Forced expression of MSR repeat transcripts above a threshold limit breaks heterochromatin organisation

Abstract

Mouse heterochromatin is characterised by transcriptionally competent major satellite repeat (MSR) sequences and it has been proposed that MSR RNA contributes to the integrity of heterochromatin. We establish an inducible dCas9-effector system in mouse embryonic fibroblasts, where we can modulate MSR transcription through the targeting of a dCas9-Repressor or a dCas9-Activator. With this system, we can define a threshold limit of >300-fold deregulation of MSR transcript levels, above which the structural organisation of heterochromatin becomes disrupted. MEF cells expressing MSR RNA above this threshold limit are not viable and the defects in heterochromatin organisation and chromosome segregation cannot be reverted. This study highlights the importance of restricting MSR RNA output to maintain heterochromatin integrity and relates MSR transcript levels to either physiological or pathological conditions. It also reveals that the structural organisation of heterochromatin is governed by the transcriptional chromatin state and associated MSR RNA of the MSR repeats.

Fig. 1. Targeting of inducible MSR-dCas9-effector components to modulate RNA output from heterochromatin.

a Diagram of inducible MSR-dCas9-effector components: MSR-dCas9-Control, MSR-dCas9-Activator (VPR) and MSR-dCas9-Repressor (KM). Also shown is the DNA sequence of a single unit (234 bp) of the MSR consensus, with the sequence of the dCas9 guide RNA highlighted in bold red. b Western blot analysis for the detection of MSR-dCas9-Control, MSR-dCas9-Repressor and MSR-dCas9-Activator after doxycycline (dox) induction. GAPDH expression is shown as a loading control. Induction conditions that result in comparable expression of MSR-dCas9-Control, MSR-dCas9-Repressor and MSR-dCas9-Activator used in this study are indicated with red asterisks. c Immunofluorescence for the localisation of MSR-dCas9-Control, MSR-dCas9-Repressor and MSR-dCas9-Activator. MEF cells were counterstained with DAPI. Scale bar is 5 μm. Images were acquired from three independent experiments. d RT-qPCR analysis for major satellite repeat (MSR), minor satellite repeat and LINE L1Md_A repeat transcripts in inducible MSR-dCas9-Control, MSR-dCas9-Repressor and MSR-dCas9-Activator MEF cells. Expression is normalised to Hprt and is relative to MSR-dCas9-Control (-dox) (mean±SD). n = 3 independent experiments. Asterisks indicate statistically significant differences (*, p ≤ 0.05, **, p ≤ 0.001, ****, p ≤ 0.0001, ns, not significant, two-sided multiple t tests). e MA-plots depicting differential repeat element expression in inducible MSR-dCas9-Control, MSR-dCas9-Repressor and MSR-dCas9-Activator MEF cells, as determined by RNA sequencing (RNA-seq). Major satellite repeats (GSAT_MM), minor satellite repeats (SYNREP_MM) and LINE-1 (L1Md_A) repeats are marked on the plots. Up-regulated repeats are represented by orange dots and down-regulated repeats by blue dots. Grey dots represent not statistically significant changes. n = 3 independent experiments.

Fig. 2. Perturbed heterochromatin organisation in MEF cells expressing MSR-dCas9-effector components.

a Immuno-DNA FISH analysis for MSR sequences of inducible MSR-dCas9-Control, MSR-dCas9-Repressor and MSR-dCas9-Activator MEF cells. In addition to the DNA-FISH signal, 3D masks were created to quantify changes in sphericity and volume of MSR-positive domains (right panels). MSR-positive domains analysed: MSR-dCas9-Control n = 509 (–dox), n = 469 (+dox); MSR-dCas9-Repressor n = 404 (–dox), n = 371 ( + dox); MSR-dCas9-Activator n = 344 (–dox), n = 254 ( + dox). Box plots shown are centred around the median value with an interquartile range (IQR) defined by the 1st and 3rd quartiles and whiskers extending to 1.5xIQR. Outliers are plotted as individual points. Asterisks indicate statistically significant differences (***, p ≤ 0.001, ****, p ≤ 0.0001, two-sided unpaired Mann–Whitney test). Scale bar is 5 μm. b Double immunofluorescence in inducible MSR-dCas9-Control, MSR-dCas9-Repressor and MSR-dCas9-Activator MEF cells for localisation of dCas9-effector components and HP1α. Nuclei were counterstained with DAPI. Linescans were used to visualise the colocalisation of DAPI (blue), Cas9 (orange), and HP1α (grey) signals as shown in the panels on the right. For each sample n ≥ 30 cells were analysed. Scale bar is 5 μm. c Double immunofluorescence in inducible MSR-dCas9-Control, MSR-dCas9-Repressor and MSR-dCas9-Activator MEF cells for localisation of MSR-dCas9-effector components and initiating RNAPII (Ser5phos) (left panel) or for MSR-dCas9-effector components and H3 panAc (right panel). Nuclei were counterstained with DAPI. For each sample n ≥ 30 cells were analysed. Scale bar is 5 μm.

Fig. 3. Upregulated MSR repeat transcripts remain associated with MSR-targeted dCas9-Activator domains and form RNA:DNA hybrids.

a Immuno-RNA FISH in inducible MSR-dCas9-Control, MSR-dCas9-Repressor and MSR-dCas9-Activator MEF cells for localisation of MSR-dCas9-effector components and forward (purine-rich) MSR repeat transcripts (left panel) or for MSR-dCas9-effector components and reverse (pyrimidine-rich) MSR repeat transcripts (right panel). Nuclei were counterstained with DAPI. In parallel, cells were incubated with RNase A ( + RNase A) prior to MSR probe hybridisation. For each sample n ≥ 60 cells were analysed. Scale bar is 5 μm. b RNA:DNA hybrid dot blot with the S9.6 antibody on nucleic acid samples from MSR-dCas9-Control, MSR-dCas9-Repressor and MSR-dCas9-Activator MEF cells uninduced or induced with doxycycline (dox). Samples were also treated with or without RNase H to determine RNase H sensitivity of the immunoprecipitated nucleic acids. The membrane was then stained with SYBR Gold to quantify total nucleic acids. As a control for RNase H activity, a cDNA sample was included. The histogram on the right is the quantification of the S9.6 signal from the dot blot normalised to the SYBR Gold signal (mean ± SD). n = 3 independent experiments. Asterisks indicate statistically significant differences (**, p ≤ 0.0019, two-way ANOVA, Šídák’s test). c RNA:DNA hybrid immunoprecipitation (RDIP) with the S9.6 antibody on genomic DNA from MSR-dCas9-Control, MSR-dCas9-Repressor and MSR-dCas9-Activator MEF cells uninduced or induced with doxycycline (dox). Samples were treated with or without RNase H to determine RNase H sensitivity of the immunoprecipitated nucleic acids. RNA: DNA hybrids were quantified by qPCR using primers against MSR, minor satellite repeats and LINE-1 (L1Md_A) repeats. For each histogram, values were normalised relative to MSR-dCas9-Control (–dox, –RNase H) (mean ± SD). n = 3 independent experiments. Asterisks indicate statistically significant differences (*, p ≤ 0.0102, ***, p ≤ 0.001, ****, p ≤ 0.0001, two-way ANOVA, Šídák’s test).

Fig. 4. Heterochromatin disruption requires a threshold limit for MSR transcript deregulation.

a Western blot analysis for the detection of MSR-dCas9-Activator after induction with increasing concentrations of doxycycline (dox). GAPDH expression is shown as a loading control. b RT-qPCR analysis for MSR transcripts from MSR-dCas9-Activator MEF cells induced with increasing concentrations of doxycycline (dox). Expression is normalised to Hprt and is relative to MSR-dCas9-Control (-dox) (mean±SD). n = 3 independent experiments. The red segment indicates the threshold range of MSR transcript deregulation (between 300–800-fold), within which heterochromatin starts to become disrupted. (*, p ≤ 0.0481, ***, p ≤ 0.0003, one-way ANOVA, Dunnett’s test). n = 3 independent experiments. c Confocal imaging of DAPI-dense regions (DAPI counterstaining) in MSR-dCas9-Activator MEF cells induced with increasing concentrations of doxycycline (dox). The percentages reflect the fraction of cells with either undispersed (white) or dispersed (yellow) DAPI-dense regions. (*, p ≤ 0.0222, ****, p ≤ 0.0001, one-way ANOVA, Dunnett’s test). Scale bar is 5 μm. For each condition n ≥ 240 cells were analysed from n = 3 independent experiments. Quantification of the imaging data is shown in the bar graph below (mean ± SD). d Bar graph of normalised counts from HiSeq RNA sequencing comparing MSR expression in Suv39h dn and H3K9 KMT 6KO MEF cells (Montavon et al. 2021) (blue), in uninduced and induced MSR-dCas9-Activator MEF cells (this study) (white and grey) and in mouse mammary tumours (Zhu et al. 2018) (orange) (mean ± SD). Fold increase of MSR RNA in the Suv39h dn and H3K9 KMT 6KO samples was calculated relative to the wt sample (indicated above the bars). Fold increase of MSR RNA in the MSR-dCas9-Activator (-dox) and MSR-dCas9-Activator (+dox) samples was calculated relative to the MSR-dCas9-Control (–dox) sample (indicated above the bars). The dashed line indicates the average normalised counts for Hprt expression across all three datasets. The red segment indicates the threshold range of MSR transcript deregulation derived from proportional conversion of RT-qPCR values shown in (b) and is equivalent to 13,000–37,000 normalised RNA-seq counts.

Fig. 5. RNA transcription and hyperacetylated chromatin drive heterochromatin disruption.

a Western blot analysis for expression of MSR-dCas9-Activator in MEF cells (see flow diagram for induction and inhibition conditions). GAPDH expression is shown as a loading control. b RT-qPCR analysis for MSR transcripts in MEF cells treated as in (a). Values were normalised to 5S RNA and are relative to MSR-dCas9-Control (–dox, –DRB) (mean ± SD) (****, p ≤ 0.0001, one-way ANOVA, Tukey’s test). n = 3 independent experiments. c Confocal imaging of DAPI-dense regions in MSR-dCas9-Activator MEF cells that were treated as in (a). Scale bar is 5 μm. The percentages reflect the fraction of cells with either undispersed (white) or dispersed (yellow) DAPI-dense regions. For each condition n ≥ 200 cells were analysed from n = 3 independent experiments. Quantification of the imaging data is shown on the right (mean ± SD). d Western blot analysis for expression of MSR-dCas9-Activator and pan-acetylated H3 in MEF cells (see flow diagram for induction and inhibition conditions). GAPDH and H3 expression are shown as loading controls. e RT-qPCR analysis for MSR transcripts in MEF cells treated as in (d). Values were normalised to Hprt and are relative to the MSR-dCas9-Control (-dox,-A-485) (mean ± SD). n = 3 independent experiments. f Immunofluorescence of MSR-dCas9-Activator MEF cells treated as in (d). Cells were immunolabelled for Cas9 and pan-acetylated H3 and counterstained with DAPI. Scale bar is 5 μm. The percentages reflect the fraction of cells with either dispersed (yellow) or undispersed (white) DAPI-dense regions. For each condition n ≥ 135 cells were analysed from three independent experiments. Quantification is shown on the right (mean ± SD). g Confocal imaging of DAPI-dense regions after combined inhibition of RNAPII and HAT enzymes in induced MSR-dCas9-Activator MEF cells (see flow diagram). Scale bar is 5 μm. The percentages reflect the fraction of cells with either dispersed (yellow) or undispersed (white) DAPI-dense regions. The total number of cells analysed per condition are as follows: (–DRB,–A-485) n = 159, (–DRB,1 µM A-485) n = 197, (–DRB,5 µM A-485) n = 172, (–DRB,10 µM A-485) n = 205, ( + DRB,–A-485) n = 157, ( + DRB,1 µM A-485) n = 186, (–DRB,5 µM A-485) n = 172, (–DRB,10 µM A-485) n = 220. n = 3 independent experiments. Quantification of the imaging data is displayed on the right (mean ± SD) (**, p ≤ 0.0044, two-way ANOVA, Šídák’s test).

Fig. 6. Delayed mitotic progression and chromosome mis-segregation in MEF cells expressing MSR-dCas9-Activator.

a Western blot analysis for detection of induced MSR-dCas9-Control, MSR-dCas9-Repressor and MSR-dCas9-Activator in unsynchronised and in RO-3306 synchronised MEF cells. Following release from RO-3306 inhibition, cell samples were collected every 30 min for an interval of up to 2.5 h. Mitotic progression was verified by probing for histone H3S10 phosphorylation (H3S10ph). GAPDH and H3 are shown as loading controls. b Profiles for mitotic progression of RO-3306 synchronised MSR-dCas9-Control, MSR-dCas9-Repressor and MSR-dCas9-Activator MEF cells without doxycycline (-dox, black circles) or with doxycycline ( + dox, red circles) induction. Following release from RO-3306 inhibition, cell samples were collected every 30 min for an interval of up to 2.5 h. Percentages of mitotic cells were quantified by immunofluorescence for H3S10 phosphorylation. For each sample n ≥ 60 cells were analysed. c Confocal imaging of chromosomes in metaphase, anaphase or telophase stages from MSR-dCas9-Control (left panel), MSR-dCas9-Repressor (middle panel) and MSR-dCas9-Activator (right panel) MEF cells without doxycycline (-dox) or with doxycycline ( + dox) induction. Chromosomes were counterstained with DAPI. For each mitotic stage, two representative images are shown. Scale bar is 10 μm. Yellow arrows indicate lagging chromosomes, chromosome bridges, multiple mitotic spindles and non-segregated chromosome sets in MEF cells expressing MSR-dCas9-Activator. Quantification of the mitotic defects is shown in Supplementary Fig. S7.

Fig. 7. Induced expression of MSR-dCas9-Activator irreversibly compromises cell viability.

a Bar graphs depicting percentages of live (white), apoptotic (blue) or dead (orange) cells in MSR-dCas9-Control, MSR-dCas9-Repressor and MSR-dCas9-Activator MEF cell populations either 0 h (left panel) or 24 h (right panel) after doxycycline removal. Cell viability was analysed for unsynchronised or RO-3306 synchronised cells and also when cells were RO-3306 synchronised and released or RO-3306 synchronised and blocked. Highlighted by light green shading (for the 24 h time point) are cell culture conditions where cell samples were released into mitosis. b Heatmap of differentially expressed genes between MSR-dCas9-Control and MSR-dCas9-Activator MEF cells. Uninduced, induced and post-induction (3, 6, 12 and 24 h after dox removal) conditions were compared. Differentially expressed genes were filtered with a basemean >100 reads, >2-fold expression change and a p-adjusted value < 0.05. k-means clustering revealed that differentially expressed genes between MSR-dCas9-Control and MSR-dCas9-Activator during the recovery phases group into three clusters. n = 3 independent experiments. c Gene ontology (GO) analysis of cluster 2. Dot plots display the top 7 GO terms. n = 3 independent experiments. GO analysis of clusters 1 and 3 are shown in Supplementary Fig. 8c. d Confocal imaging (DAPI counterstaining) of induced MSR-dCas9-Control, MSR-dCas9-Repressor and MSR-dCas9-Activator MEF cell populations 24 h after doxycycline (dox) removal. Scale bar is 10 μm. Indicated with arrows are micronuclei and multi-nucleated cells (yellow), apoptotic bodies (blue) and enlarged nuclei (green) from cells induced to express MSR-dCas9-Activator.