Ser170/172 (anti-TIF-IApS170/172). (D) Transcription of rDNA depends on CK2 activity. RNA from untreated or mock- or DMAT-treated (25 and 50 μM, 2 h) HEK293T cells was isolated, and levels of 45S pre-rRNA were monitored on Northern blots. As a loading control, blots were reprobed for β-actin mRNA. (E) Phosphorylation of Ser170/172 is required for TIF-IA activity. Flag-tagged TIF-IA, TIF-IAS170/172A, or TIF-IAS170/172D was immunopurified from HEK293T cells and assayed for transcriptional activity in a TIF-IA-responsive nuclear extract from confluent cells. A silver-stained SDS gel of wild-type and mutant TIF-IAs used in the transcription assays is shown below. (F) Activation of a Pol I minigene depends on TIF-IA phosphorylation at Ser170/172. HEK293T cells were cotransfected with a ribosomal minigene construct (pHrP2-BH) and the indicated amounts of expression vectors encoding Flag-TIF-IA or Flag-TIF-IAS170/172A. Transcript levels were analyzed on Northern blots, normalized to β-actin mRNA, and represented in the bar diagram below. The immunoblot shows the expression levels of Flag-TIF-IA and Flag-TIF-IAS170/172A. (G) Wild-type TIF-IA, but not TIF-IAS170/172A, activates transcription of a Pol I-driven luciferase reporter. HEK293T cells were transfected with pHr-IRES-Luc-T4 and expression vectors encoding Flag-TIF-IA or Flag-TIF-IAS170/172A. The bar diagram shows the level of luciferase activity determined in three independent experiments. The immunoblot shows the expression levels of Flag-TIF-IA and Flag-TIF-IAS170/172A.
- FIG. 2.
Phosphorylation of TIF-IA at Ser170/172 is required for nucleolar integrity and cell proliferation. (A) Mutation of TIF-IA at Ser170/172 causes nucleolar disruption. TIF-IAfl/fl MEFs expressing no ectopic TIF-IA (top panels), Flag-tagged TIF-IA (middle panels), or TIF-IAS170/172A (bottom panels) were infected with MSCV (TIF-IAfl/fl) or MSCV-Cre (TIF-IA−/−). Six days after infection, Pol I and both endogenous and recombinant TIF-IA were visualized by indirect immunofluorescence with antibodies against Pol I (RPA194) and TIF-IA. DAPI, 4′,6′-diamidino-2-phenylindole. (B) Nucleolar localization of TIF-IAS170/172A depends on the presence of endogenous TIF-IA. TIF-IAfl/fl MEFs expressing creERT2 as well as no ectopic TIF-IA (top panels), GFP-tagged TIF-IA (middle panels), or TIF-IAS170/172A (bottom panels) were either mock treated (TIF-IAfl/fl) or treated with 4-OHT to induce depletion of endogenous TIF-IA (TIF-IA−/−). After 4 days, cells were fixed and localization of ectopic TIF-IA was monitored by GFP fluorescence. (C) Ser170/172 phosphorylation is required for cell proliferation. TIF-IAfl/fl MEFs expressing creERT2 were either mock or 4-OHT treated (TIF-IAfl/fl or TIF-IA−/−), and cell numbers were determined at the indicated time points. Error bars are from counting of three different culture dishes. (D) Abrogation of Ser170/172 phosphorylation arrests cells in the G0/G1 phase of the cell cycle. TIF-IAfl/fl or TIF-IA−/− MEFs were stained with propidium iodide and processed for fluorescence-activated cell sorting analysis. The percentage of cells in the G0/G1 phase was quantified with Modfit LT software. Error bars are from two independent experiments. (E) Mutation of TIF-IA at Ser170/172 leads to cell cycle arrest. Lysates from TIF-IAfl/fl or TIF-IA−/− cells were analyzed on immunoblots with antibodies against the indicated cyclins and Pol I (RPA116).
- FIG. 3.
Ser170/172 phosphorylation facilitates the release of TIF-IA from Pol I. (A) Phosphorylation of TIF-IA by CK2 impairs the interaction between TIF-IA and Pol I. Flag-tagged wild-type or mutant TIF-IA was immunoprecipitated from NIH 3T3 cells, and coprecipitated TIF-IB/SL1 and Pol I were visualized on Western blots using antibodies against TAFI95, RPA116, or PAF53. (B) Inhibition of CK2 increases the association of TIF-IA with Pol I. Pol I was immunoprecipitated with anti-RPA194 antibodies from mock- or DMAT-treated (25 and 50 μM) HEK293T cells, and Pol I and coprecipitated TIF-IA were assayed on immunoblots. (C) Inhibition of CK2 does not enhance the association of TIF-IAS170/172A with Pol I. HEK293T cells overexpressing Flag-tagged wild-type or mutant TIF-IA were either mock treated or treated with DMAT (50 μM, 2 h). Pol I was immunoprecipitated, and the levels of Pol I and coprecipitated Flag-TIF-IA were assayed on immunoblots. (D) Phosphorylation at Ser170/172 releases TIF-IA from Pol I. Flag-tagged TIF-IA and TIF-IAS170/172A were immunoprecipitated with anti-Flag antibodies and eluted with the Flag peptide. Pol I/TIF-IA complexes were precipitated from the eluate with anti-RPA194 antibodies. Ser170/172 phosphorylation of similar amounts of TIF-IA bound to Pol I (a) and “free” TIF-IA in the eluate (b) was visualized on immunoblots using the phospho-specific TIF-IApS170/172 antibody.
- FIG. 4.
Covalent association of TIF-IA with Pol I impairs rDNA transcription. (A) TIF-IA/RPA43 is incorporated into Pol I. Western blots of TIF-IAfl/fl MEFs infected with MSCV (TIF-IAfl/fl) or MSCV-Cre (TIF-IA−/−) showing Cre-mediated depletion of endogenous TIF-IA and expression of exogenous TIF-IA. Blots were probed with antibodies specific for TIF-IA, Cre recombinase, the Flag epitope, or β-actin. In the bottom panel, Flag-tagged TIF-IA, TIF-IAS170/172A, and TIF-IA/RPA43 were immunoprecipitated, and coprecipitated Pol I (RPA116) was visualized on immunoblots. (B) Fusion of TIF-IA with Pol I inhibits cell proliferation. TIF-IAfl/fl MEFs expressing TIF-IA/RPA43 and creERT2 were either mock or 4-OHT treated (TIF-IAfl/fl or TIF-IA−/−), and cell numbers were determined at the indicated time points. Error bars represent numbers from three different experiments. (C) Tethering of TIF-IA to RPA43 causes nucleolar disruption. TIF-IAfl/fl MEFs coexpressing creERT2 and GFP-TIF-IA/RPA43 were mock or 4-OHT treated (TIF-IAfl/fl or TIF-IA−/−), and localization of TIF-IA/RPA43 was monitored by GFP fluorescence. DAPI, 4′,6′-diamidino-2-phenylindole. (D) Overexpression of TIF-IA/RPA43 leads to mislocalization of Pol I. MEFs expressing creERT2 as well as Flag-TIF-IA/RPA43 were either mock treated (TIF-IAfl/fl) or treated with 4-OHT (TIF-IA−/−), and distribution of Pol I (RPA194) was visualized by indirect immunostaining. (E) TIF-IAS170/172A and TIF-IA/RPA43 do not rescue Pol I transcription in TIF-IA-deficient cells. TIF-IAfl/fl MEFs expressing ectopic TIF-IA, TIF-IAS170/172A, or TIF-IA/RPA43 were depleted from endogenous TIF-IA by MSCV-Cre infection and pulse-labeled with FUrd, and FUrd incorporation into nascent RNA was monitored by indirect immunofluorescence. Cells were costained with anti-Cre antibodies. FUrd staining of at least 200 Cre-positive cells was quantified using NIS-Elements BR imaging software (Nikon); the value for TIF-IA−/− MEFs expressing wild-type TIF-IA was set to 100%.
- FIG. 5.
Dissociation of TIF-IA from elongating Pol I is required for efficient rDNA transcription. (A) Enhanced association with Pol I decreases the mobility of TIF-IA. Graphs of FRAP analyses showing the recovery kinetics of GFP-tagged TIF-IA, TIF-IAS170/172A, TIF-IA/RPA43, or RPA43 in HeLa cells. In each case values from at least 15 cells were averaged. (B) Inhibition of CK2 decreases the mobility of TIF-IA. FRAP recovery kinetics of GFP-TIF-IA in HeLa cells that were either mock treated (black curve) or treated with 50 μM DMAT for 2 h (red curve). (C) Inhibition of S170/172 phosphorylation reduces the dynamics of Pol I. FRAP recovery kinetics of GFP-RPA194 was monitored in mock- or DMAT-treated HeLa cells. (D) Ablation of Ser170/172 phosphorylation leads to enrichment of Pol I at the rDNA promoter. Cross-linked chromatin from mock- or DMAT-treated NIH 3T3 cells was immunoprecipitated with antibodies against TIF-IA or Pol I, and precipitated DNA was assayed by quantitative PCR, amplifying the rDNA promoter (−160/−1, primer pair A) or the coding region (+451/+670 and +8124/+8268, primer pairs B and C, respectively). The bar diagram shows the amounts of precipitated rDNA in mock-treated (green bars) and DMAT-treated (red bars) cells normalized to input rDNA. Mean values from three independent experiments are shown. (E) The association of H3K9me3, but not H3K4me3, with the pre-rRNA coding region is altered upon inhibition of CK2. Data from ChIP experiments showing rDNA occupancy of H3K4me3 and H3K9me3 in mock-treated (green bars) and DMAT-treated (red bars) cells. Immunoprecipitated DNA was assayed by quantitative PCR, amplifying the coding region (positions +8124 to +8268). The levels of methylated H3K4 and K9 were normalized to that of histone H3, and values for mock-treated cells were set as 1. Error bars represent standard deviations of three independent experiments.
- FIG. 6.
FCP1 stimulates rDNA transcription by counteracting CK2-mediated phosphorylation of TIF-IA. (A) FCP1 associates with the rDNA promoter. Cross-linked chromatin from HEK293T cells was immunoprecipitated with antibodies against FCP1 (dark bars) or control immunoglobulin Gs (light bars), and precipitated DNA was assayed by quantitative PCR with primers amplifying the rDNA promoter (primer pair A) or the coding region (primer pairs B and C). Amounts of precipitated DNA normalized to input rDNA are shown. Mean values are from three independent experiments. (B) FCP1 interacts with Pol I. Pol I was immunoprecipitated from HEK293T cells, and coprecipitated FCP1 was monitored on Western blots with anti-FCP1 antibodies. (C) FCP1 interacts with TIF-IA. Lysates from HEK293T cells overexpressing Flag-FCP1 were incubated with anti-Flag antibodies, and coprecipitation of TIF-IA with Flag-FCP1 was analyzed on Western blots. (D) FCP1 dephosphorylates Ser170/172 of TIF-IA in vitro. Flag-tagged FCP1 was immunopurified from HEK293T cells, and phosphatase activity was assayed using p-nitrophenyl phosphate as a substrate. To monitor dephosphorylation of Ser170/172, TIF-IA was 32P labeled with CK2 and incubated with Flag-FCP1 or the same activity of CIAP. Phospho-TIF-IA was visualized by autoradiography (upper panel), and total TIF-IA was visualized by Ponceau staining (lower panel). (E) FCP1 dephosphorylates Ser170/172 in vivo. The expression of GFP-FCP1 and Flag-TIF-IA in HEK293T cells was analyzed on Western blots (top two panels). After immunoprecipitation with anti-Flag antibodies, phosphorylation of TIF-IA was monitored with antibodies that specifically recognize phospho-Ser170/172 (pS170/172) and phospho-Ser44 (pS44). The arrowhead marks the specific band recognized by the anti-phospho-Ser44 antibody. (F) Overexpression of FCP1 increases the formation of Pol I/TIF-IA complexes. Pol I was precipitated from HEK293T cells expressing Flag-tagged TIF-IA or TIF-IAS170/172A in the absence or presence of Flag-FCP1, and coprecipitated Flag-TIF-IA and Flag-FCP1 were visualized on Western blots. (G) Knockdown of FCP1 disrupts Pol I/TIF-IA complexes. U2OS cells were transfected with control duplex siRNA directed against GFP (Ctrl siRNA) or with FCP1-specific siRNA. Cell were lysed, and Pol I was immunoprecipitated with anti-RPA194 antibodies. Amounts of coprecipitated TIF-IA as well as levels of FCP1, RPA116, and TIF-IA in the inputs were monitored on Western blots. (H) FCP1 is required for Pol I transcription. Cells were transfected with control or FCP1-specific siRNAs. The knockdown of FCP1 was monitored on immunoblots. The synthesis of pre-rRNA was measured by quantitative reverse transcription-PCR and normalized to glyceraldehyde-3-phosphate dehydrogenase. Mean values are from two independent experiments. (I) Modulation of FCP1 levels affects Pol I transcription. HEK293T cells were transfected with the indicated amounts of expression vectors encoding Flag-FCP1, and pre-rRNA was measured by quantitative reverse transcription-PCR. Mean values are from three independent experiments. The Western blots below show the expression levels of Flag-FCP1 and Pol I (RPA116).
- FIG. 7.
Model depicting the role of reversible phosphorylation of TIF-IA in the Pol I transcription cycle. UBF and TIF-IB/SL1 bound to the rDNA promoter recruit Pol Iβ, the initiation-competent subpopulation of Pol I containing CK2, to rDNA (step 1). After transcription initiation and promoter escape, CK2 phosphorylates Pol I-associated TIF-IA at Ser170/172 (step 2). This phosphorylation weakens the association of TIF-IA, leading to dissociation of the Pol I/TIF-IA complex (step 3). Then, Ser170/172 is dephosphorylated by FCP1, which enables TIF-IA to assemble into a new initiation complex and to reinitiate transcription (step 4).
