ATRX is a regulator of therapy induced senescence in human cells

Senescence is a state of stable cell cycle exit with important implications for development and disease. Here, we demonstrate that the chromatin remodeling enzyme ATRX is required for therapy-induced senescence. ATRX accumulates in nuclear foci and is required for therapy-induced senescence in multiple types of transformed cells exposed to either DNA damaging agents or CDK4 inhibitors. Mobilization into foci depends on the ability of ATRX to interact with H3K9me3 histone and HP1. Foci form soon after cells exit the cell cycle, before other hallmarks of senescence appear. Eliminating ATRX in senescent cells destabilizes the senescence-associated heterochromatic foci. Additionally, ATRX binds to and suppresses expression from the HRAS locus; repression of HRAS is sufficient to promote the transition of quiescent cells into senescence and preventing repression blocks progression into senescence. Thus ATRX is a critical regulator of therapy-induced senescence and acts in multiple ways to drive cells into this state.


Supplementary Figure 2. ATRX is required for senescence induced by MDM2 knockdown.
The LS8817 scr and LS8817 shATRX cells (left) were described in Figure 1. Similarly generated LS8313 scr and LS8313 shATRX cells (right) were prepared. (A-B) These cells were transduced with an shRNA vector targeting MDM2 and the effect on protein expression (A) and accumulation of SA-β-gal (B) were determined. ATRX was detected with the D-5 antibody. All data are represented as mean ± SEM from at least 3 independent experiments. * indicates p < 0.05 using a two-sided Students t-test. Figure 3. ATRX facilitates senescence in response to CDK4 inhibition in the ATRX-deficient cell line U2OS. U2OS cells were stably transfected with wild type ATRX as described in the legend to Figure 3. (A) The expression of cyclin A and phosphorylated Rb before and after drug treatment in parental U2OS cells and U2OS cells expressing ATRX was determined by immunoblot. (B-C) The accumulation of SA-β-gal (B) and SAHF (C) were measured in each condition. Representative SAHF images are shown in panel C. (D) Clonogenic growth was measured by crystal violet staining three weeks after the removal of the drug and replating the cells at low density in the absence of drug. Representative images are shown (left) and colony number is quantified (right). (E) Telomere restriction fragment length assay was performed on 1μg of genomic DNA in the indicated cell lines. All data are represented as mean ± SEM from at least 3 independent experiments. * indicates p < 0.05 using a two-sided Students t-test. Scale bar indicates 20 microns.

Supplementary Figure 4. The number of ATRX foci increases in senescent but not quiescent LS8817 cells. (A)
ATRX immunofluorescence was carried out in LS8817 cells in which senescence was induced by MDM2 knockdown as described in the legend to Supplementary Figure 2. Representative images are shown below. (B) LS8817 scr and LS8817 shATRX were treated with doxorubicin as described in the legend to Figure 1 and ATRX foci detected by immunofluorescence, counted, and average number plotted. Representative images are shown below. (C) LS8817 cells were serum starved (0.5%) for 5 days. The proliferation of these cells was assessed by BrdU incorporation. Senescence was assessed by the accumulation of SA-β-gal positive cells. The number of ATRX foci as measured as described previously. All data are represented as mean ± SEM from at least 3 independent experiments. * indicates p < 0.05 using a two-sided Students t-test. Scale bars indicate 20 microns.

Supplementary Figure 5. ATRX foci accumulate in different transformed cells that undergo CDK4i induced senescence but not during quiescence. (A) SNB19 glioma cells
were transduced with targeting vectors as described in the legend to Figure 1 and subsequently treated with PD0332991 and the accumulation of ATRX foci (left) and SA-β-gal positive cells (right) was measured. Representative images of ATRX foci are shown (bottom). (B) Three lung cancer derived cell lines were treated with PD0332991 for 7 days and the accumulation of ATRX foci (left) and SA-β-gal positive cells (right) was determined. Representative images of ATRX foci are shown (bottom). (C) MCF7 breast cancer cells were treated with PD0332991 for 7 days (left) or serum starved (0.5% serum) for 5 days (right). The accumulation of ATRX foci and SA-β-gal was measured as described in panel A. Representative images of ATRX foci are shown (bottom). All data are represented as mean ± SEM from at least 3 independent experiments. * indicates p < 0.05 using a two-sided Students t-test. Scale bars indicate 20 microns. Figure 6. Loss of ATRX does not affect PML foci. (A) PML foci were stained in LS8817 cells treated with CDK4 inhibitor and the number of foci was plotted as described in the legend to Figure 2A. (B) The presence of PML foci was assessed in LS8817 scr and LS8817 shATRX cells after treatment with PD0332991 for seven days. Graph represents the foldchange in the number of foci per cell in PD0332991 treated cells vs. untreated controls. (C) LS8817 cells were treated as described in the diagram in Figure 4A. PML foci were detected and quantified as described previously. Representative images are shown (left) and the number of PML foci per cell is quantified (right All data are represented as mean ± SEM from at least 3 independent experiments. * indicates p < 0.05 using a two-sided Students t-test. Scale bar indicates 20 microns.  . (B-E) The effect of reducing HRAS on the accumulation of SA-β-gal (B), SAHF (C), or ATRX foci (D) and the induction of SASP mRNAs as measured by qPCR (E) was determined as described.

Supplementary
Representative SAHF images are shown in panel C. All data are represented as mean ± SEM from at least 3 independent experiments. * indicates p < 0.05 using a two-sided Students t-test. Scale bar indicates 20 microns. Figure 9. Reducing ATRX suppresses the SASP in response to CDK4 inhibition in LS8817 cells. ATRX was reduced in LS8817 cells as described in the legend to Figure 1 and cells were treated with PD0332991 for seven days (7D PD). (A) Conditioned media was collected and cytokine expression measured as described in the methods. Representative images are shown (left) and induction of secreted SASP factors is shown (right). (B) mRNA was harvested from the indicated LS8817 cells as described in the methods and the induction of expression of the indicated SASP factors was measured.

Forward Primer
Reverse Primer