End-targeting proteomics of isolated chromatin segments of a mammalian ribosomal RNA gene promoter

Abstract

The unbiased identification of proteins associated with specific loci is crucial for understanding chromatin-based processes. The proteomics of isolated chromatin fragment (PICh) method has previously been developed to purify telomeres and identify associated proteins. This approach is based on the affinity capture of endogenous chromatin segments by hybridization with oligonucleotide containing locked nucleic acids. However, PICh is only efficient with highly abundant genomic targets, limiting its applicability. Here we develop an approach for identifying factors bound to the promoter region of the ribosomal RNA genes that we call end-targeting PICh (ePICh). Using ePICh, we could specifically enrich the RNA polymerase I pre-initiation complex, including the selectivity factor 1. The high purity of the ePICh material allowed the identification of ZFP106, a novel factor regulating transcription initiation by targeting RNA polymerase I to the promoter. Our results demonstrate that ePICh can uncover novel proteins controlling endogenous regulatory elements in mammals.

Figure 1. Rules for designing capture probes for the formation of stable oligo-DNA hybrids.

(a) Outline of the plasmid pull-down assay. Linearized plasmids (grey line) with or without the target DNA sequence (black thick line) are mixed with LNA probes conjugated to spacers (straight red line) and desthiobiotin (yellow circle). After denaturation, the mixture is incubated at 37 °C for hybridization. Hybrids are captured using streptavidin beads. (b) Purification of the telomere sequence containing plasmid with the telomere-specific probe. Each fraction (Input, Flow-through (Flow.), Eluate (El.)) was analysed by agarose gel electrophoresis and EtBr staining. The positions of plasmid containing telomere DNA and the empty vector are indicated. Linear plasmid with telomere sequence after digestion with SpeI was used for an assay (lane 1–3 in the left panel), and linear plasmid digested by ScaI was used for the other assay (lane 4–6 in the right panel). The captured band is less intense and upshifted due to the presence of the 750-bp long L-loop. The plasmid map shows the location of telomere sequence and the two restriction sites used (SpeI and ScaI). (c) rDNA-containing plasmid and empty plasmid were cut in the backbone region with SacI (S) to yield the target site for probes 1 and 2 inside the construct, or were digested with EcoRI (E) to yield the target site on the end of the fragment. (I, input; F, Flow-through; E, Eluate). Right panel: Histogram quantifying target capture (error bars represent standard deviations. Asterisks denote significance, *P<0.001 from a student-t test). Plasmid maps not on scale.

Figure 2. rDNA promoter chromatin purification.

(a) The map of probe hybridization sites on the rDNA promoter region. An arrow: 47S pre-rRNA transcription initiation site. Scale bars (in bp) are shown below; 0 indicates the 5′ end of the pre-rRNA. Red bars: TTF-1-binding sites. Purple bars: repetitive enhancer elements are indicated. Black thick lines (T0 and 5′ETS) probes used in the Southern blots of b and Supplementary Fig. S2b. Grey lines: DpnII fragments from the promoter region. The probes that are designed to bind around DpnII sites are represented in red for ePICh with 6 probes (RDN6) and in blue for ePICh with 23 probes (RDN23). (b) DNA from ePICh material analysed by agarose gel electrophoresis. DNA was detected by EtBr staining (left panel) and quantified by Southern blotting with probes specific for the 5′ETS region, the TTF-1-binding site (see a) and telomere. Input represents 10 and 5% of the lanes 1 and 2 of the starting material. ePICh purified DNA using scrambled probe (lane 4), RDN6 probe (lane 5) and RDN23 probe (lane 6) were analysed. Bottom bar graph, absolute quantification of rDNA amount in ePICh material. The signal of the materials from ePICh with each probe were compared with a standard plasmid DNA containing the DpnII fragment from 5′ETS and the DpnII fragment from T0 shown in a. The numbers above the graph bars represent the relative enrichment of the analysed rDNA fragments compared with major satellites, as measured by real-time PCR. (c) Silver staining of proteins obtained from ePICh with scrambled, RDN6 and RDN23 probe sets in mouse erythrocytes leukaemia (MEL) cells. 10% of the ePICh material was analysed. Red arrowheads show specific bands detected using RDN6 and RDN23 probes. (d) Western blot analysis monitoring ePICh enrichment. The second biggest subunit of RNA polymerase I (RPA116) and UBF in materials from ePICh with each probe set were assayed. As a negative control for background enrichment, the RNA Polymerase II RPB1 subunit was probed. Input represents 0.0004% of the chromatin extracts. 5% of the ePICh material were loaded per lane.

Figure 3. rDNA gene promoter chromatin proteome.

(a) Categorization of the 116 proteins enriched in ePICh with both RDN6 and RDN23 into nine functional groups: specific subunits of the RNA polymerase I core complex (RNA pol I), Pre-initiation complex for pol I transcription (pre-initiation), ribosomal RNA processing and modification proteins (rRNA processing), ribosomal proteins (RP), proteins localized in the nucleolus (Nucleolar protein), chromatin regulatory protein (Chromatin), proteins involved in DNA replication and repair (Replication and Repair), unclassified proteins (Unclassified), streptavidin and mitochondria proteins (Contaminants). (b) List of proteins enriched at the rDNA promoter according to the number of identified peptides and arranged in descending order. Scale bars (in peptide number) are shown below.

Figure 4. Validation of ZFP106 as a true and important rDNA promoter-binding protein.

(a) Chromatin immunoprecipitation of ZFP106 using anti-ZFP106 antibody and anti-UBF antibody. Immunoprecipitated (IP) DNA relative to input were quantified by real-time quantitative PCR. Top: Map of regulatory sequence elements on rDNA unit. The thick black line shows the position of PCR amplicons. (b) ChIP-chop experiment to measure DNA methylation level 275 bp downstream of the rDNA promoter in IP DNA with anti-UBF, anti-ZFP106 and input DNA. See the details in Supplementary Fig. 5. (c) Western blotting analysis for ZFP106 and fibrillarin after knockdown by three independent shRNA constructs. Whole protein extract after knockdown with shControl (lane 1: 100%, lane 2: 50%, lane 3: 20% of the extract) and the protein extract in the cells expressing three shRNA for ZFP106 were analysed (lanes 4–6). (d) Northern blotting analysis of pre-rRNA levels. Pre-RNA, 45–47S rRNA and 34S rRNA were detected by the ETS3 oligoprobe and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) messenger RNA (mRNA) was detected by a specific oligoprobe. The bars represent pre-rRNA levels (45–47S and 34S) normalized to GAPDH mRNA from three experiments. (e) ChIP analysis of the second biggest subunit of RNA polymerase I (RPA116) and UBF on rDNA in the cells expressing shControl and two shRNA for ZFP106. Immuno precipitated DNA relative to input were quantified by real-time quantitative PCR for the rDNA promoter region (Pro) and then normalized to control reactions from shControl-transfected cells. Three independent experiments were performed and error bars represent standard deviation. Stars denote significances, *P<0.01, and **P>0.2 from a Student’s t-test.