Senescence is an endogenous trigger for microRNA-directed transcriptional gene silencing in human cells

Abstract

Cellular senescence is a tumour-suppressor mechanism that is triggered by cancer-initiating or promoting events in mammalian cells. The molecular underpinnings for this stable arrest involve transcriptional repression of proliferation-promoting genes regulated by the retinoblastoma (RB1)/E2F repressor complex. Here, we demonstrate that AGO2, RB1 and microRNAs (miRNAs), as exemplified here by let-7, physically and functionally interact to repress RB1/E2F-target genes in senescence, a process that we call senescence-associated transcriptional gene silencing (SA-TGS). Herein, AGO2 acts as the effector protein for let-7-directed implementation of silent-state chromatin modifications at target promoters, and inhibition of the let-7/AGO2 effector complex perturbs the timely execution of senescence. Thus, we identify cellular senescence as the an endogenous signal of miRNA/AGO2-mediated TGS in human cells. Our results suggest that miRNA/AGO2-mediated SA-TGS may contribute to tumour suppression by stably repressing proliferation-promoting genes in premalignant cancer cells.

Figure 1. Identification of AGO-bound E2F target genes and heterochromatin-bound miRs

(a-c),AGO ChIP-on-chip promoter profiling. Chromatin immunoprecipitation (ChIP) from RAS^V12-induced senescent- (S) and pre-senescent, empty vector control (C) cells was performed using pan-anti-AGO antibody. Purified DNA fragments from matched genomic inputs and ChIP samples were amplified, fluorochrome-labeled and hybridised to Nimblegen promoter arrays. a, Shown is the total number of AGO2-bound promoters in control and RAS^V12-induced senescent cells, and a Venn diagram depicting specific and common AGO2-promoter targets. b, Venn diagram depicting number of AGO-bound (AGO) E2F target promoters (E2F) in control (C, upper panel) and senescent (S, lower panel) cells; P < 0,001 performing Chi-square test. c, Correlation between AGO-bound E2F-promoters and respective E2F target gene expression level; Up, AGO-bound E2F target genes up-regulated in senescence; Down, AGO-bound E2F target genes down-regulated in senescence. (d-f), Degree of association between miR sequence counts observed in d, unfractionated, cellular RNA and AGO- or e, H3K9me₂-RIP and f, between AGO- and H3K9me₂-RIPs. Normalised logarithmic counts per million (log₁₀ cpm) cpm for each miR from the respective sample were scatter plotted (see also Supplementary Table S4); (r²), correlation coefficient of determination, r, Pearson correlation coefficient.

Figure 2. AGO2 accumulates in the nucleus of senescent cells and is recruited to promoters of repressed E2F-target genes

a, Western blot analysis of cytosolic and nuclear fractions from Ras^V12-induced-senescent (S) and pre-senescent, empty vector control (C) WI38 fibroblasts using anti-AGO2, -Lamin B and -GAPDH-specific antibodies. Lamin B and GAPDH detection serve as positive controls for nuclear and cytosolic fractions; WCL, whole-cell lysate. b, Localisation of AGO2 in detergent pre-extracted pre-senescent, control (C) and senescent- (S) cells as determined by indirect immunofluorescence microscopy. DAPI was used to counterstain nuclear DNA; scale bar, 10μm. Colocalisation of AGO2 and macroH2A c, in nuclei of melanocytic nevus and d, in cytosol of melanoma as determined by indirect immunofluorescence; scale bar, 20μm. e, Global AGO2 enrichment in heterochromatin of senescent cells as detected by histone-association assay. Pre-senescent, control (C) and senescent- (S) cells were fixed by PFA, sonicated and lysed. Euchromatin- and heterochromatin-bound AGO2 was identified by co-immunoprecipitation (Co-IP) using anti-histone H3K9_ac (euchromatin), -H3K27_me3, H3K9_me2 (heterochromatin) and IgG control antibodies. Immunoprecipitates were Western-blotted with anti-AGO2 and anti-histone H3 antibodies. Ratios were calculated densitometrically. f, AGO2 and histone modification profiling of E2F-responsive promoters. qChIP was performed in pre-senescent, empty vector control (C) and Ras^V12-induced senescent- (S) cells using anti-AGO2, -H3K27_me3, -H3K9_me2, -H3K4_me3 and IgG control antibodies followed by qPCR of CDC2, CDCA8, PCNA and CCNA2 promoters. Data are means ± s.d.; n=3; P < 0,05. Experiments were performed in quadruplicates.

Figure 3. AGO2 cooperates with Rb to regulate E2F target gene expession

(a-b) AGO2 and Rb or HDAC1 were co-immunoprecipitated in cell lysates prepared from Ras^V12-induced a, senescent WI38 fibroblasts or b, Rb-inducible (TET-Off) senescent SAOS-2 (Rb-negative) cells. IgG, HDAC1 and AGO2-co-immunoprecipitates (Co-IP) were Western-blotted with anti-AGO2, -HDAC1 and -Rb antibodies. Cellular lysates from Rb-inducible (TET-Off) SAOS-2 (Rb-negative) cells were prepared and immunoprecipitated with anti-AGO2 and -Rb antibodies followed by Western blotting using a mix of anti-AGO2 and -Rb antibodies. WCL, whole cell lysate. -Rb, no Rb expression TET+; +Rb, Rb expression TET-. Asterisk indicates IgG heavy chain, arrow indicates specific HDAC1 band. (c-d), AGO2 and Rb co-repress cyclin-E promoter. c, Wild-type cyclin-E promoter-luciferase reporter assay in C33A cells. All indicated plasmid amounts in ng. d, Same assay using a cyclin-E promoter with mutated E2F-recognition site; E2F1-wt, promoter reporter with E2F1-binding sites; E2F1-mt, promoter reporter with mutated E2F1-binding sites. Data are means ± s.d.; n=5; P < 0,05. Experiments were performed in triplicates. Expression of proteins were confirmed by Western blot (data not shown). RLU, relative light units. (e-f), Depletion of AGO2 derepresses and releases Rb from E2F target genes. Cells were retrovirally infected with pBABE-Ras^V12 and, 2 days post-drug selection, cells were transiently transfected with siAGO1, siAGO2 or siScramble control (siC)(see also Supplementary Fig. S4c). e, qRT-PCR on total RNA prepared from respective samples at day 3 of siRNA treatment. Also shown is a Western blot for AGO2 levels in si-treated cells; Rb, loading control. Ratio calculated densitometrically. Data are means ± s.d.; n=3; P < 0,05. Experiments were performed in quadruplicates. f, H3K27_me3- and Rb-qChIP on CDCA8, CCNE2, CDC2 and CCNA2 promoters was performed at day 3 of siRNA treatment. Data are means ± s.d.; n=3; P < 0,05. Experiments were performed in quadruplicates. (g-i), Depletion of Rb does not have an impact on AGO2 presence at E2F target genes. Cells were retrovirally infected with pBABE-Ras^V12 and 2 days post-drug selection cells were transiently transfected with siScramble control (siC) or siRb for 3 days. g, AGO2-qChIP for E2F-target genes. Data are means ± s.d.; n=3; P < 0,05. Experiments were performed in quadruplicates. h, Indirect immunofluorescence of Rb in siC- or siRb-treated cells; scale bar, 20μm. i, qRT-PCR on total RNA prepared from si-treated samples. Data are means ± s.d.; n=3; P < 0,05. Experiments were performed in quadruplicates.

Figure 4. Depletion of AGO2 delays senescence arrest in W38 fibroblasts

a, Proliferation curves of WI38 fibroblasts under physiological oxygen concentration of 3% superinfected either with shControl (shC):Ras^V12, shAGO2-1/2:Ras^V12 or HPV16E7:Ras^V12. (b-d), Analysis of shC:Ras^V12 and shAGO2s:Ras^V12 cell populations for different proliferation and senescence markers at day 14 of life span study. b, Representative photomicrographs of cell density and morphology; scale bar 10μm. Bart chart for c, fraction of EdU incorporating cells and d, percentage of SAHF positive cells. Also shown is SAHF formation as determined by fluorescence microscopy using DAPI counterstaining of DNA in shC:Ras^V12 and shAGO2:Ras^V12 cells; scale bar 20μM. e, Proliferation curve of WI38 fibroblasts at physiological oxygen concentration of 3% either stably infected with shRNA control (shC), two independent shAGO2 constructs (shAGO2-1 and -2) or viral oncogene HPV16E7 (E7). Growth curves were initiated when shC vector control cell population ceased proliferation. (f-g), Analysis of shC control and shAGO2-silenced cell populations for different proliferation and senescence markers at day 21 of life span study. f, Bar chart for fraction of EdU incorporating cells. g, Representative micrographs showing SA-β-Gal staining and graph for percentage of SA-β-Gal positive cells; scale bar 15μM. Data are means ± s.d.; n=3; P < 0,05. Experiments were performed in duplicates.

Figure 5. Overexpression of AGO2 induces proliferative arrest with features of premature senescence

a, Life span study of FACS-sorted GFP-positive WI38 fibroblasts transduced with vector control (C) or MigR-AGO2. b-e, Analysis of cell populations overexpressing empty vector control or AGO2 for different proliferation and senescence markers at day 14 of life span study. Bar chart for b, fraction of EdU incorporating cells and c, percentage of SAHF positive cells. SAHF formation d, in detergent pre-extracted control and AGO2-overexpressing cells by indirect fluorescence using DAPI DNA counterstain to visualise SAHF and e, by indirect immunostaining of H3K9_me3 immunostaining to visualise SAHF in detergent pre-extracted GFP-positive AGO2-overexpressing cells; scale bars, 20μm. Data are means ± s.d.; n=3; P < 0,05. Experiments were performed in duplicates.

Figure 6. AGO2 and let-7f cooperate to induce transcriptional gene silencing (TGS) of E2F target promoters

a, Interaction of AGO2 and H3K9_me2 (H3K9) with let-7f was validated by combined RIP and small RNA cDNA cloning using a miR-let-7f specific primer. This result is representative of three independent experiments. IgG, non-specific antibody; I, Input. b, Alignment of let-7f to CDC2 and CDCA8 target promoters. Identical nucleotides are shown in red. Asterisk, transcription start site (TSS); E2F-binding site is depicted as grey box with respect to TSS; Arrows to the right and left of miR-target are primers used for qChIP or qRT-PCR; numbers on top of indicate miR target site position with respect to TSS. (c-f), let-7f and AGO2 cooperate to implement epigenetic gene silencing at E2F target genes. c, Exogenous let-7f or scrambled miR (C) were transiently transfected at 100nM into pre-senescent WI38 fibroblasts. Transcription of CDC2, CDCA8 and Cyclophylin (CYP) mRNAs was measured by nuclear-run-on (left panel) or qRT-PCR (right panel), respectively. CYP served as normalization control. Shown is a representative experiment of three experimental repeats. d, qChIP was performed in let-7f-treated cells using anti-AGO2, - H3K27_me3 and IgG control antibodies followed by qPCR using primers amplifying CDCA8 and CDC2 promoter regions. Data are means ± s.d.; n=4; P < 0,05. Experiments were performed in quadruplicates. e, Simultaneous presence of let-7f and AGO2 on CDCA8 and CDC2 promoters was determined by combined qChIP-ChOP assay. Pre-senescent WI38 fibroblasts were transfected with 100nM biotin-labeled let-7f, control miR (siC) or let-7 followed by successive rounds of control-IgG-ChIP or AGO2-ChIP and streptavidin-affinity (SA) precipitation. SA-chromatin-precipitates were analyzed by qPCR with primers detecting E2F-responsive promoters CDCA8 and CDC2 Data are means ± s.d.; n=4; P < 0,001. Experiments were performed in duplicates. f, Presence of let-7 at target promoters is AGO2-dependent. Pre-senescent WI38 were co-transfected with siAGO2 or siC (100nM) and biotin-labeled let-7f (100nM) followed by SA precipitation; Data are means ± s.d.; n=3; P < 0,05. Experiments were performed in duplicates. (g-h), Exogenous let-7f induces cellular senescence. Pre-senescent WI38 fibroblasts were transiently transfected with 100nM let-7f or control miR (C). g, Cell numbers were determined at the indicated time points. h, Number of cells staining positive for Ki67 and senescence-associated beta-galactosidase (SA-β-Gal) at day 8 post-treatment. Data are means ± s.d.; n=3; P < 0,05. Experiments were performed in duplicates.

Figure 7. Inhibition of let-7f perturbs timely execution of senescence and SA-TGS

(a-b), WI38 fibroblasts undergoing Ras^V12-induced senescence were treated with 100nM siAGO1, siAGO2, siScramble (siC) or let-7f antogomirs (Anti-miR). Cell proliferative capacity of cells was measured by a, growth curves and b, by immunostaining using anti-Ki67 antibody. Data are means ± s.d.; n=3; P < 0,04. Experiments were performed in duplicates. c, AGO2-qChIP was performed on Ras^V12-senescent cells treated with 100nM control- (C) or let-7f antagomirs (anti-miR) on CDCA8 and CDC2 promoters. Data are means ± s.d.; n=3; P < 0,01. Experiments were performed in duplicates. d, qRT-PCR was performed on total RNA prepared from WI38 fibroblasts undergoing Ras^V12-induced senescence treated with 100nM siAGO1, control- (C), siAGO2 or let-7f antogomirs (anti-miR) to measure CDCA8 and CDC2 mRNA levels. Asterisk, statistical significance P. Data are means ± s.d.; n=3; P < 0,05. Experiments were performed in quadruplicates.

Figure 8. Model for AGO2 and miR function in SA-TGS of E2F-target genes

In cells undergoing senescence, cytosolic (cy) miR/AGO2 complexes are retained in the nucleus. We favor a model in which the complex is guided by the miR (red) to the respective target promoter(s) where it associates with promoter-associated RNA(s) (pRNA)(left panel), which has/have been shown to be critical for TGS . Indeed, our preliminary data indicate that RNA-polymerase II (Pol II)-dependent TSS-proximal pRNAs are produced at E2F-targets in proliferating cells, whereas in senescent cells these transcripts become almost extinct. Thus, it remains to be seen whether or not AGO2 slices pRNAs after association with the respective promoters. Alternatively, the complex may directly target complementary target promoter sequences (right panel). In both cases, the miR/AGO2 complex ultimately blocks productive RNA Pol-II engagement and cooperates with E2F/Rb to repress E2F target promoters through recruitment of (additional) corepressor activities including histone-methyl transferases (HMTs) and -deacetylases (HDACs) and/or by stabilizing a pre-existing E2F/Rb-associated repressor complex. Targeted promoters are ultimately stably repressed by an inactive chromatin state characterized by methylated lysines 9 and 27 on histone H3 as cells undergo senescence.