Clearance of p16Ink4a-positive senescent cells delays ageing-associated disorders

Abstract

Advanced age is the main risk factor for most chronic diseases and functional deficits in humans, but the fundamental mechanisms that drive ageing remain largely unknown, impeding the development of interventions that might delay or prevent age-related disorders and maximize healthy lifespan. Cellular senescence, which halts the proliferation of damaged or dysfunctional cells, is an important mechanism to constrain the malignant progression of tumour cells. Senescent cells accumulate in various tissues and organs with ageing and have been hypothesized to disrupt tissue structure and function because of the components they secrete. However, whether senescent cells are causally implicated in age-related dysfunction and whether their removal is beneficial has remained unknown. To address these fundamental questions, we made use of a biomarker for senescence, p16(Ink4a), to design a novel transgene, INK-ATTAC, for inducible elimination of p16(Ink4a)-positive senescent cells upon administration of a drug. Here we show that in the BubR1 progeroid mouse background, INK-ATTAC removes p16(Ink4a)-positive senescent cells upon drug treatment. In tissues--such as adipose tissue, skeletal muscle and eye--in which p16(Ink4a) contributes to the acquisition of age-related pathologies, life-long removal of p16(Ink4a)-expressing cells delayed onset of these phenotypes. Furthermore, late-life clearance attenuated progression of already established age-related disorders. These data indicate that cellular senescence is causally implicated in generating age-related phenotypes and that removal of senescent cells can prevent or delay tissue dysfunction and extend healthspan.

Conflict of interest statement

The authors declare no competing financial interests.

Figures

Figure 1. Generation and characterization of INK-ATTAC transgenic mice

a, Schematic of the INK-ATTAC construct and the mechanism of apoptosis activation. b, GFP intensity of IAT. c, qRT–PCR analysis of the indicated tissues of 10-month-old mice. ATTAC, INK-ATTAC; H/H, BubR1H/H; SkM, skeletal muscle (gastrocnemius). d, Bone marrow cells harvested from 2-month-old mice immunostained for Flag after culture in the absence or presence of rosiglitazone for 48 h. e, SA-β-Gal stained IAT collected from 9-month-old mice of the indicated genotypes. f, Expression of senescence markers in tissues of 10-month-old mice measured by qRT–PCR. All increases are statistically significant (P < 0.05). g, FACS profile of single-cell suspensions from IAT of 10-month-old mice. Brackets indicate sorting gates. h, GFP+ and GFP− cell populations from IAT analysed for relative expression of senescence markers by qRT–PCR. All increases are statistically significant (P < 0.01). i, Bright field images of MEFs sorted into GFP+ and GFP− populations after induction of senescence and then stained for SA-β-Gal. For all experiments, n = 3 untreated females per genotype. Error bars, s.d. Scale bars in b, d and i, 20 μm. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 2. BubR1H/H;INK-ATTAC mice treated with AP20187 from weaning age on show delayed onset of p16Ink4a-mediated age-related phenotypes

a, Bone marrow cells cultured in rosiglitazone for 5 days and then treated or not treated with AP20187 (AP) for 2 days before SA-β-Gal staining. Scale bar, 50 μm. b, Incidence of lordokyphosis and cataracts. c, Representative images of 9-month-old mice. d, Mean skeletal muscle fibre diameters of 10-month-old mice. ABD, abdominal muscle; Gastro, gastrocnemius muscle. e, Exercise ability of 10-month-old AP20187-treated mice relative to age-matched untreated mice. Time is running time to exhaustion; distance is distance travelled at time of exhaustion; work is the energy expended to exhaustion. f, Body and fat depot weights of 10-month-old mice. Parentheses, s.d. Mes, mesenteric; Peri, perirenal; POV, paraovarian; SSAT, subscapular adipose tissue. g, Average fat cell diameters in IAT of 10-month-old mice. h, Dermis and subdermal adipose layer thickness of 10-month-old mice. Colour codes in e, g and h are as indicated in d. Error bars, s.e.m. For all analysis n = 6 female mice per genotype (per treatment). *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 3. AP20187-treated BubR1H/H;INK-ATTAC mice have reduced numbers of p16Ink4a-positive senescent cells

a, Images of SA-β-Gal stained IAT of 10-month-old mice. b–d, Expression of senescence markers in IAT (b), gastrocnemius (c) and eye (d) of 10-month-old AP20187-treated and untreated BubR1H/H;INK-ATTAC-3 mice relative to age-matched untreated WT;INK-ATTAC-3 mice. Error bars indicate s.d.; n = 3 females per genotype per treatment. The expression of all genes is significantly decreased upon AP20187 treatment (P < 0.05) with the exception of GFP in the eye. e, BrdU incorporation rates in IAT and skeletal muscle. Error bars, s.e.m.; n = 6 females per genotype per treatment. *P < 0.05.

Figure 4. Treatment of older BubR1H/H;INK-ATTAC mice with AP20187 delays progression of p16Ink4a-mediated age-related phenotypes

a, Mean skeletal muscle fibre diameters of the indicated mice. ABD, abdominal muscle; Gastro, gastrocnemius muscle. mo, months. b, Improvement of exercise ability of the indicated mice relative to age-matched untreated mice. c, Body and fat depots weights of the indicated mice. Parentheses, s.d. Mes, mesenteric; Peri, perirenal; POV, paraovarian; SSAT, subscapular adipose tissue. d, Average size of fat cells in IAT of the indicated mice. e, Subcutaneous adipose layer thickness of the indicated mice. f, SA-β-Gal-stained IAT. g, Expression of senescence markers in IAT and gastrocnemius of the indicated mice (n = 3 females per genotype per treatment). Expression of all genes, except those marked with NS, is significantly decreased (P < 0.05) upon late-life AP20187 treatment. Colour codes in d and e are as indicated in a. Error bars indicate s.e.m. except in g where they indicate s.d. For analyses in a–f: n = 5 5-month-old BubR1H/H;INK-ATTAC-5 −AP females; n = 9 10-month-old BubR1H/H;INK-ATTAC-3 +AP and −AP females; n = 7 10-month-old BubR1H/H;INK-ATTAC-5 +AP females; and n = 8 10-month-old BubR1H/H;INK-ATTAC-5 −AP females. *P < 0.05, **P < 0.01, ***P < 0.001. NS, not significant.