Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2015 Dec 3;163(6):1375-87.
doi: 10.1016/j.cell.2015.11.005.

Redox Signaling by the RNA Polymerase III TFIIB-Related Factor Brf2

Affiliations

Redox Signaling by the RNA Polymerase III TFIIB-Related Factor Brf2

Jerome Gouge et al. Cell. .

Abstract

TFIIB-related factor 2 (Brf2) is a member of the family of TFIIB-like core transcription factors. Brf2 recruits RNA polymerase (Pol) III to type III gene-external promoters, including the U6 spliceosomal RNA and selenocysteine tRNA genes. Found only in vertebrates, Brf2 has been linked to tumorigenesis but the underlying mechanisms remain elusive. We have solved crystal structures of a human Brf2-TBP complex bound to natural promoters, obtaining a detailed view of the molecular interactions occurring at Brf2-dependent Pol III promoters and highlighting the general structural and functional conservation of human Pol II and Pol III pre-initiation complexes. Surprisingly, our structural and functional studies unravel a Brf2 redox-sensing module capable of specifically regulating Pol III transcriptional output in living cells. Furthermore, we establish Brf2 as a central redox-sensing transcription factor involved in the oxidative stress pathway and provide a mechanistic model for Brf2 genetic activation in lung and breast cancer.

PubMed Disclaimer

Figures

None
Graphical abstract
Figure 1
Figure 1
Structure of the Brf2-TBP/DNA Ternary Complex (A) Overview of the Brf2-TBP/U6 promoter structure. DNA template and non-template strands are in blue and cyan respectively. Dashed lines represent disordered regions or regions not present in the crystallization construct. (B) Schematic of Brf2 domain organization. See also Figures S1, S2, and S7 and Tables S1 and S2.
Figure 2
Figure 2
Architecture of the Human Pol III PIC Model of a Pol III PIC (Vannini and Cramer, 2012) generated using the Brf2-TBP/DNA complex reveals that the path of the downstream DNA points toward the Pol III-specific subunits C39 and C62, and resembles the path observed in yeast and human Pol II PIC (He et al., 2013, Mühlbacher et al., 2014). See also Figure S7.
Figure 3
Figure 3
Brf2/DNA Sequence-Specific Interactions (A) Close-up view of the TATA box (yellow), downstream flanking region and sequence-specific interactions with Brf2. DNA template and non-template strands are in blue and cyan respectively. (B) Substitutions at positions +3 and +4 of the wild-type (circled in red) U6-2 promoter decrease binding of a R110A mutant, in particular when a T nucleobase is present at position +3 on the non-template strand (in cyan). R110A versus wild-type (WT) is the ratio between the percentage of binding of the mutant versus wild-type Brf2 proteins. (C) Close-up view of the TATA box (yellow), upstream flanking region and sequence-specific interactions with Brf2. DNA template and non-template strands are in blue and cyan respectively. (D) Substitutions at positions −3 and −4 of the wild-type (circled in red) U6-2 promoter reveal more efficient complex formation with a pyrimidine nucleobase and a C nucleobase at positions −3 and −4 of the template strand, respectively. (B and D) The intensity of the complex formed with TBP, U6-2 non mutated sequence and wild-type Brf2 (lane 1) was used as a reference for relative quantification. Indicates samples that were quantified relative to a distinct wild-type sequence reference not shown on the figure. Representative gels of three independent experiments. The data shown are the mean values and SE of three independent experiments. In the insets, 10 μl of a typical binding reaction (25 μl total) with Brf2 wild-type or Brf2 mutants were loaded on a SDS-PAGE gel and stained with Coomassie-blue, confirming that equal amounts of protein of comparable quality were used for EMSA assays. See also Figures S2 and S7.
Figure 4
Figure 4
The Brf2 Molecular Pin (A) The Brf2 TBP anchor domain but not the molecular pin is essential for Brf2-TBP interaction in absence of the DNA, as shown by a pull-down assay. (B) The Brf2 TBP anchor domain and the molecular pin are essential for the formation of a Brf2-TBP/DNA complex, as shown in an EMSA. (C) Close-up view of the Brf2 molecular pin at the interface between the Brf2 C-cyclin repeat, TBP, and the DNA. See also Figures S3 and S7.
Figure 5
Figure 5
Brf2 Redox Regulation (A) Close-up view of C361 at the ternary interface between the Brf2 C-cyclin repeat, TBP, and the upstream edge of the TATA box. Yellow dots represent the van der Waals radius of the sulfur atom. (B) Representative EMSA of Brf2-TBP/DNA complexes upon pre-incubation of Brf2 proteins with the alkylating agent iodoacetamide. The IC band was used for loading normalization. Indicates addition of the reducing agent after the oxidative treatment during complex assembly. (C) Representative EMSA of Brf2-TBP/DNA complexes upon removal of reducing agent (DTT) and incubation over time. The IC band was used for loading normalization. Indicates addition of the reducing agent after the oxidative treatment during complex assembly. (D) Representative EMSA of Brf2-TBP/DNA complexes upon pre-incubation of Brf2 proteins with H2O2. The IC band was used for loading normalization. Indicates addition of the reducing agent after the oxidative treatment during complex assembly. (E) Representative EMSA of Brf2-TBP/DNA complexes upon pre-incubation of Brf2 proteins with gradients of oxidized/reduced glutathione (GSSG:GSH). The IC band was used for loading normalization. Indicates addition of the reducing agent after the oxidative treatment during complex assembly. See also Figures S3, S4, and S7.
Figure 6
Figure 6
Brf2-Dependent Transcription Is Redox Regulated in Living Cells (A) qRT-PCR analysis shows that Brf2-dependent-transcripts (SeCys p-tRNA, RPPH1, RNA7SK, and U6 snRNA) are globally downregulated during oxidative stress, while a Brf1-dependent transcript (Leu p-tRNA) remains unchanged. (B) SeCys p-tRNA levels are strongly reduced in cells challenged with t-BHP relative to the unchallenged cells (as highlighted by gray and black lines, respectively) in a dose- and time-dependent manner, as measured by qRT-PCR. (C) SeCys p-tRNA levels rapidly recover upon removal of the exogenous oxidative stress inducer, as measured by qRT-PCR. Wash indicates replacement of media containing t-BHP with fresh media. (D) Effects of overexpression of Brf2 and Brf2 mutants (inset) on SeCys p-tRNA levels during oxidative stress, as measured by qRT-PCR. The numbers indicated on the histograms represent the percentage of reduction of selenocysteine tRNA levels, while if numbers are indicated above the histograms they represent the percentage of increase. Cumulative data of at least three experiments, mean + SEM. Unpaired t test: p < 0.05; ∗∗p < 0.005; ∗∗∗p < 0.0005; ∗∗∗∗p < 0.0001. p > 0.05 were deemed not significant and values were not reported. See also Figures S5 and S7.
Figure 7
Figure 7
Selenoproteins Levels and Resistance to Oxidative Stress Are Regulated in a Brf2-Dependent Manner (A) Manipulation of Brf2 protein levels affects selenoproteins expression levels during oxidative stress in MRC5 and A549 cells. In the upper insets, a western blot analysis of Nrf2 confirms induction of oxidative stress with 50 μM and 100 μM t-BHP in MRC5 and A549 cells, respectively. A western blot analysis of Brf2 immunoprecipitation from 107 MRC5 or A549 cells is shown in the lower insets (IP). (B) Overexpression of Brf2 in MRC5 cells challenged with t-BHP results in decreased apoptosis as measured by FACS analysis via annexin V-FITC/PI staining. The y axis represents the % of apoptotic cells, including both cells in early (annexin V-positive and PI-negative) and late (annexin V-positive and PI-positive) apoptosis. (C) Effects of overexpression of Brf2 wild-type and mutants on acquired resistance to apoptosis in MRC5 cells as measured by FACS analysis via annexin V-FITC/PI staining. The y axis represents the % of apoptotic cells, including both cells in early (annexin V-positive and PI-negative) and late (annexin V-positive and PI-positive) apoptosis. (D) Lowering Brf2 protein levels by siRNA in A549 cells challenged with t-BHP results in an increased cellular commitment to apoptosis as measured by FACS analysis via annexin V-FITC/PI staining. Inset: a western blot analysis of Brf2 immunoprecipitation from 107 A549 cells shows siRNA-induced Brf2 protein level reduction. The y axis represents the % of apoptotic cells, including both cells in early (annexin V-positive and PI-negative) and late (annexin V-positive and PI-positive) apoptosis. See also Figures S5 and S7.
Figure S1
Figure S1
General Conservation of the Architecture of TFIIB and TFIIB-like Factors, Related to Figure 1 (A) Schematic of the architecture of the Brf2-TBP/DNA complexes and sequences of the DNA scaffold used for crystallization. (B) Superimposition of Brf2-TBP/DNA and TFIIB-TBP/DNA (PDB: 1C9B). The structures were superimposed using TBP as a template for structural alignment. (C) Superimposition of the Brf2-TBP/U6-2 (blue), Brf2-TBP/TRNAU1 (grey) and Brf2-TBP/RPPH1 (green) structures. The structures were superimposed using TBP as a template for structural alignment. (D) Final electron density contoured at 1.2 s surrounding a tract of double-stranded DNA. (E) Brf2 sequence conservation and domain organization of Homo sapiens (H.s.) Mus musculus (M.m.), and Danio rerio (D.r.).
Figure S2
Figure S2
Sequence Alignments of Brf2 Promoters and Protein, Related to Figures 1 and 3 (A) Brf2-dependent promoters (40 nucleobase long centered around putative TATA boxes) were aligned using MEME (Bailey et al., 2009) by searching for a 16 nucleobase long consensus motif. (B) Sequence and domain conservation between Brf2, Brf1 and TFIIB. Color-coding is as in Fig. 1. Brf2-Brf1 alignments are based on sequence conservation, while Brf2-TFIIB is based on a structural alignment. (C) Two views of the specific interaction of Brf2 R110 and A108 with a TG (U6-2 in pale green) and TC (RPPH1 in light grey) dinucleotide step. In presence of the TG sequence, the T on the nontemplate strand (in red) is left unstacked at its downstream edge. In presence of a TC sequence, no local distortions of the DNA are observed. The two structures were superimposed by structurally aligning the Brf2 N-terminal cyclin repeats.
Figure S3
Figure S3
Modular Functions of the Brf2 CTD, Related to Figures 4 and 5 (A) EMSA with serial Brf2 C-terminal deletion mutants showing that the region comprised between residues 289-311 of Brf2 is involved in direct binding to the upstream transcription factor SNAPc. (B) A conserved surface of TBP is utilized by different TBP associated factors. The TBP surface buried upon interaction with the associated factor is colored in pink for human Brf2 (orange), in cyan for yeast Brf1 (PDB id: 1NGM, turquoise), in green for yeast TFIIA (PDB id: 1NH2, green) and in blue for yeast TAF1 (PDB id: 4B0A, blue). (C) Structural conservation between Brf2 C361 part of the molecular pin, and C59 part of a short helical motif of the p-50 subunit of the NF-kB transcription factor (PDB id: 1NFK). (D) Brf2 oxidative-mimic mutation C361D does not hinder Brf2-TBP complex formation in absence of the DNA, as shown by pull-down assay. “IN” indicates the input and “empty resin” the eluted untagged TBP binding non-specifically to the resin. (E) EMSA shows that formation a functional Brf2-TBP/DNA complex is severely impaired in Brf2 oxidative-mimic mutant C361D. “IN” indicates the input and “empty resin” the eluted untagged TBP binding non-specifically to the resin. (F) Fluorescence polarization saturation binding assay shows virtually no reduction in affinity of the Brf2 C361A mutant and an approximately 50-fold reduction in affinity of Brf2 C361D mutant for TBP/DNA complexes.
Figure S4
Figure S4
Mass Spectrometry Analysis of Brf2 Redox Modifications, Related to Figure 5 Biologically relevant oxidation states of C361 were confirmed by MS/MS as either unmodified, trapped with glutathione (-SS-Glu) or dimedone (Dmd). The precursor ions, errors and ion scores are indicated below the annotated fragmentation mass spectra. For clarity, only prominent fragment ions are marked.
Figure S5
Figure S5
Brf2 is a Redox Sensor in Living Cells, Related to Figures 6 and 7 (A) SeCys m-tRNA levels are reduced during oxidative stress in a Brf2-dependent manner, as monitored via four-leaf clover PCR (Honda et al., 2015). Samples labeled empty vector and Brf2 represent transient over-expressions. (B) Overexpression of Brf2 in MCF10A cells challenged with t-BHP results in decreased apoptosis as measured by FACs analysis via Annexin V-FITC/PI staining. (C) Overexpression of Brf2 affects selenoproteins expression levels during oxidative stress in MCF10A cells. (D) Effects of overexpression of Brf2 and Brf2 mutants on acquired resistance to apoptosis in MCF10A cells as measured by FACs analysis via Annexin V-FITC/PI staining.
Figure S6
Figure S6
Brf2-Dependent Reduction of SeCys p-tRNA and Enhanced Apoptosis in A549 Cells, Related to Figure 7 (A) Two individual Brf2 siRNAs cause a severe reduction of SeCys p-tRNA in A549 cells challenged with t-BHP, an effect that is fully rescued by concomitant overexpression of a siRNA resistant form of Brf2. (B) Two individual Brf2 siRNAs elicit a strong sensitization towards t-BHP in A549 cells, an effect that is fully rescued by concomitant overexpression of a siRNA resistant form of Brf2.
Figure S7
Figure S7
Mechanism of the Redox-Dependent Brf2 Blockade during Oxidative Stress and Carcinogenesis, Related to Figures 1, 2, 3, 4, 5, 6, and 7 During normal growth conditions (A) Brf2- and Nrf2-dependent transcripts are synthesized at basal levels. Upon moderate oxidative stress (B), Nrf2 is activated and Nrf2-dependent transcripts upregulated. Concomitantly, Brf2-dependent transcription, including SeCys tRNAs, is rapidly downregulated via redox-dependent modifications of Brf2. The pre-existing pool of SeCys tRNA is sufficient to sustain synthesis of selenoproteins. Upon prolonged oxidative stress (C), SeCys tRNA levels become limiting while, simultaneously, selenoprotein’s mRNAs continue to be highly upregulated by Nrf2. In this scenario, compromised synthesis of selenoproteins drives the cells into apoptosis. In cancer cells (D), the Nrf2 pathway is constitutively activated and contributes to the observed resistance of cancerous cells to higher than normal concentrations of reactive oxygen species. Under these circumstances, Brf2 overexpression is required to overcome the innate redoxdependent blockade, ensuring elevated synthesis of SeCys tRNAs and, ultimately, enabling cancer cells to evade apoptosis under prolonged oxidative stress.

References

    1. Anandapadamanaban M., Andresen C., Helander S., Ohyama Y., Siponen M.I., Lundström P., Kokubo T., Ikura M., Moche M., Sunnerhagen M. High-resolution structure of TBP with TAF1 reveals anchoring patterns in transcriptional regulation. Nat. Struct. Mol. Biol. 2013;20:1008–1014. - PMC - PubMed
    1. Anestal K., Prast-Nielsen S., Cenas N., Arner E.S. Cell death by SecTRAPs: thioredoxin reductase as a prooxidant killer of cells. PloS ONE. 2008;3:e1846. - PMC - PubMed
    1. Arimbasseri A.G., Rijal K., Maraia R.J. Comparative overview of RNA polymerase II and III transcription cycles, with focus on RNA polymerase III termination and reinitiation. Transcription. 2013;5:e27639. - PMC - PubMed
    1. Biaglow J.E., Miller R.A. The thioredoxin reductase/thioredoxin system: novel redox targets for cancer therapy. Cancer Biol. Ther. 2005;4:6–13. - PubMed
    1. Brigelius-Flohe R., Flohe L. Basic principles and emerging concepts in the redox control of transcription factors. Antioxid. Redox. Signal. 2011;15:2335–2381. - PMC - PubMed

Publication types