WO2016191740A1 - Methods for drug discovery - Google Patents

Abstract

The present invention provides assays for identifying drug candidates that regulate cellular senescence. In particular, the present invention relates to the use of ATRX foci as a biomarker for determining whether one or more drug candidates regulate senescence in a cell line. In one non-limiting embodiment, the present invention provides assays to identify compounds that can be used to prevent and/or reduce senescence. The present invention is based, at least in part, on the discovery that the number of ATRX foci increases in cells that undergo senescence. Accordingly, in non-limiting embodiments, the present invention provides for assays and kits for identifying compounds that may be useful in treating age-related diseases.

Description

METHODS FOR DRUG DISCOVERY

CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to United States Provisional

Application Serial No. 62/167,152, filed May 27, 2015, to which priority is claimed and the contents of which is incorporated herein in its entirety.

GRANT INFORMATION

This invention was made with government support under Grant No. P50CA140146 awarded by the National Institutes of Health. The government has certain rights in the invention.

1. INTRODUCTION

This present invention relates to the amount of ATRX foci as a biomarker for determining whether cells are undergoing senescence. In particular, the amount of ATRX foci can be used as a biomarker to identify drug candidates for use in combinatorial cancer treatments and/or as a biomarker for identifying compounds that prevent or inhibit senescence. As such, evaluating ATRX foci may be used as part of a drug discovery process.

2. BACKGROUND OF THE INVENTION

Alpha-thalassemia/mental retardation syndrome X-linked (ATRX) is encoded by the atrx gene. ATRX is a SWI/SNF helicase/ATPase that can regulate gene expression via chromatin remodeling and is associated with pericentric and telomeric heterochromatin (McDowell et al. PNAS (1999); Eustermann et al. NSMB (2011)). Its primary clinical indication is mutations in the mental retardation syndrome a-thalassemia/MR, X-linked (ATRX syndrome) (Picketts et al. Am. J. Human Genet. (1996)).

Although ATRX can interact with several proteins that are involved in senescence including PML bodies (Xue et al. PNAS (2003); Luciani et al. J. Cell Science (2006)), HP1 proteins (McDowell et al. PNAS (1999); Eustermann et al. NSMB (2011)) and macroH2A (Ratnakumar et al. Genes and Dev. (2012)), ATRX has never been directly associated with senescence, and studies have shown that ATRX negatively regulates macroH2A (a facilitator of senescence-associated heterochromatic foci formation) incorporation into chromatin (Ratnakumar et al. Genes and Dev. (2012)).

There is a link between age-related pathologies and senescence. The appearance of senescent cells in premature aging skin can lead to age-related dermal and epidermal thickening and loss of collagen (Waaijer et al. Aging Cell (2012)). Senescence in astrocytes can promote age-related neurodegeneration giving rise to cognitive impairment and contribute to Alzheimer's and Parkinson's diseases (Bitto et al. Exp. Cell Res. (2010)). Senescent chondrocytes play a role in osteoarthritis (Roberts et al. Eur. Spine J. (2006)) and senescent endothelial cells and smooth muscle cells may contribute to atherosclerosis (Matthews et al. Circ. Res. (2006)). A causal link between senescence and aging was first established in a progeroid mouse model in which pl6-positive senescence cells could be eliminated early in mouse life and even later in life and many age-related dysfunctions were delayed by the elimination of these senescent cells (Baker et al. Nature (2012); Baker et al. Nature. 530(7589): 184-189 (2016)).

Efforts towards drug discovery continue to use vast technical and financial resources to identify and develop new and useful drugs. However, the identification of new and useful drug therapies for cancer and age-related diseases and pathologies continues to be difficult. The development of more effective cancer therapies is important for increasing a patient's survival and can improve treatment success. Furthermore, a therapy for treating age-related diseases can affect a person's quality of life and can delay the onset of such diseases. Therefore, there remains a need in the art for assays of identifying compounds that may be effective in such treatment regimens.

3. SUMMARY OF THE INVENTION

The present invention relates to assays and compositions for identifying compounds that regulate cellular senescence. In particular, the disclosed assays relate to the use of the amount of ATRX foci as a biomarker for identifying drug candidates that induce senescence or inhibit and/or reduce senescence and/or geroconversion in a cell line. In certain embodiments, the present invention provides assays and compositions that use ATRX foci as a biomarker to identify drug candidates for use in treating age-related diseases. In certain embodiments, the present invention relates to assays that use ATRX foci as a biomarker to identify drug candidates for use in cancer treatments and, in particular, for identifying combinations of drugs for use in cancer therapy. The present invention is based, at least in part, on the discovery that the number of ATRX foci increases in cells that undergo senescence.

In certain non-limiting embodiments, the present invention provides for an assay for identifying drug candidates that reduce (retard, inhibit) and/or prevent senescence, e.g., for use in treating an age-related disease. In certain embodiments, the assay comprises (a) treating one or more cells with a compound and/or modality that induces senescence; (b) treating the one or more cells with a drug candidate; and (c) determining the number of ATRX foci per cell in the compound and drug candidate-treated cells, as compared to a reference, where a decrease in the number of ATRX foci per cell or the absence of an increase in the number of ATRX foci per cell compared to the reference is indicative that the drug candidate is likely to be useful as a therapy for treating an age-related disease. In certain embodiments, the reference sample is one or more cells treated with the compound that induces senescence alone. In certain embodiments, the one or more cells are responder cells, e.g., LS8817 cells.

In certain embodiments, the compound and/or modality that induces senescence and the drug candidate are added to the one or more cells simultaneously. In certain embodiments, the compound and/or modality that induces senescence and the drug candidate are added to the one or more cells at different timepoints. In certain embodiments, the compound and/or modality that induces senescence is a CDK4 inhibitor. In certain embodiments, the compound and/or modality that induces senescence is a modality that reduces HRAS expression. In certain embodiments, the compound and/or modality that induces senescence is a modality that reduces MDM2 expression. In certain embodiments, the drug candidate is selected from the group consisting of small chemical molecules, biologies, and peptides.

In certain embodiments, an assay for identifying drug candidates for use in treating an age-related disease can comprise (a) treating one or more cells with a CDK4 inhibitor, where the cells enter a quiescent state upon treatment with the CDK4 inhibitor ; (b) treating the one or more CDK4-inhibitor treated cells with shMDM2; (c) treating the one or more cells with a drug candidate; and (d) determining the number of ATRX foci per cell in the CDK4 inhibitor, shMDM2 and drug candidate-treated cells, as compared to a reference, where a decrease in the number of ATRX foci per cell or the absence of an increase in the number of ATRX foci per cell compared to the reference is indicative that the drug candidate is likely to be useful as a therapy for treating age-related disease. In certain embodiments, the one or more cells are treated with the drug candidate and the shMDM2

simultaneously. In certain embodiments, the one or more cells are treated with the drug candidate and the shMDM2 at different timepoints. In certain embodiments, the reference sample is one or more cells treated with the CDK4 inhibitor and the shMDM2 in the absence of the drug candidate. In certain embodiments, the one or more cells are non-responder cells, e.g., LS8107 cells.

In certain embodiments, the shMDM2 is expressed from a vector present in the one or more cells under the control of doxycycline. In certain embodiments, the CDK4 inhibitor is PD0332991. In certain embodiments, the drug candidate is selected from the group consisting of small chemical molecules, biologies, and peptides.

In certain embodiments, an assay for identifying drug candidates for use in treating an age-related disease, can comprise (a) treating one or more cells with a senescence-inducing compound and/or modality, where the cells enter a senescent state upon treatment with the compound and/or modality that induces senescence; (b) exogenously expressing MDM2 in the one or more cells treated with the senescence- inducing compound; (c) reducing the exogenous expression of MDM2 and treating the one or more cells with a drug candidate; and (d) determining the number of ATRX foci per cell as compared to a reference, where a decrease in the number of ATRX foci per cell or the absence of an increase in the number of ATRX foci per cell compared to the reference is indicative that the drug candidate is likely to be useful as a therapy for the treating age-related disease. In certain embodiments, the one or more cells are responder cells, e.g., LS8817 cells. In certain embodiments, the compound and/or modality that induces senescence is a CDK4 inhibitor.

In certain embodiments, the number of ATRX foci per cell is determined by immunofluorescence. In certain embodiments, the ATRX foci is detected by an ATRX-specific antibody. In certain embodiments, the ATRX foci is detected by an ATRX-specific antibody. For example, and not by way of limitation, the ATRX-specific antibody is the antibody sold by Bethyl, Catalog No. A301-045A ("the '045 Ab"), a fragment thereof (e.g., a Fab fragment, a Fab₂, or the variable region comprised in a chimeric molecule) or an antibody that competitively inhibits binding of the '045 Ab to ATRX. The present invention further provides kits for identifying drug candidates for use in treating an age-related disease. In certain embodiments, the kit comprises a means for detecting ATRX foci. In certain embodiments, the means for detecting ATRX foci is an ATRX-specific antibody. For example, and not by way of limitation, the ATRX-specific antibody is the antibody sold by Bethyl, Catalog No. A301 -045 A ("the '045 Ab"), a fragment thereof (e.g., a Fab fragment, a Fab₂, or the variable region comprised in a chimeric molecule) or an antibody that competitively inhibits binding of the '045 Ab to ATRX.

In certain embodiments, a kit of the present invention can further comprise one or more cells. In certain embodiments, the one or more cells are non- responder cells, e.g., LS8107 cells. In certain embodiments, the one or more cells are responder cells, e.g., LS8817 cells. In certain embodiments, the one or more cells comprise a vector comprising shMDM2 under the control of doxycycline. In certain embodiments, the one or more cells comprise a vector comprising MDM2 under the control of doxycycline. In certain embodiments, the one or more cells comprise a vector comprising shURAS under the control of doxycycline. In certain

embodiments, the kit can further comprise a senescence inducing compound and/or modality, e.g., a CDK4 inhibitor.

4. BRIEF DESCRIPTION OF THE FIGURES

FIGURE 1 depicts a schematic representation of an assay according to a non-limiting embodiment of the present invention.

FIGURE 2 shows that ATRX is required for senescence induced by the inhibition of CDK4 by PD0332991 (also referred to herein as "PD").

FIGURE 3 shows that the number of ATRX foci increased upon treatment with doxorubicin as detected by immunofluorescence. Representative immunofluorescence images are shown on top and the mean and standard deviation of the foci counts from at least 50 cells in each of three independent experiments are plotted below (*p<0.01).

FIGURE 4 shows LS8817 cells expressing either a scrambled (shSCR) or ATRX (shATRX) shRNA that were treated with 100 μΜ doxorubicin for five days and the effect on the accumulation of senescence associated β-galactosidase (β-gal) positive cells and expression of p53 and ATRX were determined. This experiment was repeated twice with independent transductants. FIGURE 5A-H shows that replicative senescence of untransformed cells is associated with an increase in the number of ATRX foci per cell. (A) Expression of phosphorylated Rb, a marker of cell proliferation, decreased upon passaging the cells to a high passage number. At passage 25, accumulation of perinuclear associated β-galactosidase (SA-P-gal) was observed (B) and the number of ATRX foci per cell significantly increased at passage 25 as compared to cells at passage 11 (C). (D) Primary human WI38 fibroblasts were cultured to replicative exhaustion. (E-H) Samples were taken at different passages (P9, PI 4, P19 and P24) for BrdU incorporation (E), immunoblot (F), accumulation of SA-P-gal positive cells (G) and cytological analysis of ATRX foci (H).

FIGURE 6A-C shows that the treatment of cells with PD0332991 followed by shMDM2 resulted in a significant increase in the number of ATRX foci within cells that do not senesce (i.e., quiesce) in response to PD0332991 treatment compared to cells that were not treated with PD0332991 or were treated with

PD0332991 alone. (A) shMDM2 expression resulted in a decrease in MDM2 expression. (B) Treatment of cells with PD0332991 followed by shMDM2 resulted in the accumulation of SA-p-gal. (C) Treatment of cells with PD0332991 followed by shMDM2 resulted in an increase in the number of ATRX foci per cell.

FIGURE 7A-H shows that the expression of Flag-MDM2 blocks CDK4 inhibition-induced senescence. (A) Schematic of the tet-on-MDM2-Flag construct and the experimental conditions. (B) Treatment of cells with PD0332991 in the presence of exogenous MDM2 prevented an increase in the number of ATRX foci per cell compared to treatment with PD0332991 alone. (C) Treatment of LS8817 cells with 10 μΜ doxycycline to induce FLAG-MDM2 expression (10 μΜ Dox), treatment with 10 μΜ doxycycline for 2 days to induce FLAG-MDM2 expression followed by the addition 0.1 μΜ PD0332991 in the presence of doxycycline for an additional 7 days (0.1 μΜ PD/10 μΜ Dox), treatment with 10 μΜ doxycycline for 2 days to induce FLAG-MDM2 expression followed by the addition of 0.1 μΜ

PD0332991 in the presence of doxycycline for 2 days to arrest the cells in quiescence and the removal of doxycycline while treating the cells with 0.1 μΜ PD0332991 alone for another 5 days (0.1 μΜ PD/- Dox), or treatment with 0.1 μΜ PD0332991 for 7 days (0.1 μΜ PD). SA-P-gal accumulation and the number of ATRX foci were analyzed under the conditions described. (D) Treatment of LS8817 cells with 10 μΜ doxycycline for 2 days to induce MDM2 expression followed by the addition 0.1 μΜ PD0332991 in the presence of doxycycline for an additional 2 days (0.1 μΜ PD/10 μΜ Dox) resulted in the reduction in BrdU incorporation as compared to control cells. (E) Kinetics of SA-P-gal accumulation and ATRX foci as a function of time (days). Red, synchronous cells; Blue, asynchronous cells. Colored lines in the x-axis indicate the length of time it takes for the phenotype to transition from low to high levels. (F) The appearance of different phenotypes can be measured in the synchronous system. LS8817 tet-on-FMDM2 cells were treated with doxycycline for two days before addition of 0.1 μΜ PD0332991. Within a day, the cells exited the cell cycle and failed to accumulate BrdU, and Rb phosphorylation and cyclin A levels decreased. By forcing the expression of MDM2 with a doxycycline regulated promoter, these cells entered a G1/G0 quiescent state. Removing doxycycline resulted in the reduction in MDM2 levels and an increase in the number of ATRX. These cells can be induced to reenter the cell cycle when CDK4i was removed, representing a second quiescent state (QII). After 24 hours, the cells cannot reenter the cell cycle upon removal and enter senescence (Sen_entiy) but do not yet express SA-P-gal and SAHF, which occurs in the next 24 hours (Sen_maint). The foci shown (bottom) are as follows: (1) foci formation in untreated cells (cycling); (2) LS8817 tet-on-FMDM2 cells treated with doxycycline and PD0332991 (QI); (3) cells treated with doxycycline and PD0332991 followed by doxycycline removal (QII); (4) 24 hours after the timepoint of (3)(Sen_entry); (5) 48 hours after the timepoint of (3)(Senmaint); (6) sample (5), in which ATRX was knocked down for 10 days; and (7) sample (5), in which

doxycycline was readded and cells were incubated for 10 days. (G) Expression of the tet-on-MDM2-Flag construct in the presence of PD0332991 and doxycycline in the glioma cell line S B 19 and the non-small cell lung cancer cell line H1975 limited ATRX foci formation as determined by analyzing the number of ATRX foci. (H) Optimization of a cell-based system that can be used in the disclosed methods.

FIGURE 8A-G shows that the reduction in MDM2 expression induces senescence. (A) Schematics of the tet-on-shMDM2 constructs and the experimental conditions. (B) Treatment of LS8817 cells with doxycycline to induce expression of shMDM2 resulted in an increase in the average number of ATRX foci per cell compared to the control. (C) In the liposarcoma cell line LS8107, treatment with 1 μΜ PD0332991 did not induce a shift in ATRX foci formation, but treatment with 10 μΜ doxycycline and the reduction of MDM2 expression resulted in an increase of the number of ATRX foci per cell. (D) The expression of shMDM2 in the presence or absence of PD0332991 resulted in a decrease in MDM2 expression. (E) Senescence was not induced in LS8107 tet-on shMDM2 cells treated with 1 μΜ PD0332991 for 7 days; whereas, senescence was induced in such cells when treated with 1 μΜ

PD0332991 for 2 days followed by 10 μΜ doxycycline for 5 days as measured by an increase in ATRX foci per cell and SA-P-Gal activity. (F) Treatment of LS8313 cells with doxycycline for 5 days to induce expression of shMDM2 resulted in an increase in the number of ATRX foci per cell as compared to the control. (G) Analysis of a putative senescence inhibitor. Data was captured on an In cell 6000 analyzer demonstrating that senosuppressors can be identified. Dose dependent activity and the number of cells detected with DAPI (relative to PD0332991 alone) is shown. This platform allows the acquisition of up to four additional parameters.

FIGURE 9A-K. ATRX is necessary for senescence. (A-E) LS8817 cells were transduced with a scrambled (shSCR) or an ATRX specific (shATRX) lentiviral knockdown vector and subsequently treated with 100 nM doxorubicin for 7 days, and the accumulation of SA-P-gal positive cells (A) and SAHF -positive cells (B) was measured in three independent experiments and the mean and standard deviation were plotted. Expression of three of the liposarcoma senescence-associated secretory phenotype (SASP) genes identified in Kovatcheva et al. (2015) were analyzed by qPCR (C). Immunoblot was used to confirm that ATRX levels were reduced and that reducing ATRX did not affect the accumulation of p53, a marker of the doxorubicin induced DNA damage response (D). Immunofluorescence was used to confirm that reducing ATRX did not affect the doxorubicin-induced accumulation of 53BP1 or γΗ2ΑΧ foci (E). (F) The LS8817 cells described above were

subsequently transduced with another shRNA expression vector targeting MDM2 and the effect on accumulation of SA-P-gal (top), protein abundance (middle), and SAHF formation (bottom) were determined. (G) The same as in panel F but with LS8313 cells. SAHF was not examined in these cells. (H-K) U20S cells were transfected with a vector expressing wild type ATRX and stable transformants selected with G418 and sorted using a flow cytometer to recover a GFP-low population as described in the methods. These cells were subsequently treated with the CDK4 inhibitor PD0332991 (PD) for 7 days and expression of proteins was measured by immunoblot (H), the accumulation of SA-P-gal (I) and SAHF (J) was measured cytologically by staining, and clonogenicity was measured by crystal violet staining three weeks after the removal of the drug and replating at low density in the absence of drug (K). * indicates p <0.05.

FIGURE lOA-C. ATRX is required for DNA damage induced senescence in LS0082 cells. (A-C) LS0082 cells were transduced with a scrambled (shSCR) or an ATRX specific (shATRX) lentiviral knockdown vector and

subsequently treated with 100 nM doxorubicin for 7 days, and the accumulation of SA-P-gal positive cells (A) was measured in three independent experiments and the mean and standard deviation were plotted. Immunoblot was used to confirm that ATRX levels were reduced and that reducing ATRX did not affect the accumulation of p53, a marker of the doxorubicin induced DNA damage response (B).

Immunofluorescence was used to confirm that reducing ATRX did not affect the doxorubicin-induced accumulation of 53BP1 or γΗ2ΑΧ foci (C).

FIGURE 11 A-L. ATRX accumulates in nuclear foci in senescent cells. ATRX mutants were transfected into U20S cells and stable transformants were isolated as described in the legend to Figure 9. (A) A diagram indicating the domains and amino acid residues numbered as annotated on UniProt is shown. Arrows indicate the paired sequencing primers used to confirm the mutations. (B) The accumulation of SA-P-gal positive cells was scored 7 days after PD0332991 (PD) treatment. Values for parallel empty vector transfected cells are shown. (C) The amount of ATRX mRNA was measured by qPCR (top) using a probe that binds within the middle of the transcript (sequence in Table 1) and ATRX protein by immunoblot (bottom). (D) The number of ATRX foci was counted in individual LS8817 cells 2, 4, and 7 days after treatment with PD0332991. The number of ATRX foci was also counted in control untreated cycling cells. Bar graphs indicate the percentage of cells that have the indicated number of foci. At least 150 cells were counted on each day. (E) LS8817 cells were treated with PD0332991 for seven days and the co-localization of ϋΡΙγ and PML foci with ATRX foci was determined by immunofluorescence.

Representative images are shown. White arrows indicate ATRX foci that do not co- localize with ϋΡΙγ. (F) The fraction of ATRX foci colocalized with ϋΡΙγ and PML foci was quantified. Circles in the Venn diagrams are drawn to scale relative to the number of the indicated foci in each condition. (G) Each indicated liposarcoma cell line was treated with PD0332991 for 7 days and ATRX foci analyzed as described in the legend to Figure 1 ID. The average number of foci per cell is plotted. The effect of PD0332991 on whether it induces quiescence or senescence is noted. (H) ATRX immunofluorescence was carried out in LS8817 cells in which senescence was induced by MDM2 knockdown as described in the legend to Figure 9F. Knocking down ATRX abrogates the appearance of foci detectable with an ATRX antibody. (I) ATRX immunofluorescence was carried out in LS8817 cells treated with doxorubicin as described in the legend to Figure 9A either with or without ATRX knockdown as indicated. (J) ATRX immunofluorescence was carried out in U20S cells transfected with each mutant or wild type protein. The average number of ATRX foci per cell is plotted (above) and representative images are shown (below). (K) ATRX

immunofluorescence was carried out in two normal human diploid fibroblasts (EVIR90 and WI38) induced to senesce by treatment with lOOnm doxorubicin (doxo), γ- irradiation (IR) or expression of KRAS^V12. Cells were evaluated 7 days following exposure to each stressor. SA-P-gal was also scored (top), the number of ATRX foci per cell was counted (top) and representative images of each are shown (bottom). (L) LS8817 cells were serum starved (0.5%) for 5 days. These cells exited the cell cycle as monitored by BrdU (top left), but did not senesce as monitored by a failure to downregulate MDM2 (bottom left), or accumulate SA-P-gal positive cells (top right) and the number of ATRX foci fails to increase (top middle and bottom right), indicates p < 0.05.

FIGURE 12A-L. ATRX foci accumulates in different transformed cells that undergo CDK4i induced senescence but during quiescence, autophagy or differentiation. (A) The glioma cell line S B19 undergoes senescence when treated with PD0332991 (Kovatcheva et al. 2015). Reducing ATRX (left) compromises the capacity of PD0332991 to induce senescence as measured by SA-P-gal (right). (B)The breast cancer cell line MCF7 was treated with PD0332991 for 7 days to induce senescence (Kovatcheva et al. 2015) and ATRX was analyzed as described in the legend to Figure 11. (C) Three lung cancer derived cell lines were treated with PD0332991 for 7 days and their capacity to undergo senescence was scored by SA-β- gal (upper left). Immunoblots are shown below to confirm that MDM2 loss was correlated to CDK4i induced senescence independent of p53 status in this cell type. Clongenicity assays demonstrating senescence of these three lines were previously reported in Kovatcheva et al. (2015). The number of ATRX foci (upper right) were analyzed as described in the legend to Figure 11. (D-G) MCF7 cells were serum starved as described (Campisi, Cellular senescence: putting the paradoxes in perspective. Curr Opin Genet Dev 21, 573 107-112 (2011)). The conversion of LC3-I to LC3-II was monitored by immunoblot (D) and puncta staining of GFP-LC3 (E). The accumulation of SA-P-gal (F) and the number of ATRX foci (G) were also measured. (H-J) The same as in (D-G) but in MCF-IOA cells. (K-L) ATRX does not accumulate in differentiated adipocytes. LS8107 cells were induced to differentiate in Gimble media for 10 days. Differentiation markers were assessed by RT-qPCR (K) and ATRX foci were monitored by immunofluorescence (L).

FIGURE 13A-H. ATRX is required for establishment and maintenance of ΤΙΡΙγ SAHF in senescent cells. (A) ATRX was reduced in cells as described in the legend to Figure 9, and cells were then treated with PD0332991 for seven days. The number of ΤΙΡΙγ (left) and PML (right) foci per cell were counted. Graphs are expressed as a fold-change in the number of foci per cell in PD0332991 treated cells vs untreated controls. (B) A schematic of the experiment used to test the requirement of ATRX in senescence maintenance as explained in the accompanying text. (C-H) Cells treated as described in Figure 13B were fixed and analyzed for BrdU

incorporation (C), SA-P-gal accumulation (D), ATRX foci formation (E), SASP markers (F), clonogenic growth potential (G) and ΤΙΡΙγ (SAFIF) foci formation (H). In Figure 13E and Figure 13H representative images are shown (right) and foci are quantified (left). * indicates p < 0.05.

FIGURE 14A-C. Loss of ATRX does not affect PML foci. (A) An increase in PML foci is detected as early as one day following CDK4 inhibition.

Graphs are as described in the legend to Figure 1 ID. The loss of ATRX in senescence cells induced by CDK4 inhibition was assessed. The number of PML foci per cell was quantified (B) and representative images are shown (C). * indicates p < 0.05.

FIGURE 15A-D. ATRX affects E2F and EZH2 target gene expression in senescent cells. (A) Hierarchical clustering based on expression of all genes sequenced by RNA-seq. Each row represents a single sample, and the conditions are indicated above; RNA-seq was performed in duplicate on each condition. (B) Venn Diagram indicating the number of genes that were differentially expressed with a fold change of at least -1.8 or 1.8 and an FDR < 0.05 across PD-treated vs. control samples in unperturbed and ATRX-deficient LS8817 878 cells; the genes that were found in common across both cell types are indicated. (C) Enrichr analysis of the top predicted transcription factors that regulate the up-regulated and down-regulated gene lists from Figure 15B. The negative log of the p-value for the enrichment scores is plotted. A similar analysis of GO categories is provided in Table 3. (D) Gene set enrichment analysis (GSEA) was performed on across PD-treated vs control samples in unperturbed and ATRX-deficient LS8817 cells, specifically analyzing E2F4, GO DNA repair and EZH2 gene signatures. The profiles are shown with their

corresponding normalized enrichment score (NES) and FWER p-values.

FIGURE 16 A- J. ATRX directly binds to and represses HRAS in response to CDK4 inhibition. (A) Venn diagram indicating the number of ATRX- specific summits identified by ChIP sequencing performed in untreated (cycling) LS8817 cells, senescent LS8817 cells treated with either PD0332991 for seven days (CDK4i) or doxorubicin for five days, and quiescent cells induced by growth in low serum for five days (0.5% serum starved). The senescence-specific summits are circled in blue. (B) Pie chart summarizing the distribution of the 166 senescence specific summits within gene bodies, associated with promoters or in intergenic regions. (C) Enrichr analysis of the top predicted transcription factors that regulate the "gene body" and "promoter" associated genes from (B). The negative log of the p- value for the enrichment scores is plotted. The specific genes that comprise each gene set can be found in Table 2. (D) qPCR analyses of expression of the ATRX bound genes identified to be downregulated or "off in senescence" (top), and upregulated or "on in senescence" (bottom). Expression was measured in untreated and PD0332991 treated (7days) senescent and quiescent cells as indicated in the figure, and the ratio of expression in CDK4 treated cells divided by that in the untreated cells plotted. The solid line indicates no change in expression; dashed lines indicate cutoffs to demonstrate significant repression or activation (>1.8 fold change). (E) ATRX was stably reduced in LS8817 cells and expression of genes was measured as described above. (F) U20S cells were stably transfected with wild type ATRX and expression of genes was measured as described above. (G) Small-scale ATRX ChIP experiments were performed in untreated controls and senescent LSS8817 and LS0082 cells, as well as quiescent LS8107 cells. The relative enrichment of ATRX at the indicated loci was analyzed by qPCR. (H) Genome browser view of ATRX enrichment at the HRAS locus. Orange bars indicate the peaks called via the IDR algorithm and green bars indicate DNA sequences predicted to form G-quadruplex structures. (I) Small-scale ATRX ChIP was performed in LS8817 2, 4 and 7 days following treatment with PD0332991 and in untreated cycling cells (CTRL). The level of ATRX enrichment specifically at the HRAS locus as determined by qPCR is plotted. (J) The same as in (D) but ATRX was immunoprecipitated from U20S cells transfected with either wild type (WT) or mutant (C240G, V588E) ATRX as described in Figure 11. * indicates p < 0.05.

FIGURE 17. URAS is repressed in a variety of different transformed cell lines undergoing senescence. Senescence was induced in LS8817 cells by knockdown of MDM2 as previously described (Baker et al. Naturally occurring pl6(Ink4a)-positive cells shorten healthy lifespan. Nature 530, 184-189 (2016)). In SNB19 glioma cells and A549 and H1975 lung cancer cells senescence was induced by treatment with PD0332991 for 7 days as described (Kovatcheva et al. (2015)). In H358 lung cancer cells, quiescence was induced by treatment with PD0332991 for seven days. Gene expression was analyzed as described in Figure 16.

FIGURE 18A-E. Reduction of HRAS is sufficient to drive geroconversion: the transition from quiescence into senescence. (A) HRAS, KRAS, and NRAS were stably knocked down with two independent hairpins each in cycling LS8817 cells and transcript levels were measured by qPCR. Expression levels were normalized to 1 in shSCR cells. (B-E) LS8817 cells were induced to quiesce via serum starvation for five days and subsequently infected with shRNA encoding lentiviruses. After an additional five days of selection, the accumulation of SA-P-gal (B), HRAS mRNA (C) and SAHF (D) were measured. Long term clonogenicity when replated in complete medium was also measured (E). * indicates p < 0.05.

FIGURE 19 depicts a schematic representation of an assay for identifying compounds that block, reduce and/or inhibit senescence according to a non-limiting embodiment of the present invention.

FIGURE 20 depicts a schematic representation of an assay for identifying compounds that block, reduce and/or inhibit geroconversion according to a non-limiting embodiment of the present invention.

FIGURE 21 depicts a schematic representation of an assay for identifying compounds that block, reduce and/or inhibit geroconversion according to a non-limiting embodiment of the present invention.

FIGURE 22A-F. Loss of HRAS can drive geroconversion in LS8017 cells and LS8817 cells lacking ATRX that have been induced to quiesce via CDK4 inhibition. LS8107 cells were treated with PD0332991 for seven days and then infected with lentiviral vectors targeting HRAS. PD treatment was maintained as viral selection occurred. (A) The efficacy of knockdown was assessed by qPCR. (B) Loss of HRAS did not affect PD-induced cell cycle exit, as measured by BrdU incorporation. (C-D) The effect of HRAS loss on senescence was measured by SA-β- gal (C) and SAHF (D). (E-F) LS8817 cells in which ATRX was stably knocked down were treated with PD0332991 for 7 days (7D PD) in order to induce quiescence. Cells were then infected with a lentiviral vector targeting a non-specific sequence (shSCR) or HRAS. The number of SA-p-gal positive (E) and SAHF positive (F) cells are plotted. * indicates p < 0.05.

FIGURE 23. Graphical summary of ATRX and its role in senescence.

5. DETAILED DESCRIPTION OF THE INVENTION For clarity and not by way of limitation the detailed description of the invention is divided into the following subsections:

(i) ATRX foci as a biomarker;

(ii) methods of use; and

(iii) kits.

5.1 ATRX FOCI AS A BIOMARKER

Alpha-thalassemia/mental retardation syndrome X-linked is denoted

ATRX herein.

The present invention discloses ATRX foci as a biomarker for senescence. The term "ATRX foci," as used herein, refers to ATRX-positive punctate structures that can be visualized within a cell.

In a specific, non-limiting embodiment, ATRX foci may be detected using an immunodetection reagent specific for an ATRX protein.

In a specific, non-limiting embodiment, an ATRX protein may be a human ATRX protein having the amino acid sequence as set forth in NCBI database accession no. NP_000480.

ATRX proteins for non-human species are known or can be determined according to methods known in the art, for example, where the sequence is the allele represented in the majority of the population.

In a specific, non-limiting embodiment, an ATRX protein may be a mouse ATRX protein having the amino acid sequence as set forth in NCBI database accession no. NP 033556. In a specific, non-limiting embodiment, an ATRX protein may be a rat ATRX protein having the amino acid sequence as set forth in NCBI database accession no. XP_003754859.

Methods for detecting and/or determining the number of ATRX foci can include, but are not limited to, immunofluorescence and immunoglobulin- mediated assays, and other techniques known in the art.

In certain, non-limiting embodiments, immunohistochemistry can be used for detecting ATRX foci. For example, an antibody that binds ATRX ("ATRX- specific antibody") or a fragment thereof can be brought into contact with, for example, a thin layer of cells, followed by washing to remove unbound antibody, and then contacted with a second, labeled antibody, e.g., secondary antibody. Labeling can be by fluorescent markers, enzymes, such as peroxidase, avidin or radiolabeling. The labeling can be scored visually using microscopy and the results can be quantitated. Non-limiting examples of antibody fragments include Fab, Fab', F(ab')₂, Fv, single chain Fv (scFv) or a variable region comprised in a chimeric molecule.

In non-limiting embodiments, methods of ATRX detection may utilize the ATRX-specific antibody sold by Bethyl, Catalog No. A301-045A ("the '045 Ab"), a fragment thereof, or an antibody that competitively inhibits binding of the Ό45 Ab to ATRX. In certain non-limiting embodiments, methods of ATRX detection may utilize the ATRX-specific antibody sold by Santa Cruz Biotechnology, Catalog No. sc-55584 (the "D5 antibody"), a fragment thereof, or an antibody that

competitively inhibits binding of the D5 antibody to ATRX.

5.2 METHODS OF USE

The present invention relates to assays for identifying compounds that regulate cellular senescence by analyzing ATRX foci as a biomarker. As discussed in the Examples section below, senescence in cancer cells and untransformed cells correlates with an increase in the number of ATRX foci in each cell, and disruption of ATRX prevents senescence.

"Senescence," as used herein, refers to a cell state in which the cell has little or no proliferative capacity as compared to quiescence, where a cell retains the capacity for proliferation. In certain embodiments, "senescence" or a "senescent state" refers to a cellular state where an increase in the expression of at least one marker, or at least two markers, or at least three markers of the senescent phenotype selected from the group consisting of SA-P-gal, senescence-associated

heterochromatin foci and elaboration of the senescence-associated secretory program is observed in the cell.

In certain embodiments, the present invention provides for assays for identifying drug compounds that induce senescence. In certain embodiments, such drug compounds can be used in combination with other cancer agents as a cancer therapy.

In certain embodiments, the present invention further provides for assays that can be used for identifying compounds that prevent, inhibit, minimize and/or reduce senescence. In certain embodiments, such drug compounds can be used to prevent, minimize, inhibit and/or reduce senescence induced by aging and can be used to treat age-related diseases. In certain embodiments, the present invention further provides for assays that can be used for identifying compounds that prevent, inhibit, minimize and/or reduce geroconversion.

In certain embodiments, the disclosed assays can be used to screen large libraries of compounds. In non-limiting embodiments, the assays of the disclosed invention can be used to prioritize large numbers of new compounds for further drug development and/or can identify new compounds that can be used in combination with compounds currently being used clinically.

Candidate compounds (also referred to herein as drug candidates) to be screened in the currently disclosed assays include pharmacologic agents already known in the art as well as compounds previously unknown to have any

pharmacological activity. One non-limiting example of a library that includes compounds that can be screened using the disclosed assays is an FDA approved library of compounds that can be used by humans. Synthetic compound libraries are commercially available from a number of companies including Maybridge Chemical Co. (Trevillet, Cornwall, UK), Comgenex (Princeton, N.J.), Brandon Associates (Merrimack, N.H.) and Microsource (New Milford, Conn.), and a rare chemical library is available from Aldrich (Milwaukee, Wis.). Alternatively or additionally, libraries of natural compounds in the form of bacterial, fungal, plant and animal extracts can be used in the disclosed assay, for example, from Pan Laboratories or MycoSearch. For example, and not by way of limitation, the drug candidates can include medicaments; vitamins; mineral supplements; substances used for the treatment and/or prevention of cancer; or prodrugs, which become biologically active or more active after they have been placed in a physiological environment.

Additional non-limiting examples of drug candidates include small molecules, antibiotics, antivirals, antifungals, enediynes, heavy metal complexes, hormone antagonists, non-specific (non-antibody) proteins, sugar oligomers, aptamers, oligonucleotides (e.g., antisense oligonucleotides that bind to a target nucleic acid sequence (e.g., mRNA sequence)), siRNA, shRNA, peptides, proteins, radionuclides, and transcription-based pharmaceuticals. In non-limiting embodiments, potential drug candidates can include nucleic acids, peptides, small molecule compounds (e.g., pharmaceutical compounds), and peptidomimetics. Candidate compounds can be naturally occurring compounds or synthetic compounds. For example, and not by way of limitation, the candidate compounds can be isolated from microorganisms, animals or plants, or can be produced recombinantly or synthesized by chemical methods known in the art.

The assays of the present invention may be performed in multiwell formats, in microtiter plates, in multispot formats or in arrays. In certain non-limiting embodiments, the cells for use in the present invention can be cultured, grown and/or analyzed in 96-well microtiter plates. In certain non-limiting embodiments, the cells for use in the present invention can be cultured, grown and/or analyzed in 384-well microtiter plates.

In certain, non-limiting embodiments, immunohistochemistry can be used for detecting ATRX foci in the presently disclosed methods. For example, an ATRX-specific antibody can be brought into contact with, for example, a thin layer of cells, followed by washing to remove unbound antibody, and then contacted with a second, labeled, antibody, e.g., secondary antibody. Labeling (e.g., of an antibody) can be by fluorescent markers, e.g., fluorophores, enzymes, such as peroxidase, avidin or radiolabeling. Alternatively, or additionally, an ATRX-specific antibody that is conjugated to a fluorophore can be brought into contact with the cells, followed by washing to remove unbound antibody, without the need for a second, labeled antibody. Non-limiting examples of fluorophores include rare earth chelates or fluorescein and its derivatives, rhodamine and its derivatives, dansyl, umbelliferone, luciferases, e.g., firefly luciferase and bacterial luciferase (see U.S. Patent No.

4,737,456, incorporated by reference herein in its entirety) and luciferin. The number of ATRX foci per cell can be scored visually using microscopy and the results can be quantitated. In a specific non-limiting embodiment, the '045 Ab, or an antibody that competitively inhibits binding of the '045 Ab to ATRX, can be used to detect ATRX foci. In a specific non-limiting embodiment, the D5 antibody, or an antibody that competitively inhibits binding of the D5 antibody to ATRX, can be used to detect ATRX foci.

In certain embodiments, an antibody that competitively inhibits binding of the '045 Ab or the D5 antibody to ATRX refers to an antibody that blocks binding of the '045 Ab or the D5 antibody to ATRX in a competition assay by about 50% or more, e.g., about 55% or more, about 60% or more, about 65% or more, about 70%) or more, about 75% or more, about 80% or more, about 85% or more, about 90% or more, about 95% or more, about 98% or more or about 99% or more, and conversely, the '045 Ab or the D5 antibody blocks binding of the antibody to ATRX in a competition assay by about 50% or more, e.g., about 55% or more, about 60% or more, about 65% or more, about 70% or more, about 75% or more, about 80% or more, about 85% or more, about 90% or more, about 95% or more, about 98% or more or about 99% or more. An exemplary competition assay is described in

"Antibodies," Harlow and Lane (Cold Spring Harbor Press, Cold Spring Harbor, NY)(1988).

In certain embodiments, the number of ATRX foci per cell can be determined at least one day, at least two days, at least three days, at least four days, at least five days, at least six days, at least seven days, at least eight days or more after treatment with the drug candidate. For example, and not by way of limitation, the number of ATRX foci per cell can be determined 3 days after treatment with the drug candidate (see Figure 7H).

5.2.1. ASSAY FOR IDENTIFYING COMPOUNDS THAT INDUCE

SENESCENCE

The present invention provides assays for identifying drug candidates that can be used in a combinatorial cancer treatment. In particular, the present invention provides assays for identifying potential drug candidates that can be used to induce senescence in a cancer cell line. In certain embodiments, the assays of the present invention comprise analyzing the number of ATRX foci per cell observed in the cancer line following treatment with the drug combination.

The present invention provides assays for identifying drug candidates that may be effective as therapeutic agents for treating cancer early in the drug development and discovery process. Non-limiting examples of such cancers include soft tissue sarcomas, melanoma, breast cancer, lung cancer, liposarcoma, basal cell carcinoma and glioma (or glioblastoma). In non-limiting embodiments, the assays of the disclosed invention can be used to prioritize large numbers of new compounds for further drug development and/or can identify new compounds that can be used in combination with cancer agents that are currently being used clinically.

The disclosed assay of the present invention further provides a high- throughput screening method for identifying potential drug combinations that can be used to induce senescence in a cancer cell line. For purposes of illustration and not limitation, Figure 1 is a schematic representation of an exemplary assay for identifying potential drug combinations according to the disclosed invention. In certain non-limiting embodiments, the assay of the present invention 100 includes treating one or more cells with a first drug candidate 101. The amount of the compound that is applied to the cells can depend on the type of compound used as the first drug candidate and the number of cells being treated. In non-limiting embodiments, cells can be treated with a first drug candidate at a concentration of about 1 nM to about 1 M. For example, and not by way of limitation, the cells can be treated with a first drug candidate at a concentration from about 10 nM to about 100 μΜ, from about 100 nM to about 10 uM, from about 500 nM to about 10 μΜ, from about 750 nM to about 10 uM, from about 750 nM to about 5 μΜ or from about 750 nM to about 1 uM. For example, and not by way of limitation, a first drug candidate can include PI3-kinase inhibitors and estrogen receptor antagonists.

Non-limiting examples of drug candidates that can be analyzed using the disclosed assays are described above. In certain embodiments, candidate compounds to be screened in the currently disclosed assay include known cancer chemotherapy agents such as, but not limited to, actinomycin, all-trans retinoic acid, azacitidine, azathioprine, bleomycin, bortezomib, carboplatin, capecitabine, cisplatin, chlorambucil, cyclophosphamide, cytarabine, daunorubicin, docetaxel, doxifluridine, doxorubicin, epirubicin, epothilone, etoposide, fluorouracil, gemcitabine,

hydroxyurea, idarubicin, imatinib, irinotecan, mechlorethamine, mercaptopurine, methotrexate, mitoxantrone, oxaliplatin, paclitaxel, pemetrexed, teniposide, tioguanine, topotecan, valrubicin, vinblastine, vincristine, vindesine and vinorelbine. In particular non-limiting embodiments, the candidate compound is not a CDK4 inhibitor. For example, and not by way of limitation, the first and/or second drug candidates are not CDK4 inhibitors. In certain embodiments, the first candidate drug is not a CDK4 inhibitor. In certain embodiments, the second candidate drug is not a CDK4 inhibitor.

In non-limiting embodiments, cells for use in the disclosed assay can be a "non-responder cell." In certain non-limiting embodiments, a non-responder cell is a cell that when treated with an amount of the first drug candidate effective for inducing senescence in a responder cell does not increase expression of at least one marker, or at least two markers, or at least three markers, of the senescent phenotype selected from the group consisting of senescence-associated beta-galactosidase (SA-β- gal), senescence-associated heterochromatin foci and elaboration of the senescence- associated secretory program and/or does not increase the number of ATRX foci in the cell, e.g., nucleus.

A "responder cell," as used herein, refers to a cell that when treated with an effective amount of a drug candidate, e.g., a first drug candidate, or a compound that induces senescence, increases expression of at least one marker, or at least two markers, or at least three markers, of the senescent phenotype selected from the group consisting of SA-P-gal, senescence-associated heterochromatin foci and elaboration of the senescence-associated secretory program and/or increases the number of ATRX foci in the nucleus. In non-limiting embodiments, the mean level of nuclear ATRX foci increase may be at least 30%.

In non-limiting embodiments, cells for use in the disclosed assay can be any cell line that undergoes quiescence in response to CDK4 inhibition (or in response to treatment with the first drug candidate) but does not progress towards senescence as measured by accumulation of either SA-P-gal, senescence associated heterochromatic foci and/or elaboration of the senescence associated secretome. In certain non-limiting embodiments, a cell line that undergoes quiescence in response to CDK4 inhibition does not increase expression of at least one marker, or at least two markers, or at least three markers, of the senescent phenotype selected from the group consisting of SA-P-gal, senescence-associated heterochromatin foci and elaboration of the senescence-associated secretory program and/or does not increase the number of ATRX foci in the nucleus and/or exhibits stable or increased levels of MDM2 protein, relative to the level without treatment with the CDK4 inhibitor.

In certain non-limiting embodiments, the cells for use in the disclosed assay can include LS8107, LS7785-1, LS7785-10, LS8313, H358 and/or H3122 cells. In a non-limiting embodiment, the cells can be LS8107 cells. In a non-limiting embodiment, the cells can be LS8313 cells. In non-limiting embodiments, the cells can be cancer cells from a patient, or a population of cells cultured from cancer cells from a patient.

In certain non-limiting embodiments, the cell lines used in the present assay can be a cancer cell line that does not undergo senescence in response to treatment with the compound used as the first drug candidate. For example, and not by way of limitation, a cell line that does not undergo senescence can be a cell line that does not increase expression of at least one marker, or at least two markers, or at least three markers, of the senescent phenotype selected from the group consisting of SA-P-gal, senescence-associated heterochromatin foci and elaboration of the senescence-associated secretory program and/or does not increase the number of ATRX foci in the nucleus relative to the level observed in the absence of treatment with the first drug candidate.

After treatment with the first drug candidate, the one or more cells can subsequently be treated with a second drug candidate 102. Non-limiting examples of candidate compounds that can be used as the second drug candidate are disclosed above. In certain embodiments, the first drug candidate and the second drug candidate are different compounds. In certain embodiments, the cells can be treated with a second drug candidate at least one day, at least two days, at least three days, at least four days, at least five days, at least six days, at least seven days, at least eight days or more after treatment with the first drug candidate. In a specific non-limiting embodiment, the one or more cells can be treated with the second drug candidate two days after treatment with the first drug candidate. The amount of the second drug candidate applied to the cells depends on the type of compound used as the second drug candidate. In non-limiting embodiments, cells can be treated with a second drug candidate at a concentration of about 1 nM to about 1 M. For example, and not by way of limitation, the cells can be treated with a second drug candidate at a concentration from about 10 nM to about 100 μΜ, from about 100 nM to about 10 μΜ, from about 500 nM to about 10 uM, from about 750 nM to about 10 μΜ, from about 750 nM to about 5 uM or from about 750 nM to about 1 μΜ. In certain embodiments, the cells can be treated with a second drug candidate at a concentration of about 1 μΜ. Following treatment with the second drug candidate, the assay method can further include determining the number of ATRX foci per cell 103, where an increase in the number of ATRX foci per cell in response to treatment with the second drug candidate (e.g., as compared to a reference standard indicates that the second drug candidate may be useful when administered in combination with the first drug candidate during the treatment of a subject that has cancer. In non-limiting embodiments, determining the number of ATRX foci can be performed at least two days, at least three days, at least four days, at least five days, at least six days, at least seven days, at least eight days or more after treatment with the second drug candidate.

In non-limiting embodiments, an increase in the number of ATRX foci may be appreciated by comparing the number of ATRX foci per cell in the non- responder cells following treatment with the second drug candidate to a reference standard. In certain, non-limiting embodiments, the reference standard can include non-responsive cells that have been treated with the first drug candidate alone.

Alternatively or additionally, the reference standard can include non-responsive cells that have not been treated with a first drug candidate or a second drug candidate. In non-limiting embodiments, an increase in the percentage of cells that have a number of ATRX foci that increase about 50%, about 60%, about 70% or more per cell on average in response to treatment with both the first and second drug candidates as compared to a reference standard is indicative that the combination may be useful in treating subjects having cancer.

"In combination with" or "in conjunction with," as used interchangeably herein, means that the first drug candidate and the second drug candidate are administered to a subject as part of a treatment regimen or plan. In certain embodiments, being used in combination does not require that the first drug candidate and the second drug candidate are physically combined prior to

administration or that they be administered over the same time frame. For example, and not by way of limitation, the first drug candidate and the second drug candidate can be administered concurrently to the subject being treated, or can be administered at the same time or sequentially in any order or at different points in time.

5.2.2. ASSAY FOR IDENTIFYING COMPOUNDS THAT

PREVENT. MINIMIZE. INHIBIT AND/OR REDUCE SENESCENCE The present invention provides assays for identifying potential drug candidates that can be used to minimize, prevent, inhibit and/or reduce senescence in a cell line. In certain embodiments, the presently disclosed assays can be used to identify compounds that can prevent or inhibit a quiescent cell from transitioning to a senescent state and/or can result in a senescent cell to reenter the cell cycle and/or enter a quiescent state. The assays of the present invention comprise analyzing the number of ATRX foci per cell observed in the cell line following treatment with a drug candidate.

The present invention provides assays for identifying drug candidates that may be effective as therapeutic agents for treating diseases that are associated with senescence (i.e., "age-related diseases"). For example, and not by way of limitation, the drug candidates may be effective as therapeutic agents for treating age- related diseases. Non-limiting examples of such age-related diseases include heart disease, atherosclerosis, intervertebral disc degeneration, sarcopenia, emphysema, glomerular disease, chronic obstructive pulmonary disease (COPD), cataracts, macular degeneration, hypertension, glaucoma, skin aging, neurodegeneration, reduced organ function, e.g., liver, Alzheimer's, Parkinson's, arthritis, e.g., osteoarthritis, dementia and diabetes (e.g., Type 2 diabetes). Additional non-limiting examples of such diseases include graying of the hair, loss of muscle mass and loss of subcutaneous fat.

For purposes of illustration and not limitation, Figure 19 is a schematic representation of an exemplary assay for identifying potential drug candidates that can reduce, minimize, inhibit and/or prevent senescence according to the disclosed invention 200. In certain embodiments, the assay for identifying compounds that can inhibit, minimize, prevent, inhibit and/or reduce senescence can include treating one or more cells with a drug candidate in combination with a compound and/or modality that induces senescence in the cells (i.e., a "senescence-inducing compound") 201/202. For example, and not by way of limitation, the senescence-inducing compound is a compound or modality that can result in an increase in the expression of at least one marker, or at least two markers, or at least three markers, of the senescent phenotype selected from the group consisting of SA-P-gal, senescence- associated heterochromatin foci and elaboration of the senescence-associated secretory program and/or an increase the number of ATRX foci in the cell relative to the level observed in the absence of treatment with the compound or modality. In certain embodiments, the senescence-inducing modality can be gamma radiation.

In non-limiting embodiments, the cells can be treated with the senescence-inducing compound at least about 6 hours, about 8 hours, about 10 hours, about 12 hours, about 14 hours, about 16 hours, about one day, about two days or about three days after plating, e.g., within a microtiter plate, e.g., a 384-well microtiter plate (see Figure 7H). In certain embodiments, the cells can be treated with the senescence-inducing compound at about 12 hours after plating (see Figure 7H).

In certain embodiments, the senescence-inducing compound can be doxorubicin. In non-limiting embodiments, cells can be treated with doxorubicin at a concentration of about 1 nM to about 1 M. For example, and not by way of limitation, the cells can be treated with doxorubicin at a concentration from about 10 nM to about 100 uM, from about 100 nM to about 10 μΜ, from about 500 nM to about 10 μΜ, from about 750 nM to about 10 μΜ, from about 750 nM to about 5 μΜ or from about 750 nM to about 1 μΜ. In certain embodiments, the cells can be treated with doxorubicin at a concentration of about 100 nM. In certain embodiments, the cells can be treated with doxorubicin at a concentration of about 10 μΜ or 100 μΜ.

In certain embodiments, the senescence-inducing compound can be a CDK4 inhibitor. Non-limiting examples of CDK4 inhibitors include compounds that inhibit the kinase activity of CDK4. Additional non-limiting examples of CDK4 inhibitors include ATP-competitive inhibitors of CDK4. In particular non-limiting embodiments, the CDK4 inhibitor is derived from pyridopyrimidine or

indolocarbazole compounds. Further non-limiting examples of CDK4 inhibitors include Palbociclib Isethionate, LEE011, LY2835219, PD0332991, and Flavopiridol Hydrochloride. In certain embodiments, the CDK4 inhibitor is PD0332991.

Additional CDK4 inhibitors are disclosed in U.S. Patent Nos. 6,630,464 and

6,818,663, and U.S. Patent Application No. U.S. 2012/244, 110. Further non-limiting examples of CDK4 inhibitors include antisense oligonucleotides, shRNA molecules, and siRNA molecules that specifically inhibit the expression or activity of CDK4. One non-limiting example of a CDK4 inhibitor comprises an antisense, shRNA, or siRNA nucleic acid sequence homologous to at least a portion of a CDK4 nucleic acid sequence, wherein the homology of the portion relative to the CDK4 sequence is at least about 75 or at least about 80 or at least about 85 or at least about 90 or at least about 95 or at least about 98 percent, where percent homology can be determined by, for example, BLAST or FASTA software. In certain non-limiting embodiments, the complementary portion may constitute at least 10 nucleotides or at least 15 nucleotides or at least 20 nucleotides or at least 25 nucleotides or at least 30 nucleotides and the antisense nucleic acid, shRNA or siRNA molecules may be up to 15 or up to 20 or up to 25 or up to 30 or up to 35 or up to 40 or up to 45 or up to 50 or up to 75 or up to 100 nucleotides in length. Antisense, shRNA, or siRNA molecules may comprise DNA or atypical or non-naturally occurring residues, for example, but not limited to, phosphorothioate residues.

In certain embodiments, the senescence-inducing compound can be a modality that results in a reduction in MDM2 expression, e.g., antisense

oligonucleotides, shRNA molecules, and siRNA molecules that specifically inhibit the expression or activity of MDM2. In certain embodiments, the modality that results in a reduction in MDM2 expression can be an siRNA or shRNA that targets MDM2. In certain embodiments, the modality, e.g., shMDM2, can be under the control of tetracycline or its derivative doxycycline (e.g., tet-shMDM2), and the use of the modality includes contacting the cells with doxycycline, e.g., to induce expression of the shMDM2. In certain embodiments, the region of MDM2 targeted by the siRNA or shRNA comprises the nucleotide sequence

GCAAGGTGTTCAGATTGTATAA (SEQ ID NO: 1) or

ACAGGAACTTGGTAGTAGTCAA (SEQ ID NO:2).

In certain embodiments, the senescence-inducing compound can be a modality that results in a reduction in HRAS expression, e.g., antisense

oligonucleotides, shRNA molecules, and siRNA molecules that specifically inhibit the expression or activity of HRAS. In certain embodiments, the modality that results in a reduction in HRAS expression can be an siRNA or shRNA that targets HRAS. In certain embodiments, the modality, e.g., shHRAS, can be under the control of tetracycline or its derivative doxycycline (e.g. , tet-HRAS), and the use of the modality includes contacting the cells with doxycycline, e.g., to induce expression of the HRAS. In certain embodiments, the region of HRAS targeted by the siRNA or shRNA comprises the nucleotide sequence CGGAAGCAGGTGGTCATTGAT (SEQ ID NO:3) or GTGTGTGTTTGCCATCAACAA (SEQ ID NO:4). In certain embodiments, the senescence-inducing compound can be a compound that interferes with receptor tyrosine kinase activity, PI3K/AKT activity or MEK activity, which can affect the activity and/or function of HRAS. In non-limiting embodiments, the cells can be treated with the drug candidate and the senescence-inducing compound simultaneously. Alternatively, the cells can be treated with the drug candidate at least one day, at least two days, at least three days, at least four days, at least five days, at least six days, at least seven days, at least eight days or more after treatment with the senescence-inducing compound 202. In certain embodiments, the cells can be treated with the drug candidate 2 days after with the senescence-inducing compound (see Figure 7H). Additionally, the cells can be treated with the drug candidate at least one day, at least two days, at least three days, at least four days, at least five days, at least six days, at least seven days, at least eight days or more before treatment of the cells with the senescence-inducing compound.

In certain embodiments, the assay can further include determining the number of ATRX foci per cell following treatment with the drug candidate 203. In certain embodiments, if a decrease in the number of ATRX foci per cell is observed or if the number of ATRX foci per cell does not increase in response to treatment with the drug candidate in combination with the senescence-inducing compound then the drug candidate may be useful in preventing and/or inhibiting the induction of senescence or reducing senescence and/or be useful as a therapy for treating age- related diseases.

In non-limiting embodiments, cells for use in the disclosed assay can be a cell that when treated with an effective amount of a senescence-inducing compound increases expression of at least one marker, or at least two markers, or at least three markers, of the senescent phenotype selected from the group consisting of SA-P-gal, senescence-associated heterochromatin foci and elaboration of the senescence-associated secretory program and/or increases the number of ATRX foci in the nucleus. In certain embodiments, these cells are also referred to herein as "responder cells," as disclosed above. Non-limiting examples of such cells include LS8817, LS141, LS0082, A549, MCF7, SNB19 and H1975. In non-limiting embodiments, the mean level of nuclear ATRX foci increase per cell may be at least 30%.

In certain embodiments, the assay for identifying drug candidates that can minimize, prevent, inhibit and/or reduce senescence can comprise treating one or more cells that are in a senescent state, e.g., due to replicative senescence, with a drug candidate. In certain embodiments, senescence can be induced by the reduction in MDM2 expression in the cell, e.g., a responder cell and/or a non-responder cell. For example, and not by way of limitation, MDM2 expression can be reduced by the introduction of a modality that reduces MDM2 expression, e.g., a shRNA targeting MDM2 (i.e., shMDM2) or an siRNA targeting MDM2 (as noted above), into the cell, e.g., a LS8817 cell. In certain embodiments, the cell line can be the LS8313 cell line. In certain embodiments, the reduction in MDM2 expression can occur simultaneously with the treatment of the cells with the drug candidate. Alternatively, the cells can be treated with the drug candidate and the modality that reduces MDM2 expression at different timepoints. For example and not by way of limitation, the cells can be treated with the drug candidate at least one day, at least two days, at least three days, at least four days, at least five days, at least six days, at least seven days, at least eight days or more after the reduction in MDM2 expression. In certain non-limiting embodiments, the assay can further include determining the number of ATRX foci per cell following treatment with the drug candidate where a decrease in the number of ATRX foci per cell in response to treatment with the drug candidate indicates that the drug candidate may be useful in transitioning the cells from a senescent state to a quiescent state and/or to induce the cells to reenter the cell cycle. In certain embodiments, the assay can further include determining whether the one or more cells reenter the cell cycle by counting the total number of cells, e.g., by DAPI staining.

The presently disclosed invention further provides assays for preventing, inhibiting, minimizing and/or reducing geroconversion in cells. As discussed in the Examples section below, geroconversion is the transition from a quiescent state to a senescence state and is associated with an increase in the number of ATRX foci in each cell. For purposes of illustration and not limitation, Figure 20 is a schematic representation of an exemplary assay for identifying potential drug candidates that can reduce, minimize, inhibit and/or prevent geroconversion according to the disclosed invention. In certain embodiments, the assay for identifying drug candidates that can reduce, minimize, inhibit and/or prevent geroconversion, i.e., the transition from a quiescent state to a senescent state, can comprise treating one or more cells with a senescence-inducing compound, e.g., a CDK4 inhibitor, 301, where the cell line undergoes quiescence in response to treatment with the senescence- inducing compound but does not progress towards senescence as measured by accumulation of either SA-P-gal, senescence associated heterochromatic foci and/or elaboration of the senescence associated secretome. In certain non-limiting embodiments, a cell line that undergoes quiescence in response to treatment with the senescence-inducing compound does not increase expression of at least one marker, or at least two markers, or at least three markers, of the senescent phenotype selected from the group consisting of SA-P-gal, senescence-associated heterochromatin foci and elaboration of the senescence-associated secretory program and/or does not increase the number of ATRX foci in the nucleus and/or exhibits stable or increased levels of MDM2 protein, relative to the level without treatment with the compound. In certain embodiments, these cells are also referred to herein as "non-responder cells," as described above. Non-limiting examples of such a cell line include LS8107, LS7785-1, LS7785-10, LS8313, H358 and H3122.

In certain embodiments, the assay can further include reducing MDM2 expression in the one or more senescence-inducing compound-treated cells in the presence of a drug candidate 302. In non-limiting embodiments, the reduction in MDM2 expression can occur simultaneously with the treatment of the cells with the drug candidate. Alternatively, the cells can be treated with the drug candidate at least one day, at least two days, at least three days, at least four days, at least five days, at least six days, at least seven days, at least eight days or more after the reduction in MDM2 expression. In certain embodiments, the cell line can comprise a modality that reduces MDM2 expression, e.g., a shRNA targeting MDM2 (i.e., shMDM2) or an siRNA targeting MDM2. In certain embodiments, the modality, e.g., shMDM2, can be under the control of tetracycline or its derivative doxycycline (e.g., tet-shMDM2), and the expression of the modality includes contacting the cells with doxycycline, e.g., to induce expression of the shMDM2. In certain embodiments, the assay can further include determining the number of ATRX foci per cell following treatment with the drug candidate 303, where a decrease in the number of ATRX foci per cell or where no increase in the number of ATRX foci per cell is observed in response to treatment with the drug candidate in the presence of the modality that targets MDM2 expression, as compared to a reference standard, indicates that the drug candidate may be useful in preventing the induction of senescence or reducing senescence.

For purposes of illustration and not limitation, Figure 21 is a schematic representation of an exemplary assay for identifying potential drug candidates that can reduce, minimize, inhibit and/or prevent geroconversion according to the disclosed invention. In certain embodiments, the assay for identifying drug candidates that can reduce, minimize, inhibit and/or prevent geroconversion, can comprise treating one or more cells with a senescence-inducing compound, e.g., a CDK4 inhibitor, in the presence of exogenously-expressed MDM2, e.g., human MDM2, 401/402, where the cell line undergoes quiescence in response to treatment with the senescence-inducing compound and MDM2 expression but does not progress towards senescence. In certain embodiments, exogenous expression of MDM2 can be initiated after treatment of the cells with the senescence-inducing compound 402. In certain embodiments, the cell line can be a responder cell line. Non-limiting examples of such a cell line for use in this assay includes LS8817, LS141 and LS0082 cell lines, e.g., a responder cell line. Additional non-limiting examples of cell lines that can be used in this assay include the glioma cell line SNB19 and the non-small cell lung cancer cell line

HI 975. In certain embodiments, the cell line can comprise a modality that increases MDM2 expression, e.g., a construct that comprises full-length MDM2. In certain embodiments, the exogenously-expressed MDM2 can be tagged, for example, with a Flag-tag, a Myc-tag or an HA-tag. In certain embodiments, the modality can be under the control of tetracycline or its derivative doxycycline (e.g. , tet-MDM2), and the expression of MDM2 is obtained by contacting the cells with doxycycline. In certain embodiments, the assay can further include the removal of doxycycline from the cells and the treatment of the cells with a drug candidate 403. In non-limiting

embodiments, the removal of doxycycline can occur simultaneously with the treatment of the cells with the drug candidate. Alternatively, the cells can be treated with the drug candidate at least one day, at least two days, at least three days, at least four days, at least five days, at least six days, at least seven days, at least eight days or more after the removal of doxycycline. In certain embodiments, the cells can be treated with the drug candidate at least one day, at least two days, at least three days, at least four days, at least five days, at least six days, at least seven days, at least eight days or more prior to removal of doxycycline from the cells.

In certain embodiments, the assay can further include determining the number of ATRX foci per cell following treatment with the drug candidate and the removal of doxycycline 404, where a decrease in the number of ATRX foci per cell or where no increase in the number of ATRX foci per cell is observed in response to treatment with the drug candidate in the absence of doxycycline, as compared to a reference standard, indicates that the drug candidate may be useful in preventing and/or inhibiting the induction of senescence and/or reducing senescence (e.g., reducing, blocking and/or inhibiting geroconversion). In non-limiting embodiments, determining the number of ATRX foci can be performed within at least two days, at least three days, at least four days, at least five days, at least six days, at least seven days, at least eight days or more after treatment with the drug candidate. In non- limiting embodiments, a decrease in the number of ATRX foci or the lack of an increase in the number of ATRX foci may be appreciated by comparing the number of ATRX foci per cell in the cells following treatment with the drug candidate to a reference standard. In certain, non-limiting embodiments, the reference standard can include cells that have been treated with the senescence-inducing compound alone. Alternatively or additionally, the reference standard can include senescence cells that have not been treated with the drug candidate or senescence-inducing compound. In certain embodiments, the reference standard can include cells that have been treated with a senescence-inducing compound followed by treatment with a modality that reduces MDM2 expression. In certain embodiments, the reference standard can include cells that have been treated with a senescence-inducing compound in the presence of exogenously expressed MDM2. In non-limiting embodiments, a decrease in the percentage of cells or no increase in the percentage of cells that have an increased number of ATRX foci per cell in response to treatment with the drug candidate as compared to a reference standard is indicative that the drug candidate may be useful in treating subjects having an age-related disease. In certain

embodiments, an increase in the average number of ATRX foci per cell of less than about 50%, less than about 45%, less than about 40%, less than about 35%, less than about 30%), less than about 25%, less than about 20%, less than about 15%, less than about 10%), less than about 5%, less than about 2% or less than about 1% upon treatment with the drug candidate, e.g., as compared to a reference standard, is indicative that the drug candidate may be useful in treating subjects having an age- related disease.

The amount of the drug candidate that is applied to the cells can depend on the type of compound used as the drug candidate and the number of cells being treated. Non-limiting examples of drug candidates or candidate compound libraries that can be analyzed using the disclosed assays are discussed above. In certain non-limiting embodiments, cells can be treated with a drug candidate at a concentration of about 1 nM to about 1 M. For example, and not by way of limitation, the cells can be treated with the drug candidate at a concentration from about 10 nM to about 100 uM, from about 100 nM to about 10 μΜ, from about 500 nM to about 10 μΜ, from about 750 nM to about 10 μΜ, from about 750 nM to about 5 μΜ or from about 750 nM to about 1 μΜ. In certain embodiments, the senescence- inducing compound can be applied to the cells at a concentration of about 10 μΜ.

The amount of the senescence-inducing compound that is applied to the cells can depend on the type of compound used as the senescence-inducing compound and the number of cells being treated. Non-limiting examples of senescence-inducing compounds that can be used in the disclosed assays are discussed above. In certain non-limiting embodiments, cells can be treated with a senescence-inducing compound at a concentration of about 1 nM to about 1 M. For example, and not by way of limitation, the cells can be treated with the senescence- inducing compound at a concentration from about 10 nM to about 100 μΜ, from about 100 nM to about 10 μΜ, from about 500 nM to about 10 μΜ, from about 750 nM to about 10 μΜ, from about 750 nM to about 5 μΜ or from about 750 nM to about 1 μΜ. In certain embodiments, the senescence-inducing compound can be applied to the cells at a concentration of about 1 μΜ. In non-limiting embodiments, cells can be treated with a CDK4 inhibitor, e.g., PD0332991, at a concentration of about 1 nM to about 1 M. For example, and not by way of limitation, the cells can be treated with a CDK4 inhibitor at a concentration from about 10 nM to about 100 μΜ, from about 100 nM to about 10 μΜ, from about 500 nM to about 10 μΜ, from about 750 nM to about 10 μΜ, from about 750 nM to about 5 μΜ or from about 750 nM to about 1 μΜ. In certain embodiments, the cells can be treated with a CDK4 inhibitor at a concentration of about 1 μΜ.

5.3 KITS

In non-limiting embodiments, the present invention provides for a kit for performing the assays of the present invention. For example, and not by way of limitation, the present invention provides for a kit for determining whether a drug combination induces senescence and/or whether a drug candidate prevents, minimizes, inhibits and/or reduces senescence and/or geroconversion in a cell, comprising a means for detecting ATRX foci. Methods for measuring ATRX foci are described in the sections above.

Types of kits include, but are not limited to, arrays/microarrays, ATRX-specific, antibodies or other detection reagents for detecting ATRX foci. In non-limiting embodiments, a kit may comprise at least one antibody for immunodetecting ATRX foci. In one specific non-limiting embodiment, a kit may comprise a probe or antibody suitable for detecting ATRX protein present within the foci. Antibodies, both polyclonal and monoclonal, including molecules comprising an antibody variable region or a subregion thereof, specific for an ATRX protein, may be prepared using conventional immunization techniques, as will be generally known to those of skill in the art. In non-limiting embodiments, a kit of the present invention can comprise the ATRX-specific antibody sold by Bethyl, Catalog No. A301 -045 A ("the '045 Ab"), a fragment thereof, or an antibody that competitively inhibits binding of the '045 Ab to ATRX, for detecting ATRX foci. In certain embodiments, a kit of the present invention can comprise the ATRX-specific antibody sold by Santa Cruz Biotechnology, Catalog No. sc-55584 (the "D5 antibody"), a fragment thereof, or an antibody that competitively inhibits binding of the D5 antibody to ATRX, for detecting ATRX foci.

The immunodetection reagents of the kit may include detectable labels that are associated with, or linked to, the given antibody or antigen itself. Such detectable labels include, for example, fluorescent molecules (rhodamine, fluorescein, green fluorescent protein, luciferase, Cy3, Cy5, or ROX), radiolabels (3H, 35S, 32P, 14C, 1311) or enzymes (alkaline phosphatase, horseradish peroxidase). Alternatively, a detectable moiety may be comprised in a secondary antibody or antibody fragment, which selectively binds to the first antibody or antibody fragment (where said first antibody or antibody fragment specifically recognizes ATRX).

In certain non-limiting embodiments, a kit may comprise one or more detection reagents and other components (e.g., a buffer, enzymes such as alkaline phosphatase, antibodies, and the like) necessary to carry out an assay or reaction to determine the number of ATRX foci per cell.

In a further non-limiting embodiment, the kit may further include one or more cells for performing the disclosed assays. Non-limiting examples of cells for use in the disclosed assays are described above. For example, and not by way of limitation, the kit can include responder and/or non-responder cells, as disclosed above. In certain embodiments, the kit can include LS8107, LS8313, LS8817, SNB 19 and/or HI 975 cells.

In certain embodiments, the kit can include LS8107 cells or LS8313. In certain embodiments, the kit can include LS8107 or LS8313 cells that comprise a vector comprising a shMDM2 under the control of doxycycline. In certain

embodiments, the kit can include LS8817 cells. In certain embodiments, the kit can include LS8313 cells.

In certain embodiments, the kit can include LS8817, S B 19 and/or HI 975 cells. In certain embodiments, the kit can include LS8817, S B 19 and/or HI 975 cells that comprise a vector comprising a shMDM2 under the control of doxycycline. In certain embodiments, the kit can include LS8817 cells that comprise a vector comprising MDM2 (e.g., full-length MDM2) under the control of

doxycycline. The cells within the kit can be supplied as a cell suspension or as a frozen cell sample. In non-limiting embodiments, the cells can be provided in a multiwell plate, a microtiter plate or in an array.

A kit may further include instructions for using the kit to determine the number of ATRX foci. In non-limiting embodiments, the instructions describe that an increase in the number of ATRX foci per cell upon treatment with the drug

combination is indicative that the drug combination may be useful as a combinatorial drug therapy for cancer. Alternatively or additionally, in certain non-limiting embodiments, the instructions can describe that a decrease or the absence of an increase in the number of ATRX foci per cell upon treatment with a drug candidate is indicative that the drug candidate may be useful as a therapy for treating age-related diseases.

6. EXAMPLE 1 : ATRX FOCI AS A MARKER FOR SENESCENCE

6.1 MATERIALS AND METHODS

Senescence analyses. Cells were plated at a concentration of 25,000 per well in a 4-well chamber slides (Lab-Tek) and treated for seven days with drug and stained for senescence-associated β-galactosidase (Cell Signaling kit #9860). Cell number was quantitated by DAPI staining and β-galactosidase staining quantitated as a proportion of total cells.

Immunofluorescence. Cells were washed twice with PBS and fixed with 4% paraformaldehyde at room temperature for 20 minutes. Cells were again washed twice with PBS and treated with 0.5% triton X-100 in PBS. Cells were again washed twice with PBS and blocked in 5% BSA in PBS for 30 minutes at room temperature. Cells were incubated with anti-ATRX antibody, obtained from Bethyl, Cat No. A301-045A, diluted at 1 :2000 in PBS (stock is provided at 1 mg/ml) overnight at 4 degrees Celsius. The following day, cells were washed twice with PBS and incubated with Alexa Fluor-488 anti-rabbit secondary diluted at 1 :500 in PBS (stock was provided at 2 mg/ml) for one hour at room temperature. Cells were washed twice with PBS and incubated with 0.1 μg/ml DAPI for 5 minutes at room temperature. Cells were washed three times with PBS and slides were mounted in Vectashield fluorescence mounting media. Slides were imaged using a Zeiss

Axioplan 2 Upright Microscope.

Immunoblot. Antibodies against p53 (Bp53-12), were obtained from Santa Cruz Biotechnology, and the ATRX antibody, Cat No. A301 -045 A was obtained from Bethyl. Treated cells were lysed with buffer composed of 50mM Tris- HC1, pH7.4, 250mM NaCl, 5mM EDTA, 0.5% NP40, 2mM PMSF, and supplemented with protease inhibitors. Eighty micrograms of protein were resolved by SDS-PAGE and transferred to PVDF membranes. Membranes were incubated overnight with antibodies (1 : 1000).

6.2 RESULTS

Senescence is perceived as a favorable clinical outcome due to its ability to inhibit tumor progression. ATRX plays a role in senescence and has been shown to interact with PML, macroH2A, FIP1 and hi stone H3.3 {see Eustermann et al., 2011; Lewis et al., 2010; Ratnakumar et al., 2012; Xue et al., 2003), which have been show to interact with HIRA/ASF. Immunofluorescence analysis using the Bethyl antibody was performed to determine if ATRX was recruited to foci. As shown in Figure 2, ATRX is required for senescence induced by the inhibition of CDK4. In U20S cells that do not express endogenous ATRX, inhibition of CDK4 using the CDK4 inhibitor, PD0332991, resulted in the cells entering a quiescent state versus a senescent state (Figure 2). Expression of exogenous ATRX in U20S cells, in the presence of the CDK4 inhibitor, resulted in the cells entering a senescent state as determined by the accumulation of the senescence marker, perinuclear associated β- galactosidase (SA-P-gal). The exogenous expression of a mutant form of ATRX in the U20S cells did not modulate the response to CDK4 inhibition.

As shown in Figure 3, ATRX foci were observed upon treatment with doxorubicin. LS8817 cells expressing either a scrambled (shSCR) or ATRX shRNA (shATRX) were treated with 100 μΜ doxorubicin for five days. In doxorubicin treated cells, the number of ATRX foci per cell significantly increased compared to cells that were not treated with doxorubicin (Figure 3). As shown in Figure 3, the average number of ATRX foci per cell in doxorubicin-treated cells was about 35; whereas, the average number of ATRX foci per cell in the control cells was about 8. In addition, ATRX foci were not observed in cells expressing shATRX (Figure 3).

To determine if doxorubicin induced senescence in the liposarcoma cell line, LS8817, the accumulation of SA-P-gal was measured following treatment with doxorubicin (Figure 4). As shown in Figure 4, LS8817 cells expressing either a shSCR or shATRX were treated with 100 μΜ doxorubicin for five days and the effect on the accumulation of SA-P-gal was determined (bottom). SA-P-gal staining significantly accumulated after doxorubicin treatment in cells expressing shSCR. In contrast, SA-P-gal staining did not significantly accumulate in shATRX-expressing cells after doxorubicin treatment (Figure 4).

6.3 DISCUSSION

Without being bound to a particular theory, these results indicate that senescence, regardless of the nature of the inducer, is correlated with an increase in the number of ATRX foci in each cell, and disruption of ATRX prevents senescence. Therefore, ATRX is generally required for senescence, and this is associated with its recruitment to chromatin-associated foci.

7. EXAMPLE 2: ATRX FOCI AS A MARKER FOR REPLICATIVE

SENESCENCE AND GEROCONVERSION

To determine whether ATRX foci are observed upon induction of senescence in the untransformed cell line, WI-38, the cells were continually passaged until passage number 25 to induce replicative senescence (Figure 5). As the cells were continually passaged to passage 25, cell proliferation was observed to have greatly reduced as determined by BrdU staining (Figure 5A and E). In addition, the expression of p53, which has been previously shown to increase in expression in proliferating cells, was greatly reduced in cells at passage 25 (Figure 5A and F).

To determine if continual passaging of the WI-38 cell line induced senescence, the accumulation of SA-P-gal was measured during passage 11 and passage 25. As shown in Figure 5B and G, SA-P-gal staining significantly accumulated after passage 25 as compared to passage 11, suggesting that the cells under senescence during continual passaging. In addition, the number of ATRX foci per cell significantly increased at passage 25 as compared to cells at passage 11 (Figure 5C and H). These results show that replicative senescence is associated with an increase in ATRX foci, and ATRX foci can be a marker for normal, untransformed cells undergoing senescence.

In addition, an increase in the number of ATRX foci was observed during geroconversion, which is the transition from a quiescent state to a senescence state. As shown in Figure 6, LS8107 cells, a cancer cell line that does not undergo senescence in response to treatment with the CDK4 inhibitor, PD0332991, entered a quiescent state in the presence of PD032991. Following treatment with PD032991, the cells were than treated with a shRNA directed to MDM2 to decrease MDM2 expression (Figure 6A and 8D). The loss of MDM2 expression following CDK4 inhibition resulted in the cells transitioning from a quiescent state to a senescent state as indicated by the accumulation of SA-P-gal staining (Figure 6B and 8C, E) and resulted in an increase in the number of ATRX foci per cell as compared to the cells treated with PD0332991 alone (Figure 6C and 8C, E). To ensure that geroconversion can occur in these cells, LS8107 tet-on-shMDM2 cells were treated with 1 μΜ

PD0332991 for 7 days, which did not induce senescence, and with 1 μΜ PD0332991 for 2 days followed by 10 μΜ doxycycline for 5 days, which did induce senescence as measured by an increase in ATRX foci per cell and SA-P-Gal activity (Figure 8E). Without being bound to a particular theory, these results indicate that geroconversion is correlated with an increase in the number of ATRX foci in each cell.

In the liposarcoma cell line, LS8817, the CDK4 inhibitor, PD0332991, at a concentration of 1 μΜ or 0.1 μΜ, resulted in the cells entering a senescent state. This senescent state correlated with a significant increase in the average number of ATRX foci per cell (Figure 7B and H). The exogenous expression of MDM2 by the use of a tet-on-MDM2-Flag construct within these CDK4 inhibitor-treated cells in the presence of doxycycline resulted in the cells to enter a quiescent state and prevented them from undergoing senescence (Figure 7A, C and D). The removal of doxycycline allowed the cells to enter senescence and resulted in an increase in the average number of ATRX foci per cell (Figure 7A and 7C). In addition, as shown in Figure 7C, the removal of doxycycline to reduce exogenous MDM2 expression allowed the cells to enter a senescent state as indicated by the accumulation of SA-P-gal staining. As shown in Figure 7D, upon addition of doxycycline and PD0332991, the cells no longer proliferate as indicated by the reduction in BrdU incorporation in doxycycline- treated cells. As shown in Figure 7E, the removal of doxycycline results in an increase in SA-P-gal accumulation and ATRX foci as a function of time. In addition, different phenotypes can be measured in the synchronous system (Figure 7F) giving rise to the idea that particular states exist along the path to geroconversion. As shown in Figure 7F, LS8817 cells treated with CDK4 inhibitors exit the cell cycle (and exhibit low BrdU incorporation, low phosphorylated Rb levels and low cyclin A levels). A day or two later, MDM2 expression decreases and the number of ATRX foci increase, and three to five days later, the cells gain the hallmarks of senescence that include SA-P-gal, SAHF and long term irreversible growth arrest. These data indicate that such a system can be used to identify compounds that restrict senescence entry.

The expression of exogenous MDM2 in the presence of a CDK4 inhibitor to modulate senescence entry can be used in other cell lines. As shown in Figure 7G, this system can be applied to other cell lines, including the glioma cell line S B19 and the non-small cell lung cancer cell line HI 975. The removal of doxycycline to reduce exogenous MDM2 expression in the presence of the CDK4 inhibitor, PD0332991, allowed the S B19 and H1975 cells to enter senescence as indicated by the increase in the number of ATRX foci observed in such cells (Figure 7G).

To determine if the reduction in MDM2 alone can result in the induction of senescence, LS8817 cells, LS8107 cells or LS8313 cells, which included a tet-on-shMDM2 construct, were treated with doxycycline to express shMDM2 (Figure 8 A). The targets of the shMDM2 constructs are shown in Figure 8B. As shown in Figure 8B, the reduction in the expression of MDM2 alone in LS8817 cells resulted in the cells entering a senescent state. The number of ATRX foci also increased upon the reduction in MDM2 expression (Figure 8B). In addition, and as shown in Figure 8F, the reduction of MDM2 expression alone in LS8313 cells was sufficient to induce the cells to enter a senescent state as indicated by the increase in the number of ATRX foci observed per cell as compared to the control.

The use of a cell-based system, disclosed herein, to identify compounds that block senescence is shown in Figure 8G. As shown in Figure 8G, the responsive cell line, LS8817, was treated with PD0332991 in the presence of a putative senescence inhibitor (also referred to herein as a "senosuppresor"). Data was captured on an In-Cell Analyzer 6000, and dose dependent activity and the number of cells detected with DAPI (relative to PD0332991 alone) is shown. As shown in Figure 8G, the addition of the putative senosuppressor in the presence of PD0332991 resulted in a reduction in the number of ATRX foci as compared to the control. These data confirm that such an assay to identify compounds that block senescence.

To use such as cell-based system in a high throughput assay, a number of parameters can be optimized (Figure 7H). For example, in the LS8817 cell line, cells can be plated into a 384 well plate and then treated with 1 μΜ PD0332991 12 hours later. The number of ATRX foci per cell increased when measured 5 days later (Figure 7H(i) and (ii)). In addition, cells do not have to be treated with 1 μΜ

PD0332991 multiple times within the 7 day time period in order to get an increase in ATRX foci formation, so cells can be treated only on day 1 in order to simplify automation of the screen (Figure 7H(iii)). The concentration of antibody that will be used to identify ATRX foci can be 0.5 μg/mL (Figure 7H(iv)). The imaging of slides and counting of ATRX foci also do not have to be performed manually, as the In-Cell Analyzer 6000 can be programmed to obtain similar fold changes in ATRX foci upon PD0332991 treatment as compared to control cells (Figure 7H(v)).

8. EXAMPLE 3 : ATRX REGULATES SENESCENCE

8.1 MATERIALS AND METHODS

Cell culture. The cell lines were previously described (Kovatcheva et al. (2015)) or obtained from the ATCC. All cell lines were maintained in DME HG supplemented with 10% heat-inactivated fetal bovine serum (unless otherwise indicated) and 2mM L-glutamine. Differentiation of LS8817 cells was performed using the Gimble protocol as previously described (Halvorsen et al.

Thiazolidinediones and glucocorticoids synergistically induce differentiation of human adipose tissue stromal cells: biochemical, cellular, and molecular analysis. Metabolism 50, 407-413 (2001)). HDFs were split 1 :3 every time they reached confluence and were allowed to grow until replicative exhaustion. For irradiation, cells were exposed to lOGy and then allowed to recover for 10 days prior to analysis.

Gene targeting by shRNA. shRNA were delivered in the pLKO.1 vector (Open Biosystems) and infected cells selected using puromycin (^g/ml); infection with a virus carrying a scramble control

(CAACAAGATGAAGAGCACCAA (SEQ ID NO: 9)) was used as a control in all experiments utilizing shRNA. shRNA for ATRX and MDM2 were previously characterized (Kovatcheva et al. (2015)). Lentiviral vectors were prepared for the indicated genes as follows: HRAS (265, CGGAAGCAGGTGGTCATTGAT (SEQ ID NO: 5); 266, GTGTGTGTTTGCCATCAACAA (SEQ ID NO: 6)), ZNF107 (660, CGATTCTCAAACCTAACTATA (SEQ ID NO: 7); 662,

CAGGAAACAAACACTTCAAAT (SEQ ID NO: 8)).

Generation of ATRX mutants. ATRX was mutated using standard

QuikChange procedures and PLATINUM® Taq DNA Polymerase, High Fidelity (Therm oFisher) according to manufacturer's protocols. PCR products were then digested with Dpnl (New England Biolabs) according to manufacturer's protocols before being transfected into ELECTROMAX™ STBL4™ Competent Cells (Therm oFisher) according to manufacturer's protocols.

Mutational analysis in U20S cell lines. ATRX was delivered using a construct generously provided by David Picketts. U20S cells were transfected using siLentFect Lipid Reagent (BioRad) according to manufacturer's protocols. 48 hours following transfection, cells were selected using G418 (500μg/ml). Selected cells were sorted using fluorescent activated cell sorting on a MOFLOW™ sorter

(Beckman Coulter). Untransfected U20S cells were used to mark the GFP-negative population and the GFP-low population was collected. To verify the integrity of each mutant DNA was extracted from the transfected cells (QIAGEN DNEasy) and specific regions of ATRX indicated in Figure 11 A were amplified by PLATINUM® Taq DNA Polymerase, High Fidelity (ThermoFisher) {see Table 1 for primers) and sequenced.

Chromatin immunoprecipitation followed by sequencing (ChlP- seq). ATRX ChIP was performed as previously described (Law et al. ATR-X syndrome protein targets tandem repeats and influences allele-specific expression in a size-dependent manner. Cell 143, 367-378 (2010)). Barcoded Illumina libraries were generated using the Kapa Hyper Prep Kit (Kapa Biosystems, Wilmington,

Massachusetts) according to the manufacturer's instructions, with 12 cycles of PCR amplification. Libraries were pooled and run on an Illumina HiSeq 2500, v4 chemistry, to obtain 30-40 million single read, 50 nucleotide-long reads passing filter. Reads were adapter and quality trimmed using Trim Galore!

(http://www.bioinformatics.babraham.ac.uk/projects/trim_galore), aligned with BWA mem vO.7.8 (Li and Durbin. Fast and accurate short read alignment with Burrows- Wheeler transform. Bioinformatics 25, 1754-1760 (2009)), and realigned around indels and base quality score recalibrated with the Genome Analysis Toolkit v3.1.1 (McKenna et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next generation DNA sequencing data. Genome research 20, 1297-1303 (2010)). This was repeated twice with biologic replicates at different times.

MACS v2.0.10 (Zhang et al. Model-based analysis of ChlP-Seq (MACS). Genome biology 9, R137 (2008)) was used to call peaks within the irreproducible discovery rate (DDR) framework

(https://www. encodeproj ect. org/software/idr/ ;

https://sites.google.com/site/anshulkundaje/projects/idr). Peaks were annotated with the most proximal upstream and downstream genes (hgl9 RefSeq genes) filtered for technical false positives and intergenic sequences, and then overlapped with -5kb to +lkb promoters. Known telomeric, centromeric, and repetitive sequences were also annotated within the peaks. ChlPseq data was deposited on the Gene Expression Omnibus (GEO, http://www.ncbi.nlm.nih.gov/geo/).

RNA sequencing. RNA was extracted from cells treated as described using standard protocols (RNEasy, QIAGEN). All RNA-seq experiments were performed in triplicate (all from biologic replicates at different times). RNA quality was checked on a BioAnalyzer to ensure a minimum RNA Integrity Value (RIN) of 7. Libraries were generated using 500ng of input RNA per sample according to the manufacturer's instructions for TruSeq mRNA Library Prep Kit V2 (Illumina) with 8 cycles of PCR. Libraries were pooled and run on an Illumina HiSeq 2500, high output, to obtain 30 million paired end, 50 nucleotide-long reads. The RNA-Seq reads were aligned to the human reference sequence hgl9 with the RNASeq aligner STAR (version_2.4.0c). Genes annotated in Gencode version 18 were quantified with featureCounts (subhead package version 1.4.3-pl). The raw counts were then subjected to the Bioconductor package DESeq2 to call for differential expression between the groups of samples. Enrichment of differential expression in sets of genes was determined using Gene Set Enrichment Analysis (GSEA) on gene groupings from MSigDB, as well as custom sets (Subramanian et al, 2005). RNA-seq data was deposited on the Gene Expression Omnibus (GEO,

http://www.nchi .nlm .nih.gov/geo/).

RNA-sequencing gene list derivation. RNA sequencing data analysis and comparisons were performed with Partek Software. The gene lists included all genes that showed at least a 1.8-fold change (FDR < 0.05) when comparing control and 7 day PD0332991 treated samples. GO profiling, transcription factor profiling and GSEA. Gene lists were analyzed using the publicly available Enrichr software

(http://amp.pharm.mssm.edu/Enrichr/) (Chen et al. Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool. BMC Bioinformatics 14, 128 (2013)) and the top four most significantly enriched groups (based on p-value) were reported. For GSEA, gene lists were derived from the RNA-seq data and were compared against gene lists in the publicly available Molecular Signatures Database (MSigDB) v5.1. The specific gene sets analyzed were KO DO EZH2 T ARGET S (M5301), V$E2F4DP1_01 (M10526) and DNA REPAIR (M18229). GSEA statistical analysis was carried out with publicly available software from the Broad Institute (http://www.broadinstitute.org/gsea/index.jsp) (Subramanian et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proceedings of the National Academy of Sciences of the United States of America 102, 15545-15550 (2005)).

Real time quantitative qPCR (RT-qPCR). cDNA was synthesized from ^g of each RNA sample (extracted as above) using the One Taq® RT-PCR Kit and olio-dT primers (New England BioLabs). cDNA was diluted 1 :5 and Ιμΐ of reaction was used for qPCR using 400nM of each forward and reverse primers and SYBR® Green PCR Master Mix (Life Technologies) according to manufacturer's protocols. qPCR was performed on ViiaTM 7 Real-Time PCR System (Thermo

Scientific). All primers are listed in Table 1. All RT-qPCR comparisons are by two- sided t test.

Antibodies. Antibodies against MDM2 (SMP14), total Rb (IF8), cyclin A (H432), pl6 (C20), p53 (DO-1 and Bp53-12), tubulin (C20), ATRX (H-300) and PML (PG-M3) were obtained from Santa Cruz Biotechnology, phospho-Rb 780 (#9307) from Cell Signaling, ΗΡΙγ (05-690) and γΗ2ΑΧ (05-636) from Millipore, ATRX (A301-045A) from Bethyl Laboratories, 53BP1 (abl72580) and LC3

(ab48394) from Abeam, GFP (Al 1122) from Life Technologies. Dilutions for immunofluorescence were as follows: ATRX (A301-045A) 1 :2000; ΗΡΙγ (05-690) 1 :5000; PML (PG-M3) 1 : 1000; γΗ2ΑΧ (05-636) 1 :200; 53BP1 (abl72580) 1 :2000; GFP (Al l 122) 1 : 1000.

Senescence analyses. Senescence associated-P-galactosidase, senescence associated heterochromatic foci formation and clonogenic growth assays were performed as previously described (Kovatcheva et al. (2015)). All senescence comparisons are by two-sided t test.

8.2 RESULTS

ATRX regulates senescence. Cellular senescence is triggered by different stresses that irreversibly prevent cells from further proliferation. It can occur in non-transformed cells in which oncogenic drivers induce hyperproliferation or telomere shortening during replicative passaging or in response to unresolved DNA damage and persistent chronic activation of the DNA damage response (reviewed by d'Adda di Fagagna. Living on a break: cellular senescence as a DNA-damage response. Nature reviews Cancer 8, 512-522 (2008)). It can also occur in transformed cells in which proliferation is suppressed, for example by unresolved DNA damage or CDK4 inhibition (reviewed by Acosta and Gil. Senescence: a new weapon for cancer therapy. Trends in cell biology 22, 211-219 (2012); Salama et al. Cellular senescence and its effector programs. Genes & development 28, 99-114 (2014); Sherr et al.

Targeting CDK4 and CDK6: From Discovery to Therapy. Cancer Discov. (2015)).

Knocking down ATRX in human diploid fibroblasts induces their death. While a few cells may survive, these have gross defects in telomere

maintenance. Such defects obfuscate any interpretations about the effect of eliminating ATRX on senescence in primary cells; however, ATRX deficiency can be induced in transformed cells with little replicative or telomere issues, at least during the time it takes to assess senescence. In transformed cells, ATRX was required for DNA damage induced senescence.

Reducing ATRX in LS8817 cells attenuated the doxorubicin-induced increase in the number of SA-P-gal (Figure 9A) or SAHF positive cells (Figure 9B), and attenuated the expression of the cytokine genes that define the SASP in these cells (Figure 9C). Knocking down ATRX did not lead to DNA damage induced apoptosis (negative data not shown), nor did it affect the doxorubicin-induced accumulation of p53 (Figure 9D), or the accumulation of 53BP1 or γΗ2Αχ foci (Figure 9E), markers of DNA damage. The specificity of the lentiviral vectors encoding ATRX shRNA hairpins were previously validated (Kovatcheva et al. (2015)). Similar results were also obtained in another cell line, LS0082 (Figure 10).

Loss of MDM2 is another inducer of senescence in transformed cells (Kovatcheva et al. (2015)). ATRX was also required for senescence induced by MDM2 knockdown. Reducing ATRX in LS8817 cells (Figure 9F) or LS8313 cells (Figure 9G) attenuated their ability to accumulate SA-P-gal or SAFIF positive cells when MDM2 levels were decreased. The SAHF phenotype is weak in LS8313 cells and thus was not examined. Nevertheless, the cells that failed to senesce underwent quiescence.

ATRX was also required for CDK4 inhibition induced senescence. U20S cells are Rbl 10 positive and have a deletion encompassing exons 2-19 at the ATRX loci. After addition of CDK4i these cells exit the cell cycle with reduced expression of cyclin A and phosphorylated Rb (Figure 9H). There was a small increase in the number of SA-P-gal and SAHF positive cells; however, the number of these cells was greatly increased if ATRX expression was forced from a stably transfected vector (Figures 91 and 9J). The SASP phenotype was weak in the U20S cells; nevertheless, these cells were reduced in their capacity to re-enter the cell cycle after removing the CDK4i (Figure 9K), indicative of senescence. Forced ATRX expression did not significantly alter the proliferation of these cells during

unperturbed culture, nor did it affect the viability of the cells after CDK4i induced cell cycle exit. Thus, ATRX is important for multiple types of therapy induced

senescence (TIS), regardless of whether they act in a p53-dependent (DNA damage) or p53/INK4-independent (CDK4 inhibition) manner.

The number of ATRX nuclear foci increases in senescent cells. ATRX is a SWI/SNF family helicase. The ADD domain and LxVxL domain facilitate chromatin association by fostering interactions with the H3K4meO and H3K9me3 histones and HP1, respectively (Eustermann et al. Combinatorial readout of histone H3 modifications specifies localization of ATRX to heterochromatin. Nature structural & molecular biology 18, 777-782 (2011)). A hypomorphic K1650N mutation in the helicase domain, or a mutation that disrupts the interaction of ATRX with H3K9me3 histone (C240G), or a mutation that disrupts the interaction of ATRX with HP1 (V588E) (Eustermann et al. (2011)) were each compromised in their ability to support accumulation of SA-P-gal positive cells when CDK4 inhibitors were added. However, these cells still quiesced. In contrast, the E218A mutation that disrupts the interaction of ATRX with H3K4meO histone (Eustermann et al. (2011)), and a conservative missense mutation in the helicase domain identified in a liposarcoma patient (LI 612V) did not affect the ability of ATRX to support CDK4i induced accumulation of SA-P-gal (Figure 1 IB). The expression of the mutants did not correlate with their ability to promote the accumulation of SA-P-gal (Figure 11C). Thus, the interaction of ATRX with H3K9me3 histones and the HP1 family of proteins as well as the helicase activity were necessary for ATRX to promote senescence following inhibition of CDK4.

Given the requirements for H3K9me3 histone and HP1 binding sites and the fact that both of these modifications are enriched in SAHF, whether ATRX formed nuclear foci and colocalized with the SAHF were analyzed. There was a reproducible and dramatic increase in the number of ATRX foci formed within two days after CDK4 inhibitors were added to LS8817. The number of foci continued to increase over the next 2 to 5 days (Figure 1 ID). Approximately three-quarters of these foci overlapped with ΤΙΡΙγ foci in senescent cells (Figures 1 IE, 1 IF), suggesting that ATRX might participate in formation or maintenance of the SAHF and may have additional transcriptional regulatory activities as well. Similarly, there was an increase in the number of foci in two other liposarcoma cell lines (LS141, LS0082), a breast cancer cell line (MCF7), a glioma cell line (SNB19) and two lung cancer cell lines (A549 and HI 975) that underwent CDK4i-induced senescence (Figure 11G and Figure 12). An increase in the LS8817 cell line following MDM2 knockdown (Figure 11H) or after treatment with doxorubicin (Figure 1 II) was observed. Finally, the formation of nuclear foci was correlated with the ability of ATRX to promote senescence in U20S cells following treatment with the CDK4i (Figure 11 J). Only the K1650N helicase mutant was able to form foci but did not support senescence, perhaps reflecting the importance of the chromatin remodeling activity of this enzyme.

Although whether ATRX was necessary for senescence in primary cells could not be addressed, whether ATRX foci were formed during senescence in such cells was analyzed. As indicated in Example 2, ATRX foci increased in WI38 and EVIR90 non-immortalized and non-transformed diploid human fibroblasts following exposure to doxorubicin or gamma-irradiation, conditions that are well known inducers of senescence in these cells and also elicited the accumulation of SAP-gal, a hallmark of senescence (Figure 1 IK). Likewise, an increase in ATRX foci was detected in cells in which expression of KRAS^V12 induced accumulation of SA-β- gal, and in cells that were undergoing replicative senescence during serial passaging. Thus, the increase in ATRX foci occurs in different normal and transformed cells exposed to different stimuli that can induce senescence. On the other hand, an increase in ATRX foci was not seen in three liposarcoma cell lines (LS8107, LS7785-1, and LS7785-10) and a lung cancer cell line (H358) in which CDK4i induce quiescence (Figures 11G and Figure 12).

Additionally, there was no increase in ATRX foci when LS8817 cells quiesced following serum starvation (Figure 11L), or when MCF7 and MCF-IOA cells

quiesced and became autophagic after serum starvation (Figure 12D-J), or when

LS8107 cells were differentiated into adipocytes (Figure 12K-L). Thus, the increase in ATRX foci formation occurred soon after the cells were exposed to the stimulus and was specific in cells that exited the cell cycle and are on a path destined to senescence.

Table 1. Sequences of Forward and Reverse Primers for Gene expression analysis, sequencing and ChlP-qPCR

Z F -S 07 ?-CCAACACTCAAACCTAA7TAACCA-3- 5'-6GTTTGAAGACTGGCTAAAA<3C-3'

ATRX plays a role in the maintenance of the SAHF. First described by Narita and colleagues (Narita et al. Rb-mediated heterochromatin formation and silencing of E2F target genes during cellular senescence. Cell 113, 703-716 (2003)) as 30-50 bright punctate DNA dense foci not present in quiescent cells, there is now a detailed molecular understanding of the assembly of SAHF (reviewed in Rai and Adams. Lessons from senescence: chromatin maintenance in non-proliferating cells. Biochim Biophys Acta 1819, 322-331 (2013)). First, the chromatin remodeling enzyme HIRA and the structural protein HPl are recruited to the PML nuclear body where they are presumably modified in some way to support SAHF assembly (Zhang et al. Molecular dissection of formation of senescence-associated heterochromatin foci. Molecular and cellular biology 27, 2343-2358 (2007)). This is followed by the association of HIRA with the histone chaperone Asfl, which deposits histone H3 at particular sites marked by Rb-E2F binding, and macroH2A (mH2A) is recruited and begins to drive chromatin condensation (Zhang et al. Formation of MacroH2A- containing senescence-associated heterochromatin foci and senescence driven by

ASFla and HIRA. Developmental cell 8, 19-30 (2005); Rai et al. HIRA orchestrates a dynamic chromatin landscape in senescence and is required for suppression of neoplasia. Genes & Development 28, 2712-2725 (2014)). Finally, H3K9me3 and HPl accumulate at these sites (Zhang et al. (2007)).

The formation of PML nuclear bodies plays a broad role in senescence

(Borden. Pondering the promyelocytic leukemia protein (PML) puzzle: possible functions for PML nuclear bodies. Molecular and cellular biology 22, 5259-5269 (2002); Ferbeyre et al. PML is induced by oncogenic Ras and promotes premature senescence. Genes & Development 14, 2015-2027 (2000); Salomoni and Pandolfi. The role of PML in tumor suppression. Cell 108, 165-170 (2002)). ATRX foci overlap with both the PML nuclear bodies and ΗΡΙγ foci in senescent cells (Figures 1 IE, 1 IF). Knocking down ATRX did not diminish the number of the PML bodies, but the induction of ΗΡΙγ foci was greatly diminished (Figure 13 A). PML foci accumulated within one day of exposure to the senescence stimulus (Figure 14) while ATRX foci are first detected 2 days following the stimulus. SAHF are not detected for at least four to five days. This suggests that ATRX is required later than the accumulation of PML foci and may be directly involved in the formation or maintenance of the SAHF. To determine if ATRX was required for the maintenance of SAHF, senescent LS8817 cells were transduced with a lentivirus expressing the ATRX hairpin and after selection for 10 days in the presence of CDK4 inhibitor the accumulation of ATRX, the accumulation of SA-P-gal and SAHF were measured, and the irreversibility of arrest after removal of CDK4 inhibitor (Figure 13B). Loss of ATRX in a senescent cell did not reduce the accumulation of SA-P-gal positive cells, did not reduce the expression of the SASP genes (some of which were even modestly enhanced) or the irreversibility of growth arrest after CDK4 inhibitors were removed as measured by both incorporation of BrdU or long term clonogenicity after replating in the absence of the drug (Figures 13C-G). However, the accumulation of SAHF was decreased (Figure 13H) and the PML foci remained unchanged (Figure 14). These data indicate that ATRX was required for the maintenance of the SAHF in senescent cells.

Genome wide expression analysis in quiescent and senescent LS8817 cells. Previous efforts to describe the differences between genome wide expression in quiescent and senescent cells have used a single cell type treated with different inducers leading to the different outcomes (Lackner et al. A genomics approach identifies senescence specific gene expression regulation. Aging Cell 13, 946-950 (2014); Mason et al. Molecular signature of oncogenic ras-induced senescence. Oncogene 23, 9238-9246 (2004); Nelson et al. A comparison of oncogene-induced senescence and replicative senescence: implications for tumor suppression and aging. Age (Dordr) 36, 9637 (2014); Purcell et al. Gene expression profiling of replicative and induced senescence. Cell cycle 13, 3927-3937 (2014); Rovillain et al. Activation of nuclear factor-kappa B signalling promotes cellular senescence. Oncogene 30, 2356-2366 (2011); Zhang et al. Senescence-specific gene expression fingerprints reveal cell-type dependent physical clustering of up-regulated chromosomal loci. Proceedings of the National Academy of Sciences of the United States of America 100, 3251-3256 (2003)). It was thought that an isogenic cell line that responded differently to a single inducer might provide additional insight into the transcriptional networks that distinguish these cellular outcomes. Thus, a genome wide RNA sequencing (RNA-seq) was carried out to identify genes differentially expressed in ATRX deficient and wild type LS8817 cells before and after the addition of CDK4 inhibitor and to identify how ATRX affects the transition between quiescence and senescence. Cells in the absence of ATRX (LS8817 shATRX) underwent quiescence and control (LS8817 scr) cells underwent senescence when treated with CDK4 inhibitors; thus, it was not surprising that the changes in gene expression associated with CDK4 inhibition were dramatically affected by ATRX loss (Figure 15 A). Using analysis of variance (ANOVA) to interrogate which genes were up- or down-regulated by at least a factor of 1.8 in untreated versus treated cells and an FDR<0.05, there were 4894 genes whose expression changed significantly in control cells and only 739 in ATRX knockdown cells (Figure 15B). 324 of these genes overlapped. This indicates that ATRX plays an integral role shaping the gene expression program between quiescent and senescent cells induced by CDK4 inhibition. Consistent with previous reports, the most commonly down-regulated GO terms included those relating to the cell cycle, DNA replication and repair. However, unlike those studies, up-regulation of the F-κΒ network was not observed. This may be because of cell-type or inducer-specific nuances in the SASP program (reviewed in Salama et al. Cellular senescence and its effector programs. Genes & development 28, 99-114 (2014); Kuilman et al. The essence of senescence. Genes & Development 24, 2463-2479 (2010)).

To identify which Gene Ontology (GO) categories were enriched specifically to each condition, the genes were next separated by those that were enriched or repressed in the senescent cells or quiescent cells compared to the untreated cycling cells using Enrichr (Chen et al. Enrichr: interactive and

collaborative HTML5 gene list enrichment analysis tool. BMC Bioinformatics 14, 128 (2013)). The genes that were upregulated in the drug treated LS8817 senescent cells were strongly enriched for extracellular matrix organization (GO: 0030198, - log(p value) 7.44), extracellular structure organization (GO:0043062, -log(p value) 7.41), hemophilic cell adhesion via plasma membrane adhesion molecules (GO:

0007156, -log(pvalue) 6.60, and cell-cell adhesion via plasma membrane adhesion molecules (GO:0098742, -log(pvalue) 5.84). The genes that were most strongly down-regulated were enriched for mitotic cell cycle (GO; 0000278, -log(pvalue) 61.53), DNA repair (GO:0006281, -log(pvalue) 34.49), cell cycle phase transition (GO: 0044770, -log(pvalue)33.04) and mitotic cell cycle phase transition (GO:

0044772, -log(pvalue) 32.55). The expression in the drug-treated LS8817 shATRX quiescent cells could not be analyzed because fewer than 10 of the regulated genes belonged to any single GO category. Enrichr was also used to identify which transcription factor (TF) targets were specifically affected by the loss of ATRX (Figure 15C). E2F4 target genes were enriched exclusively in the ATRX proficient cells, consistent with the hypothesis that ATRX is required to properly regulate E2F-dependent gene expression in SAFIF. Myc, Max, and p300 target genes were enriched in both, as expected since both quiescent and senescent cells have exited the cell cycle. In cells lacking ATRX, the targets of SUZ12 and SMC3 were specifically down-regulated following drug treatment.

In addition, EZH2 targets were up-regulated following drug treatment in the wild type LS8817 cells, whereas some EZH2 targets were up-regulated and others were down-regulated in shATRX cells after CDK4 inhibition. This suggests a general perturbation of EZH2 target gene expression as a result of ATRX loss.

To confirm the relationship between these GO categories and TF targets, a gene set enrichment analysis (GSEA) was used to compare the complete transcriptomes of drug treated cells to their asynchronous counterparts in both ATRX proficient and deficient populations (Subramanian et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles.

Proceedings of the National Academy of Sciences of the United States of America 102, 15545-15550 (2005)). Of the aforementioned gene sets, only E2F4 and GO DNA repair genes were significantly enriched in the cycling population, while only EZH2 genes were significantly enriched in the senescent drug treated population specifically in wild type LS8817 cells (Figure 15). In the shATRX cells, these gene sets were no longer significantly enriched, indicating that ATRX is important for their appropriate regulation following CDK4 inhibition.

HRAS is a direct target of ATRX repression in senescent cells. In order to gain this insight and identify common direct roles of ATRX in senescent cells, ChlP-sequencing in doxorubicin-treated and CDK4 inhibitor-treated LS8817 cells were performed after seven days of drug treatment. To identify binding sites specifically related to senescence ChlP-seq from quiescent serum starved and cycling LS8817 cells were also carried out. Confirming the specificity of the

immunoprecipitation, antibodies to ATRX failed to pull down sufficient DNA from U20S cells to prepare a library. Binding sites were identified using the irreproducible discovery format framework and filtering out the technical false positives and previously reported centromeric and telomeric peaks as well as other repetitive regions.

There were many ATRX binding sites specific to each individual inducer of senescence, and a few that were shared in common (Figure 16A). Because shared sites may suggest a "core" program underlying ATRX's activity in senescence, while inducer-specific sites may be important for sculpting specific nuances necessary for each, the 166 summits of ATRX binding enriched in both senescent CDK4i and doxorubicin-treated cells but absent in cycling or quiescent cells were focused on. One-third were in gene bodies or promoters (Figure 16B) representing 41 unique genes (Table 2). The remaining sites were in intergenic sequences (Table 3). This ratio of ATRX binding sites is consistent with that previously reported by others (Law et al. ATR-X syndrome protein targets tandem repeats and influences allele-specific expression in a size-dependent manner. Cell 143, 367-378 (2010)). Any relationship between the distribution of the intergenic sites along the chromosomes relative to centromeres, telomeres, or other binding sites were indiscernible. Likewise, a common structural or sequence element in these intergenic regions was not discernible, including G-quadruplexes which are known to be enriched at ATRX binding sites (Law et al. (2010)). These sequences were also not related to long non- coding RNAs, which might be involved in chromosome wide repression.

The Enrichr software was used to determine what known transcription factors regulate this subset of 41 genes. E2F1, MAX and FOXC1 were the most strongly enriched (Figure 16C and Table 2). The predominance of E2F binding sites was consistent with the hypothesis that ATRX was involved in the maintenance of the SAHF.

The expression of these 41 core genes by RNA-seq were analyzed.

Only 39 were identified. 18 of the 39 were similarly regulated in both the CDK4i and doxorubicin treated cells compared to cycling or serum starved cells (Table 2). 8 were decreased in at least three of the comparisons (KIF15, ATAD2, DHX9,

ARHGAP18, PARD3, HRAS, ZNF107, and SORCS3), and ten were increased in all four of the comparisons (RNF128, PLCB 1, HCN1, CTNDD2, TRPS1, COL26A1,

GAS6-AS1, SIN3B, HIBADH, and C20orfl95). After eliminating those changes that were specific to LS8817 senescent cells by comparing expression in another CDK4 inhibitor-induced senescent cell (LS0082) and a CDK4 inhibitor induced quiescent cell (LS8107) (Figure 16D) and confirming that these were ATRX dependent changes in LS8817 (Figure 16E) and U20S (Figure 16F) cells, only three genes remained, HRAS, SORCS3, and Z F107. ATRX binding was strongly enriched at HRAS locus in both the senescent LS8817 and LS0082 cells and not in the quiescent LS8107 cells in small scale ChlP-qPCR, and more modestly enriched at the Z F107 locus, being stronger in LS0082 than in LS8817 (Figure 16G). Evaluation of binding at the SORCS3 locus was problematic, because the sequences identified in chromatin immunoprecipitation were AT-rich making it difficult to generate qPCR primers to assess this. The suppression of HRAS was not limited to sarcoma derived cell types that have undergone senescence. HRAS (but not Z F107) expression was also suppressed in senescent glioma (S B 19) and lung cancer (A549, HI 975) cells following CDK4 inhibition, but not in a quiescent lung cancer cell line (H358) (Figure 17).

Examination of the ATRX ChlP-seq tracks around the HRAS locus in LS8817 cells showed clear focal binding enriched in senescent cells. This locus also contains a number of G-quadruplex forming sequence elements (Figure 16H), which often correlate to how strongly ATRX can affect target gene expression (Law et al. (2010)). ATRX binding to HRAS was detected strongly only in senescent cells late after the addition of CDK4 inhibitors (Figure 161). In U20S cells reconstituted with different alleles of ATRX, ATRX binding was only enriched at the HRAS locus and was not detected in cells expressing C240G and V588E mutant proteins (Figure 16J). Thus, HRAS is probably a bona fide target of ATRX and its repression may account for some of the senescence promoting activity of ATRX in response to CDK4 inhibition.

Table 2. RNA expression of genes in which ATRX binding was enriched in senescent cells.

Ratios of gene expression in CDK4i-induced or doxorubicin-induced senescent cells compared to either cycling untreated cells or quiescent cells induced by serum starvation are shown. Whether each gene belongs to the TRANSFAC/JASPAR PWM transcription factor target gene sets as described in Figure 16C is indicated. Table 3. Intergenic ATRX binding summits specific to senescent cells

c rom start end Annotation

1 91925938 91925939

102018219 102018404

105412514 105412523

209408416 209408486

239388803 239389186

249240657 249240661

13501361 13501428

18476404 18476451

68677257 68677357

89830549 89830948 Satellite repeat

89842097 89842579 Satellite repeat

90372669 90373039 Satellite repeat

170324275 170324747

234696198 234696415

5280825 5281398

176015569 176015636

3783531 3783536

12668659 12668821

32841093 32841720

49093261 49093513 Centromeric repeat

49101046 49101397 Centromeric repeat

49152326 49152682 Centromeric repeat

49644961 49645328 Centromeric repeat

72550758 72550863

174406163 174406321

1871481 1871550

18762886 18763065

10347767 10347851

10347928 10347944

10861184 10861241

49941286 49941585

55302722 55303293

97567854 97567952

123085501 123085540

152076921 152077079

152077575 152077644

12167595 12167898

28557300 28557702

95950978 95950985

115389236 115389304

115389604 11538981 Table 3 (continued)

156028585 156028662

7 303448 303662

13410402 13410825

1368078 1368166

24470223 24470403

61753367 61753692 Centromeric repeat 61754739 61754758 Centromeric repeat 61755180 61755181 Centromeric repeat 61785764 61785943 Centromeric repeat 61976292 61976485 Centromeric repeat 64686706 64686764

64687462 64687553

64688394 64688449

57957852 67957855

89587945 89588409

109256914 109256962

112436507 112436597

112437364 112437777

152595074 152595085

8 580611 581274

123511634 123511942

144830890 144831040

10 4538086 4538309

35212938 352 4343

35273735 35274491

36579319 36579973

38817263 38817894 Satellite repeat

42376312 42376513 Centromeric repeat

124406499 124406826

124845541 124845586

12 28802780 28803520

90486997 90487035

13 45998983 45999069 14 19349311 19349576

15 68132281 68132735

16 14379872 14379886

46385813 46386377 Satellite repeat 46401210 46401342 Satellite repeat 46418455 46418660 Satellite repeat 46419752 46420071 Satellite repeat 46425345 46425933 Satellite repeat Table 3 (continued)

46427376 46427635 Satellite repeat

46427947 46428131 Satellite repeat

4643741 46437826 Satellite repeat

46438756 46439060 Satellite repeat

46453996 46454222 Satellite repeat

1107027 1107069

25265249 25265420 Satellite repeat

77868506 77858574

18 14689273 14689486

23044031 23044157

28338370 28338473

28338647 28338876

74220747 74220815

246215 246233 TAR1

18715549 18715619

8716885 18717001

18719183 18719205

X 25695192 25695281

25695736 25695898

25696006 25696064

26 02926 26 02957

26103019 26 03701

26134399 26135146

58073171 58073271

61724028 61724201 Centromeric repeat

73765939 73765967

77524597 77524999

Reducing HRAS in quiescent cells can induce geroconversion.

The ability to assess the roles of mitogenic factors in senescence is limited by the fact that promoting cell proliferation will prevent cell cycle exit and thus senescence. Geroconversion is the transition of cells from quiescence into senescence, and thus a gene, mitogenic or not, can be determined if it has a role in driving geroconversion and senescence as disruption of the gene should only prevent senescence and the cell should still undergo quiescence. HRAS is a strong effector of mitogenic signaling and inhibiting its expression in cycling cells will trigger senescence. Nevertheless, whether suppressing HRAS in quiescent cells would affect their progression into senescence can be examined.

HRAS was knockdown in serum starved LS8817 cells and the accumulation of SA-P-gal and SAHF positive cells was measured, and growth after serum was restored. In addition, the effect of knocking down KRAS and NRAS as well as ZNF107 were also examined. All the targeting vectors were specific.

Knocking down one gene product did not affect the accumulation of the others (Figure 18 A). The magnitude of FIRAS repression achieved with the knockdowns was similar to that seen with CDK4 inhibition (Figure 18B). Following the loss of HRAS, but not the other RAS isoforms or Z F107, the number of SA-P-gal positive cells increased (Figure 18C). The number of SAHF positive cells also increased in the FIRAS knockdown (Figure 18D). This marker of cellular senescence was not analyzed in the other knockdowns. As expected of a senescent cell, however, clonogenic growth was reduced following restoration of serum to HRAS knockdowns but not the Z F107 knockdown cells (Figure 18E). This was repeated in two other cell lines in which ATRX function was compromised. First, in LS8817 cells in which ATRX was knocked down, and second, in LS8107 cells in which ATRX was phosphorylated and incapable of promoting senescence following treatment with CDK4 inhibitors (Kovatcheva et al. (2015)). In both circumstances, knocking down HRAS increased the number of SA-P-gal and SAHF positive cells (Figure 22E and F). Furthermore, loss of HRAS can drive geroconversion in LS8017 cells and LS8817 cells lacking ATRX that have been induced to quiesce via CDK4 inhibition (Figure 22A-D). Collectively, this indicates that the repression of HRAS contributes to the role of ATRX in senescence and that the reduction in HRAS expression induces senescence.

8.3 DISCUSSION

The chromatin remodeling enzyme ATRX plays a role driving cells into senescence in a variety of circumstances. ATRX accumulates in nuclear foci soon after cells exit the cell cycle, well before most other characteristic hallmarks of senescent cells are observed. Recruitment to such foci occurs in normal human diploid fibroblasts undergoing DNA damage induced senescence, replicative senescence, or oncogene induced senescence, and in multiple types of cancer cell lines induced to senescence by either doxorubicin or CDK4 inhibition. ATRX is a pleiotropic regulator of senescence; not only are there inducer specific binding sites and targets, it maintains the facultative heterochromatin of the SAHF and can represses the expression of HRAS late during therapy induced senescence.

At the molecular level, the recruitment of ATRX to foci depends on its interaction with H3K9me3 histone and HP1 proteins. Both of these are markers of transcriptionally repressed chromatin and major constituents of the facultative heterochromatic domain known as SAHF. Mutation of the ATPase and translocase domain does not greatly affect the formation of the foci, but it did prevent senescence. Without being bound to a particular theory, these data suggest that the ability of ATRX to remodel chromatin is important to its function.

ATRX foci were first detected early after cells exited the cell cycle, but after the increase in PML nuclear bodies was detected. Knocking down ATRX before treatment did not affect the increase in PML nuclear bodies and knocking down ATRX in senescent cells did not affect the number of PML nuclear bodies clearly indicating that ATRX operates downstream or parallel to this event. The number of ATRX foci continues to increase over time until the cells accumulate of SA-P-gal and SAHF, express the cytokine genes of the SASP, and irreversibly commit to senescence as evidenced by long term clonogenic growth arrest after the inducer is removed. Without being limited to a particular theory, it appears that targets of ATRX binding probably change as cells progress from quiescence into senescence. SAHF do not form until late in senescence, and the HRAS locus is not bound until late in senescence. Thus, earlier ATRX binding sites may be targeting inducer specific features of each senescence program. Some earlier binding sites are probably maintained and detected at later times, while others may undergo dynamic changes and be absent in the late senescent cells that were examined. That ATRX plays a critical role early as cells progress from quiescence into senescence is clear from the effect of the ATRX mutations which prevented the appearance of SA-P-gal in U20S cells. Additionally, knockdowns of ATRX prevented the accumulation of any senescence markers in response to CDK4 inhibition or doxorubicin. Only the SAHF phenotype was affected if ATRX levels were reduced once cells had already become senescent. Temporal mapping of the ATRX binding sites and correlation with other events should ultimately provide insight into the chromatin gymnastics that have been described as cells become senescent.

It is well accepted that a collection of markers is needed to identify a senescent cell, including but not limited to the accumulation of SA-P-gal, the accumulation of SAHF, and the expression of a cell type specific SASP cytokine profile (Salama et al. Cellular senescence and its effector programs. Genes & development 28, 99-114 (2014); Kuilman et al. The essence of senescence. Genes & development 24, 2463-2479 (2010)). Now added to this list is the appearance of an increased number of ATRX foci, which can be detected earlier than the others. Regardless of the nature of the cell type (non-immortal and non-transformed human diploid fibroblast, transformed cancer cell lines derived from breast, lung, glioma, prostate, and liposarcoma patients) or the inducing signal (replicative passaging, expression of oncogenic ras, DNA damage induced by doxorubicin or gamma- irradiation, or CDK4 inhibition) an increase in ATRX foci is observed soon after cells exit the cell cycle. This is an easy cytological assay and remarkably specific as foci do not increase in quiescent cells, autophagic cells or differentiated adipocytes. For example, an increase in ATRX foci can discriminate between CDK4 inhibitor treated LS8817 liposarcoma cells and MCF7 breast cancer cells that are embarking upon a senescence pathway from the same parental cells that are quiescent following serum starvation. These foci do not occur simply as a result of CDK4 inhibition as the number of ATRX foci fails to increase in plenty of cells that fail to senesce following CDK4 inhibition. ATRX foci formation is not simply delayed in quiescent cells as such foci could not be detected even after cells were held in quiescence for up to two weeks.

ATRX is an important mediator of geroconversion, the transition from quiescence to senescence, and not for the decision to exit the cell cycle. It is anticipated that screening for compounds and genetic events that are required for geroconversion will be practical using the detection of ATRX foci as a readout. Not only is such a screen temporally favorable, occurring a short time after the inducer is added, it is able to distinguish between cells embarking into a senescence program from other forms of growth arrest. The importance of this is clear when considering the effect of cytostatic chemotherapies or interfering with senescence to ameliorate age-related pathologies. In a small pilot study, it was shown that geroconversion can account for how patients with well-differentiated and dedifferentiated liposarcoma responded to single agent CDK4 inhibition (Kovatcheva et al. (2015); Dickson et al. Progression-free survival among patients with well-differentiated or dedifferentiated liposarcoma treated with CDK4 inhibitor Palbociclib: A Phase 2 clinical trial. JAMA Oncol. (2016); Dickson et al. Phase II trial of the CDK4 inhibitor PD0332991 in patients with advanced CDK4-amplified well-differentiated or dedifferentiated liposarcoma. Journal of clinical oncology: official journal of the American Society of Clinical Oncology 31, 2024-2028 (2013)). An increased understanding of the genetic requirements and molecular events underlying this transition, summarized in Figure 23, will help improve such treatments. For example, the rationale combination of combining CDK4 inhibitors with Ras pathway modulation may drive cells that would normally quiesce in response to CDK4 inhibition alone into senescence. There is abundant evidence that suppression of Ras-MAPK signaling can also induce senescence in a variety of breast, prostate, lung cancers and melanomas (Axanova et al. 1,25-dihydroxyvitamin D(3) and PI3K/AKT inhibitors synergistically inhibit growth and induce senescence in prostate cancer cells. The Prostate 70, 1658-1671 (2010); Haferkamp et al. Vemurafenib induces senescence features in melanoma cells. The Journal of investigative dermatology 133, 1601-1609 (2013); Ota et al. Sirtl inhibitor, Sirtinol, induces senescence-like growth arrest with attenuated Ras-MAPK signaling in human cancer cells. Oncogene 25, 176-185 (2006); Schick et al.

Trametinib radiosensitises RAS- and BRAF-mutated melanoma by perturbing cell cycle and inducing senescence. Radiotherapy and oncology: journal of the European Society for Therapeutic Radiology and Oncology, (2015)). Thus, understanding geroconversion may have a broad therapeutic efficiency (Blagosklonny. Cell cycle arrest is not yet senescence, which is not just cell cycle arrest: terminology for TOR- driven aging. Aging 4, 159-165 (2012); Blagosklonny. Geroconversion: irreversible step to cellular senescence. Cell cycle 13, 3628-3635 (2014); Ewald et al. Therapy- induced senescence in cancer. Journal of the National Cancer Institute 102, 1536- 1546 (2010); and Shay and Roninson. Hallmarks of senescence in carcinogenesis and cancer therapy . Oncogene 23, 2919-2933 (2004)) .

Elimination of senescent cells can ameliorate some of the pathologies associated with aging in mice (Baker et al. Naturally occurring pl6(Ink4a)-positive cells shorten healthy lifespan. Nature 530, 184-189 (2016); Baker et al. Clearance of pl6Ink4a-positive senescent cells delays ageing-associated disorders. Nature 479, 232-236 (2011)). Such pathologies are associated with the "sterile" inflammation promoted by the SASP in senescent diploid cells and thus significant effort is put forth to identify senolytic agents or understand the SASP such that it can be controlled. However, when considering such applications based on drugs such as ABT-203 or rapamycin (Chang et al. Clearance of senescent cells by ABT263 rejuvenates aged hematopoietic stem cells in mice. Nature medicine 22, 78-83 (2016); Zhu et al. The Achilles' heel of senescent cells: from transcriptome to senolytic drugs. Aging Cell 14, 644-658 (2015); Kirkland and Tchkonia Clinical strategies and animal models for developing senolytic agents. Exp Gerontol 68, 19-25 (2015)), senescent cells have been not realistically targeted, and eliminating the inflammatory response in general can have significant deleterious effects. Preventing senescence, if it can be done without inducing continued cell proliferation may be a viable option. One can readily appreciate that leaving a cell in a durable quiescent state may be clinically effective. If it does progress further to senescence, but does so at a slower rate it could allow the body's natural mechanisms for clearing senescent cells to function at an appropriate level.

Various references are cited herein, the contents of which are hereby incorporated by reference in their entireties.

Claims

WHAT WE CLAIM IS:

1. An assay for identifying drug candidates for use in treating an age-related disease, comprising:

(a) treating one or more cells with a compound and/or modality that

induces senescence;

(b) treating the one or more cells with a drug candidate; and

(c) determining the number of ATRX foci per cell in the compound and drug candidate-treated cells, as compared to a reference,

where a decrease in the number of ATRX foci per cell or an absence of an increase in the number of ATRX foci per cell compared to the reference is indicative that the drug candidate is likely to be useful as a therapy for treating an age-related disease.

2. The assay of claim 1, wherein the one or more cells is a responder cell.

3. The assay of claim 1, wherein the one or more cells is LS8817.

4. The assay of claim 1, wherein the compound and/or modality that induces senescence and the drug candidate are added to the one or more cells simultaneously.

5. The assay of claim 1, wherein the compound and/or modality that induces senescence and the drug candidate are added to the one or more cells at different timepoints.

6. The assay of claim 1, wherein the compound and/or modality that induces senescence is a CDK4 inhibitor.

7. The assay of claim 1, wherein the compound and/or modality that induces senescence is a HRAS inhibitor.

8. The assay of claim 1, wherein the compound and/or modality that induces senescence reduces HRAS expression.

9. The assay of claim 8, wherein the compound and/or modality that induces senescence is a shRNA or siRNA specific for HRAS.

10. The assay of claim 1, wherein the compound and/or modality that induces senescence reduces MDM2 expression.

11. The assay of claim 10, wherein the compound and/or modality that induces senescence is a shRNA or siRNA specific for MDM2.

12. The assay of claim 1, wherein the drug candidates are selected from the group consisting of small chemical molecules, biologies, and peptides.

13. The assay of claim 1, wherein the number of ATRX foci per cell is determined by immunofluorescence.

14. The assay of claim 1 or 13, wherein the ATRX foci is detected by an ATRX-specific antibody.

15. The assay of claim 14, wherein the ATRX-specific antibody is the antibody sold by Bethyl, Catalog No. A301 -045 A ("the '045 Ab"), a fragment thereof (e.g., a Fab fragment, a Fab₂, or the variable region comprised in a chimeric molecule) or an antibody that competitively inhibits binding of the '045 Ab to ATRX.

16. The assay of claim 1, wherein the reference sample is one or more cells treated with the compound that induces senescence alone.

17. An assay for identifying drug candidates for use in treating an age-related disease, comprising:

(a) treating one or more cells with a CDK4 inhibitor, where the cells enter a quiescent state upon treatment with the CDK4 inhibitor;

(b) treating the one or more CDK4 inhibitor-treated cells with shMDM2;

(c) treating the one or more cells with a drug candidate; and

(d) determining the number of ATRX foci per cell in the CDK4 inhibitor, shMDM2 and drug candidate-treated cells, as compared to a reference,

where a decrease in the number of ATRX foci per cell or the absence of an increase in the number of ATRX foci per cell compared to the reference is indicative that the drug candidate is likely to be useful as a therapy for the treating age-related disease.

18. The assay of claim 17, wherein the number of ATRX foci per cell is determined by immunofluorescence.

19. The assay of claim 17, wherein the one or more cells is a non-responder cell.

20. The assay of claim 17 or 19, wherein the one or more cells is LS8107.

21. The assay of claim 17, wherein the shMDM2 is expressed from a vector present in the one or more cells under the control of doxycycline.

22. The assay of claim 17, wherein the CDK4 inhibitor is PD0332991.

23. The assay of claim 17, wherein the drug candidate is selected from the group consisting of small chemical molecules, biologies, and peptides.

24. The assay of claim 17, wherein the ATRX foci is detected by an ATRX- specific antibody.

25. The assay of claim 24, wherein the ATRX-specific antibody is the antibody sold by Bethyl, Catalog No. A301 -045 A ("the '045 Ab"), a fragment thereof (e.g., a Fab fragment, a Fab₂, or the variable region comprised in a chimeric molecule) or an antibody that competitively inhibits binding of the '045 Ab to ATRX.

26. The assay of claim 1, wherein the reference sample is one or more cells treated with the CDK4 inhibitor and the shMDM2 in the absence of the drug candidate.

27. An assay for identifying drug candidates for use in treating an age-related disease, comprising:

(a) treating one or more cells with a senescence-inducing compound, where the cells enter a senescent state upon treatment with the senescence-inducing compound;

(b) exogenously expressing MDM2 in the one or more cells treated with the senescence-inducing compounds;

(c) reducing the exogenous expression of MDM2 and treating the one or more cells with a drug candidate; and

(d) determining the number of ATRX foci per cell as compared to a reference,

where a decrease in the number of ATRX foci per cell or the absence of an increase in the number of ATRX foci per cell compared to the reference is indicative that the drug candidate is likely to be useful as a therapy for the treating age-related disease.

28. The assay of claim 27, wherein the number of ATRX foci per cell is determined by immunofluorescence.

29. The assay of claim 27, wherein the one or more cells is a responder cell.

30. The assay of claim 27 or 29, wherein the one or more cells is LS8817.

31. The assay of claim 27, wherein the MDM2 is expressed from a vector present in the one or more cells under the control of doxycycline.

32. The assay of claim 27, wherein the senescence-inducing compound is a CDK4 inhibitor.

33. The assay of claim 32, wherein the CDK4 inhibitor is PD0332991.

34. The assay of claim 27, wherein the drug candidate is selected from the group consisting of small chemical molecules, biologies, and peptides.

35. The assay of claim 27, wherein the ATRX foci is detected by an ATRX- specific antibody.

36. The assay of claim 35, wherein the ATRX-specific antibody is the antibody sold by Bethyl, Catalog No. A301 -045 A ("the '045 Ab"), a fragment thereof (e.g., a Fab fragment, a Fab₂, or the variable region comprised in a chimeric molecule) or an antibody that competitively inhibits binding of the '045 Ab to ATRX.

37. The assay of claim 27, wherein the reference sample is one or more cells treated with the CDK4 inhibitor and the MDM2 in the absence of the drug candidate.

38. A kit for identifying drug candidates for use in treating an age-related disease comprising a means for detecting ATRX foci.

39. The kit of claim 38, wherein the means for detecting ATRX foci is an ATRX-specific antibody.

40. The kit of claim 39, wherein the ATRX-specific antibody is the antibody sold by Bethyl, Catalog No. A301-045A ("the '045 Ab"), a fragment thereof (e.g. , a Fab fragment, a Fab₂, or the variable region comprised in a chimeric molecule) or an antibody that competitively inhibits binding of the '045 Ab to ATRX.

41. The kit of claim 38 further comprising one or more cells.

42. The kit of claim 41, wherein the one or more cells are non-responder cells.

43. The kit of claim 42, wherein the non-responder cells are LS8107 cells.

44. The kit of claim 41, wherein one or more cells are responder cells.

45. The kit of claim 44, wherein the responder cells are LS8817 cells.

46. The kit of claim 41, wherein the one or more cells comprise a vector comprising shMDM2 under the control of doxycycline.

47. The kit of claim 41, wherein the one or more cells comprise a vector comprising shURAS under the control of doxycycline.

48. The kit of claim 41, wherein the one or more cells comprise a vector comprising MDM2 under the control of doxycycline.

49. The kit of claim 38 or 41 further comprising a senescence-inducing compound and/or modality.