Ribosomal DNA promoter recognition is determined in vivo by cooperation between UBTF1 and SL1 and is compromised in the UBTF-E210K neuroregression syndrome

Abstract

Transcription of the ~200 mouse and human ribosomal RNA genes (rDNA) by RNA Polymerase I (RPI/PolR1) accounts for 80% of total cellular RNA, around 35% of all nuclear RNA synthesis, and determines the cytoplasmic ribosome complement. It is therefore a major factor controlling cell growth and its misfunction has been implicated in hypertrophic and developmental disorders. Activation of each rDNA repeat requires nucleosome replacement by the architectural multi-HMGbox factor UBTF to create a 15.7 kbp nucleosome free region (NFR). Formation of this NFR is also essential for recruitment of the TBP-TAFI factor SL1 and for preinitiation complex (PIC) formation at the gene and enhancer-associated promoters of the rDNA. However, these promoters show little sequence commonality and neither UBTF nor SL1 display significant DNA sequence binding specificity, making what drives PIC formation a mystery. Here we show that cooperation between SL1 and the longer UBTF1 splice variant generates the specificity required for rDNA promoter recognition in cell. We find that conditional deletion of the TAF1B subunit of SL1 causes a striking depletion of UBTF at both rDNA promoters but not elsewhere across the rDNA. We also find that while both UBTF1 and -2 variants bind throughout the rDNA NFR, only UBTF1 is present with SL1 at the promoters. The data strongly suggest an induced-fit model of RPI promoter recognition in which UBTF1 plays an architectural role. Interestingly, a recurrent UBTF-E210K mutation and the cause of a pediatric neurodegeneration syndrome provides indirect support for this model. E210K knock-in cells show enhanced levels of the UBTF1 splice variant and a concomitant increase in active rDNA copies. In contrast, they also display reduced rDNA transcription and promoter recruitment of SL1. We suggest the underlying cause of the UBTF-E210K syndrome is therefore a reduction in cooperative UBTF1-SL1 promoter recruitment that may be partially compensated by enhanced rDNA activation.

Fig 1. Genetic inactivation of the Taf1b gene disrupts rRNA synthesis and nucleolar structure in conditional MEFs.

A) Schematic of the basal factors of the RNA Polymerase I (RPI/Pol1) initiation complex and their assembly on the rDNA promoter. The diagram is not intended to indicate precise factor positioning. B) General organisation of the rDNA chromatin structure in mouse. The positions of Spacer and 47S Promoter duplications, the distribution of UBTF, CTCF binding at the upstream boundary and the flanking nucleosomal IGS are indicated. C) Synthesis of 47S pre-rRNA in Taf1b conditional MEFs. 47S pre-rRNA synthesis was determined by [3H]-uridine RNA metabolic labelling following homozygous Taf1b deletion by 4-HT induction of ER-Cre. The inset in the upper panel shows a typical example of 47S pre-rRNA labelling in Taf1b^fl/fl/p53^-/-/ERcre^+/+ MEFs at different times after ER-Cre induction. The histogram shows quantification of 47S labelling in Taf1b^fl/fl/p53^-/-/ERcre^+/+ MEFs normalized to labelling in control Taf1b^wt/wt/p53^-/-/ERcre^+/+ MEFs. The data were derived from 6 independent analyses of MEFs derived from two floxed and two wild type embryos. Error bars indicate the SEM. The lower panel in C shows a parallel time course of TAF1B depletion, see S2 Fig for more detail. D) RPI, UBTF and fibrillarin (FBL) indirect immunofluorescence labelling in Taf1b^fl/fl/p53^-/-/ERcre^+/+ MEFs at different time points after 4-HT induction of Taf1b deletion, see S3 Fig for a more detail.

Fig 2. Loss of TAF1B abrogates RPI but not UBTF recruitment to the rDNA.

A) Organisation of the mouse rDNA locus indicating the positions of the Spacer (SpPr) and 47S (47SPr) promoter sequences, the RPI termination sites Tsp, T₀ and T1-10, the extent 47S pre-rRNA coding region (light green) and the encoded 18, 5.8 and 28S rRNAs. The positions of the qPCR amplicons used in ChIP analyses are indicated, as are the rDNA fragments and pMr100 probe used in psoralen accessibility cross linking (PAC). B) ChIP-qPCR analysis of TAF1B, TBP, RPI and UBTF occupancy at sites across the rDNA of Taf1b^fl/fl/p53^-/-/ERcre^+/+ MEFs before and 5 days after Taf1b inactivation by 4-HT treatment. The data derive from 3 biological ChIP replicas each analyzed by qPCR in triplicate. S5 Fig shows a similar ChIP analysis in Taf1b^fl/fl/ERcre^+/+ mESCs with mapping of TAF1B, TAF1C and TBP subunits of SL1. C) PAC reveals a gradual reduction in the amount and mobility of active rDNA chromatin following Taf1b inactivation as in B. Upper panel shows a typical psoralen time course analysis of rDNA chromatin showing the lower mobility of the active “a” and higher mobility of the inactive “i” 1.3kbp BamHI-BamHI fragment from the rDNA 47S coding region. The lower histogram panel shows the mean active rDNA fraction estimated from curve fit analysis of the 1.3, 2.4 and 4.7kbp BamHI-BamHI rDNA fragment profiles in two biological replicas. Error bars indicate the SEM.

Fig 3. TAF1B loss induces depletion of UBTF from both Spacer and 47S promoters but not from the adjacent enhancer repeats nor from the 47S gene body.

A) DChIP-Seq analysis of TAF1B and UBTF occupancy across the rDNA repeat in Taf1b^fl/fl/p53^-/-/ERcre^+/+ MEFs before (TAF1B+) and 5 days after Taf1b inactivation (TAF1B-). ΔUBTF indicates the difference map of UBTF occupancy after TAF1B depletion minus the occupancy before TAF1B depletion. B) Magnified view of the DChIP mapping in A showing detail over the promoter and enhancer regions. C) Analysis of Spacer and 47S promoter occupancies reveals a direct proportionality between TAF1B and UBTF. Four independent DChIP UBTF and TAF1B data sets displaying different levels of TAF1B depletion were quantitatively analyzed for UBTF and TAF1B promoter occupancy by peak fit, examples of which are shown in S8 Fig. The fractions of SL1 (TAF1B) and UBTF on each promoter after TAF1B depletion are plotted one against the other and reveal near linear relationships. Error bars in C show the SEM associated with peak fitting.

Fig 4. The rDNA promoters selectively recruit the UBTF1 variant.

A) Schematic representation of the domain structure of the UBTF splice variants in mouse and human indicating the N-terminal dimerization, 6 HMGbox domains and the C-terminal Acidic Domain. B) DChIP-Seq mapping profiles of exogenously expressed 3xFLAG-UBTF1 or UBTF2, and endogenous TAF1B in NIH3T3 MEFs, see also S9 Fig. The difference map of UBTF1-UBTF2 occupancies reveals a strong selectivity for UBTF1 mapping precisely over the Spacer and 47S promoters. C) Magnified view of the DChIP mapping in B showing detail over the promoter and enhancer regions. D) Peak fit analysis of UBTF1 and UBTF2 occupancies over the Spacer and 47S promoter revealed that UBTF1 was at least 4 times more prevalent at either promoter. The data derive from two biological replicas and the SEM is shown.

Fig 5. UBTF1 but not UBTF2 is recruited to the rDNA promoters in the absence of endogenous UBTF.

A) DChIP-Seq mapping profiles of exogenous 3xFlag-tagged UBTF1 or UBTF2 in UBTF conditional (Ubtf ^fl/fl/ER-Cre^+/+/p53^-/-) MEFs before (blue) and after (red) UBTF deletion. As reference, the upper track shows mapping of the endogenous UBTF shown in Fig 3. B) Quantitation of UBTF1 and 2 at the Spacer and 47S promoters as determined by peak fit analysis, S12 Fig, see Materials and Methods.

Fig 6. The E210K UBTF mutation suppresses both proliferation and 47S pre-rRNA synthesis but enhances rDNA activation.

A) Ubtf^E210K/E210K MEFs were found to proliferate significantly more slowly than isogenic wild type (ubtf^wt/wt) MEFs. Doubling time (dt) was estimated from an exponential curve fit as respectively 35h and 31h for mutant and wild type MEFs. B) 47S pre-rRNA synthesis in ubtf^E210K/E210K and wild type MEFs determined by metabolic pulse (30 min) labelling. See also S14A Fig for analysis of processing intermediates at increasing labelling times for individual MEF isolates. C) Per cell total cellular RNA content of ubtf^E210K/E210K and wildtype MEFs. The data in A, B and C derive from two or more biological replicas in each of which a minimum of two independently isolated mutant and wild type MEF cultures were analyzed in parallel. D) PAC analysis of ubtf^E210K/E210K and wild type MEFs. Upper panel shows an example of the active rDNA “a” and inactive “i” profiles for the 1.3kbp BamHI-BamHI 47S coding region fragment (see Fig 2A) and the lower panel the corresponding band intensities. E) Active rDNA fractions were estimated from the combined curve fit analysis of 1.3, 2.4 and 4.7kbp BamHI-BamHI rDNA fragment PAC profiles. The data derive from three independent Ubtf^E210K/E210K and two wild type MEF isolates in two PAC biological replicas and are plotted to show median, upper and lower data quartiles and outliers. F) and G) Analysis of UBTF1 and 2 levels in Ubtf^E210K/E210K and wild type MEF isolates. Panel F shows a typical Western analysis of UBTF variants in these MEFs and panel G quantitative estimates of relative UBTF1/UBTF2 protein and mRNA ratios in these MEFs. H) and I) Show similar estimates of relative UBTF1/UBTF2 protein and mRNA ratios in Cortex and Cerebellum tissue from matched Ubtf^wt/wt, Ubtf^wt/E210K and Ubtf^E210K/E210K adult mice. Error bars throughout indicate SEM.

Fig 7. The E210K mutation in UBTF suppresses both the RPI loading across the rDNA and preinitiation complex formation at the Spacer and 47S promoters.

A) ChIP-qPCR analysis of RPI occupancy at the Spacer promoter within the ETS region of the 47S coding region in Ubtf^E210K/E210K and wild type MEFs. B) ChIP-qPCR analysis of relative preinitiation complex formation in Ubtf^E210K/E210K and wild type MEFs. The data show the mean occupancy at amplicons SpPr and T0/Pr by TAF1B or UBTF. The data in A and B derive from 4 independent ChIP-qPCR experiments and error bars indicate the SEM, see Fig 2A and Materials and Methods for amplicon positions. C), D) and E) DChIP-Seq mapping of TAF1B, UBTF and RPI across the rDNA of Ubtf^E210K/E210K and wild type MEFs. Panels D and E show an enlargement of the rDNA promoter and Enhancer region and a difference map of UBTF occupancy (mutant—wild type MEFs). A full-width UBTF difference map is shown in S16 Fig. The data are typical of two biological replicas.

Fig 8. Diagrammatic representation of the induce-fit model for cooperative UBTF1-SL1 recognition and binding at the rDNA promoters.

A) Neither UBTF1 nor SL1 alone is able to form a stable interaction with the promoters. B) UBTF1 binding induces a transient reshaping of the promoter that allows SL1 to form weak DNA interactions. C) Tightening of UBTF contacts induces further promoter reshaping inducing a new DNA surface that closely “fits” the DNA interaction surface of SL1. The promoter and flanking DNA sequences are shown respectively in orange and dark grey, the region of UBTF known to bend DNA is shown in blue, SL1 in light grey, and the active transcription initiation site is indicated by an arrow.