Directed elimination of senescent cells by inhibition of BCL-W and BCL-XL

Abstract

Senescent cells, formed in response to physiological and oncogenic stresses, facilitate protection from tumourigenesis and aid in tissue repair. However, accumulation of such cells in tissues contributes to age-related pathologies. Resistance of senescent cells to apoptotic stimuli may contribute to their accumulation, yet the molecular mechanisms allowing their prolonged viability are poorly characterized. Here we show that senescent cells upregulate the anti-apoptotic proteins BCL-W and BCL-XL. Joint inhibition of BCL-W and BCL-XL by siRNAs or the small-molecule ABT-737 specifically induces apoptosis in senescent cells. Notably, treatment of mice with ABT-737 efficiently eliminates senescent cells induced by DNA damage in the lungs as well as senescent cells formed in the epidermis by activation of p53 through transgenic p14(ARF). Elimination of senescent cells from the epidermis leads to an increase in hair-follicle stem cell proliferation. The finding that senescent cells can be eliminated pharmacologically paves the way to new strategies for the treatment of age-related pathologies.

Figure 1. BCL-2 family members are elevated in senescent cells and provide resistance to apoptosis.

(a) IMR-90 human fibroblasts that were induced to senesce either through DNA damage (DIS), replicative exhaustion (RS) or oncogene expression (OIS), as well as controls proliferating cells (growing, G) and empty vector-transfected (V) cells, were treated for 10 h with TNF-α and CHX (TNF-α) or with vehicle (DMSO). Cell survival relative to vehicle-treated cells was determined by quantification of the remaining adherent cells. Histograms indicate the percentages of surviving senescent (DIS, RS and OIS) cells compared with G or V controls. Data are presented as mean±s.e.m of three repeats, performed in triplicates. (b) Western blot analysis of cleaved PARP, ICAD and caspase-3 proteins from senescent and control IMR-90 cells (labelled as in a), indicating levels of apoptosis after treatment with TNF-α and CHX. (c) Percentage survival of senescent and control cells (labelled as in a) 24 h after UV irradiation (80 J m⁻²). Data are presented as mean±s.e.m of three repeats, performed in triplicates. (d) Expression of BCL-2 family members (BCL-W, BCL-XL, BCL-2 and MCL-1) and of senescence effector proteins (p16, p21, p53) in senescent (DIS, RS, and OIS) and control (G, V) cells of both human (IMR-90) and mouse (MEF) origin. (e) Percentage survival of senescent and control cells (as in a) as well as in quiescent (Q) control cells after treatment for 24 h with the indicated concentrations of ABT-737, an inhibitor of BCL-W, BCL-XL and BCL-2. Data are presented as mean±s.e.m of three repeats, performed in triplicates. (f) Percentage survival of senescent (DIS and OIS) and control (G and V) cells treated with ABT-737 (10 μM) for 24 h with or without pretreatment for 6 h with the pan-caspase inhibitor z-VAD-fmk. Data are presented as mean±s.e.m of three repeats, performed in triplicates. (g) Western blot analysis of cleaved PARP and caspase-3 in the samples described in f. Data in b,d,g are representative blots of at least two independent experiments each. Data were analysed using Student's t-test. *P<0.05, **P<0.005.

Figure 2. BCL-W and BCL-XL maintain the viability of senescent cells.

(a) IMR-90 human fibroblasts (DIS, and OIS cells, as well as control G or V cells) were treated for 24 h with the indicated concentrations of Obatoclax or with vehicle (DMSO). Cell survival relative to vehicle-treated cells was determined by quantification of the remaining adherent cells. Histograms indicate the percentages of surviving senescent (DIS and OIS) cells compared with G or V controls. Data are presented as mean±s.e.m of three repeats, performed in triplicates. (b) Percentage survival of the cells described in a following 24 h treatment with the indicated concentrations of the BCL-2 inhibitor ABT-199 and of SH-SY5Y cells, which express high BCL-2 levels and serve as positive control for response to the drug. Data are presented as mean±s.e.m of three repeats, performed in triplicates. (c) Percentage survival of DIS cells transduced with siRNAs targeting BCL-2, BCL-W and BCL-XL or their combinations as indicated. (d) Western blot analysis of BCL-2, BCL-W and BCL-XL following siRNA treatment of DIS cells. (e) Percentage survival of DIS cells transduced with siRNAs targeting BCL-W, BCL-XL or both, with or without treatment with the BCL-2 inhibitor ABT-199 (10 μM). Data are presented as mean±s.e.m of three repeats, performed in triplicates. Data were analysed using Student's t-test. *P<0.05, **P<0.005, ***P<0.0005.

Figure 3. Molecular mechanisms regulating BCL-W and BCL-XL protein level in senescent cells.

(a,b) mRNA expression levels of BCL-W variants 1 and 2 (v1 and v2, respectively) in IMR-90 human fibroblasts (DIS, RS and OIS cells) as well as IMR-90 control (G or V) cells. Data are presented as mean±s.e.m of three repeats. (c) mRNA expression levels of BCL-XL in IMR-90 human fibroblasts (DIS, RS and OIS cells) as well as IMR-90 control (G or V) cells. Data are presented as mean±s.e.m of three repeats. (d) Western blot analysis of BCL-W and BCL-XL following inhibition of protein translation by CHX at the indicated time points after treatment. (e) Quantification of protein levels in d. Data are presented as mean±s.e.m of three independent experiments. (f) Polysomal profiles of control (G) and senescent (DIS) cells after sucrose gradient fractionation. The area under the curve was calculated for monosomes (fraction #1) and polyribosomes (polysomes, fractions 2−5), and indicates a lower polysomal content in DIS cells than in G cells. Data are presented as mean±s.e.m of three independent experiments. (g) Distribution of the mRNAs of GAPDH, Actin, p53 and BCL-XL in ribosomal fractions in senescent (DIS) cells and control (G) cells. The presented changes are relative to total mRNA. Values indicate the average percentage of mRNA in each fraction of total mRNA derived from three independent ribosomal fractionations. (h) Schematic representation of the bicistronic constructs. The BCL-XL construct contains the IRES of 5′-UTR BCL-XL (ref. 33). (i) Relative IRES activity is represented by the Fluc/Rluc ratio for all constructs transfected into IMR-90 cells. Data are presented as mean±s.e.m of three independent experiments. Data were analysed using Student's t-test. *P<0.05. **P<0.005.

Figure 4. ABT-737 eliminates senescent cells from lung of irradiated mice.

(a) Seven days post irradiation, mice were treated with ABT-737 (n=5) or vehicle (n=5) for 2 consecutive days. Lungs were dissected 1 day thereafter, sectioned and the sections were stained for SA-β-Gal. Scale bar, 10 μm. (b) Quantification of SA-β-Gal staining in the bronchioles area of vehicle- or ABT-737-treated mice. Data were presented as means±s.e.m. from five mice in each group. (c) One day following vehicle or ABT-737 treatment, lung were dissociated, stained for γH2AX and analysed by FACS. Percentage of γH2AX-positive cells was quantified from this analysis. Data were presented as means±s.e.m. from five mice in each group. (d) Western blot analysis of Bcl-w, p53 and p21 in the mice described in a. Two mice represent each condition, each line represents a mouse. (e) Lung sections from unirradiated mice and the mice described in a were stained for Bcl-xL, p53 and cleaved caspase-3, a marker of apoptosis. Scale bar, 10 μm. Data in b,c were analysed using Student's t-test. *P<0.05, ***P<0.0005.

Figure 5. ABT-737 treatment eliminates senescent epidermal cells and induces stem cell proliferation.

(a) Skin sections stained for SA-β-Gal (blue) of control tet-p14 (Ctrl) mice, and K5-rtTA/tet-p14 mice treated with dox for 4 weeks to activate p14^ARF, and subsequently treated with ABT-737 (p14+ABT) or vehicle (p14+V) for 4 consecutive days. (b) SA-β-Gal⁺ cells per microscopic field in mice as in a. (c) FACS analyses of SA-β-Gal activity (C₁₂FDG stain) in epidermal cells isolated from indicated mice. Gate indicates SA-β-Gal⁺ cell percentages. FSC-A—forward scatter. (d) Skin sections stained for the human p14^ARF (white, arrows) from indicated mice after 2 days of ABT-737 or vehicle treatment. K14 (green) marks the basal epidermis. (e) p14^ARF+ cells per field in mice as in c. (f) Sections of mice as in c stained for the apoptosis marker cleaved caspase-3 (CC3, arrows). (g) CC3⁺ cells per field in same mice. (h) Sections of hair follicle bulges of p14-expressing mice after 4 days of ABT-737 or vehicle treatment, stained for the proliferation marker Ki67 (green, arrows) and the bulge marker K15 (red). (i) Numbers of Ki67⁺K15⁺ cells per follicle per mouse in mice as in h. Dots represent mean number in individual mice, combining three independent experiments. (j) Representative FACS analyses of epidermal cells from indicated mice after 2 days of ABT-737 or vehicle treatment, stained for CD34, CD49f and Sca1. Charts show only Sca1^– (follicular) cells. Gate indicates percentage of CD34^+/CD49f^high hair-follicle stem cells. n=2 mice per group, experiment was done twice. Throughout, dots indicate individual mice, bars indicate mean±s.e.m. Data in b,e,f show mice combined from two independent experiments out of four conducted. *P<0.05; **P<0.005; ***P<0.0005 by Student's t-test. Scale bars, 25 μm.