• Log in with Facebook Log in with Twitter Log In with Google      Sign In    
  • Create Account
  LongeCity
              Advocacy & Research for Unlimited Lifespans

Photo

Elimination of senescent cells by β-galactosidase-targeted prodrug attenuates inflammation and restores physical ...

apoptosis senescence

  • Please log in to reply
No replies to this topic

#1 Engadin

  • Guest
  • 198 posts
  • 580
  • Location:Madrid
  • NO

Posted 27 April 2020 - 07:04 PM


.

 

 

F U L L   T I T L E :   Elimination of senescent cells by β-galactosidase-targeted prodrug attenuates inflammation and restores physical function in aged mice.

 

 

O P E N   A C C E S S   S O U R C E :   nature

 

 

 

 

 

 

Abstract
 
Cellular senescence, a persistent state of cell cycle arrest, accumulates in aged organisms, contributes to tissue dysfunction, and drives age-related phenotypes. The clearance of senescent cells is expected to decrease chronic, low-grade inflammation and improve tissue repair capacity, thus attenuating the decline of physical function in aged organisms. However, selective and effective clearance of senescent cells of different cell types has proven challenging. Herein, we developed a prodrug strategy to design a new compound based on the increased activity of lysosomal β-galactosidase (β-gal), a primary characteristic of senescent cells. Our prodrug SSK1 is specifically activated by β-gal and eliminates mouse and human senescent cells independently of senescence inducers and cell types. In aged mice, our compound effectively cleared senescent cells in different tissues, decreased the senescence- and age-associated gene signatures, attenuated low-grade local and systemic inflammation, and restored physical function. Our results demonstrate that lysosomal β-gal can be effectively leveraged to selectively eliminate senescent cells, providing a novel strategy to develop anti-aging interventions.
 
 
 
Introduction
 
Aging is the predominant risk for physiological degeneration, increased chronic morbidities, and age-specific mortality.1,2 One major hallmark of aging is the chronic accumulation of cellular senescence, a permanent state of cell-cycle arrest in response to various damaging stimuli.3,4 Cellular senescence impairs the ability of tissue regeneration and drives chronic low-grade inflammation, which exacerbates the aging process.5,6 Importantly, transplantation of senescent cells into young mice was sufficient to drive age-related pathology and cause persistent physical dysfunction.7 In contrast, deletion of senescent cells by a genetic approach attenuated age-related deterioration and extended the health-span in aged mice.8,9 These studies demonstrated that senescence is one of the major drivers of aging and that clearing senescent cells is a promising approach to treat age-related diseases and improve physical function.6,10
 
Previous studies have shown that compounds termed ‘senolytics’ could kill senescent cells.11,12,13 Reported senolytics target anti-apoptotic pathways, which are up-regulated to inhibit apoptosis in senescent cells.11,14 These senolytics have been reported to eliminate certain types of senescent cells and have shown the potential to improve physiological function in several tissues.7,12,15 However, senolytic drugs have significant limitations in killing senescent cells in terms of specificity and broad-spectrum activity because of the dynamic and highly heterogeneous nature of the senescence program, which leads to the varying sensitivity of different types of senescent cells to current senolytic drugs.6,16,17 To overcome these challenges, it is highly demanded to develop a new strategy that permits selectively deleting senescent cells in a wide spectrum of cell types or tissues for anti-aging interventions.
 
To specifically target senescent cells, we focused on one primary characteristic of senescent cells — the increased activity of lysosomal β-galactosidase, exploited as senescence-associated β-galactosidase (SA-β-gal).18,19 Notably, SA-β-gal in diverse types of senescent cells is one widely used marker for identifying senescence in vitro and in vivo, which is linked to the increased content of lysosomes.20,21,22,23 Therefore, we hypothesized that lysosomal β-gal could be utilized for the design of a galactose-modified prodrug to target senescent cells in a broader spectrum. This prodrug could be processed into a cytotoxic compound by β-gal and subsequently delete senescent cells in a specific manner, a strategy that could overcome the limitations of current senolytic drugs.
 
Here, we designed a new prodrug that was specifically cleaved by lysosomal β-gal into cytotoxic gemcitabine and induced apoptosis in senescent cells. This prodrug eliminated both mouse and human senescent cells independent of the senescence inducers and cell types. In aged mice, our compound reduced SA-β-gal-positive senescent cells in different tissues, decreased senescence- and age-associated gene signatures, attenuated low-grade chronic inflammation, and improved physical function.
 
 
 
Results
 
Design of SSK1 to kill senescent cells selectively
 
To design a lysosomal β-gal-responsive prodrug, we first screened a panel of FDA-approved drugs to select a compound with potent cytotoxicity for senescent cells as an end-product (Supplementary information, Table S1). We chose gemcitabine because (1) it killed both mouse and human senescent cells effectively and potently (Fig. 1a; Supplementary information, Fig. S1a–c and Table S1); (2) its 4-amino group is a well-established site for prodrug development;24 and (3) it exhibited reduced systemic toxicity due to its short plasma circulation time.25,26 Therefore, we synthesized senescence-specific killing compound 1 (SSK1) (Fig. 1b), in which the acetyl group and β-gal-responsive moiety improved the cellular permeability and specificity, respectively.27 Gemcitabine has been reported to be transported into cells via molecular transporters for nucleosides given its hydrophilicity,24,25 however SSK1 was modified with acetyl group and β-gal-responsive moiety suggesting its hydrophobicity (Fig. 1b). Nucleoside transporter inhibitor was unable to block the killing effect of SSK1 indicating that SSK1 entered the senescent cells independent of transporter (Supplementary information, Fig. S1e). We also validated that the prodrug SSK1 was specifically cleaved to release cytotoxic gemcitabine in senescent cells but not in non-senescent cells (Fig. 1c; Supplementary information, Fig. S1d). These results suggested that SSK1 was toxic to senescent cells and non-toxic to non-senescent cells.
 
 
41422_2020_314_Fig1_HTML.png
Fig. 1: Design of SSK1 and validation of its ability to selectively kill senescent cells.
a Quantification of cell viability of non-senescent and replication-induced senescent new born dermal fibroblasts (NBFs) incubated with increasing concentrations of gemcitabine for 3 days (n = 3). b Molecular structure of the prodrug SSK1. c Metabolism of SSK1 into gemcitabine in replication-induced senescent NBFs and their non-senescent counterparts incubated with SSK1 (0.5 µM) for 3 days (n = 3). d Quantification of cell viability of senescent and non-senescent NBFs incubated with increasing concentrations of SSK1 for 3 days (n = 4). e Cell viability of GLB1 knockdown (shGLB1-1, -2 and -3) or shControl senescent cells treated with vehicle or SSK1 (0.5 µM) for 3 days (n = 4). For cell viability analysis in (a), (d) and (e), cell numbers were quantified using Hoechst 33342 staining and dead cells were excluded by propidium iodide (PI) staining. f Detection of phos-p38 MAPK in senescent cells or non-senescent cells incubated with SSK1 (0.5 µM) for 3 days by western blot. g Representative flow cytometric plots (left) and quantification (right) of the percentage of viable (Q4: PI− annexin V−) and apoptotic (Q2 and Q3: PI+ annexin V+ and PI− annexin V+) cells in vehicle- or SSK1-treated senescent cells by annexin V and PI staining after vehicle or SSK1 (0.5 µM) treatment for 3 days (n = 3). ‘n’ represents the number of biological replicates. Data are presented as means ± SEM. Unpaired two-tailed t-test for ©, (d), and (g), two-way ANOVA test for (e), *P < 0.05, ***P < 0.001, ****P < 0.0001.
 
 
 
To test SSK1’s ability to selectively kill senescent cells, we treated primary mouse fibroblasts and their replication-induced senescent counterparts with SSK1. We found that SSK1 selectively and potently eliminated β-gal-positive senescent cells within a wide therapeutic window (Fig. 1d; Supplementary information, Fig. S1f–i). Since SSK1 was cleaved into 4-(hydroxymethyl)-2-nitrophenol and gemcitabine in senescent cells, as a control, the hydrolyzed product of SSK1 — 4-(hydroxymethyl)-2-nitrophenol showed no obvious toxicity to senescent and non-senescent cells (Supplementary information, Fig. S1j), and gemcitabine killed both cells at the similar concentration (Fig. 1a). To address whether SSK1 cleared senescent cells through β-gal rather than the higher metabolic rates in senescent cells, we tested another gemcitabine-based prodrug (LY2334737) lacking the galactose moiety and found no obvious difference in the elimination of senescent cells and non-senescent cells (Supplementary information, Fig. S1k). We further used RNA interference to decrease the expression of GLB1, the gene encoding β-gal (Supplementary information, Fig. S1l).28 Knockdown of GLB1 reduced SA-β-gal activity (Supplementary information, Fig. S1m) and showed little effect on other senescence markers, such as p16, p21 and Il1α (Supplementary information, Fig. S1n). More importantly, knockdown of GLB1 impaired the ability of SSK1 to kill SA-β-gal-positive senescent cells (Fig. 1e), suggesting that its specificity for senescent cells depended on lysosomal β-gal activity. Collectively, we leveraged lysosomal β-gal, one conserved characteristic of senescent cells to design a prodrug that specifically killed senescent cells.
 
Next, we explored the molecular mechanism of SSK1 in senescent cells. As gemcitabine has been reported to induce cell death through the activation of p38 mitogen-activated protein kinase (MAPK),29,30 we examined the phosphorylation status of p38 MAPK and its upstream MKK3/MKK6 in SSK1-treated senescent cells by western blot.31,32 After SSK1 treatment, both p38 MAPK and MKK3/MKK6 were activated by phosphorylation in senescent cells (Fig. 1f; Supplementary information, Fig. S2a, b), indicating that SSK1 could be processed into gemcitabine in senescent cells and activated the p38 MAPK signaling pathway. This was further confirmed by the treatment of p38 MAPK inhibitors Birb796, SB203580, and SB202190, which impaired SSK1’s ability to specifically kill senescent cells (Supplementary information, Fig. S2c). Thus, SSK1 killed senescent cells through the activation of the p38 MAPK signaling pathway. We also found that SSK1 was able to induce mitochondrial DNA damage in senescent cells (Supplementary information, Fig. S2d), similar to the reported ganciclovir, which also belongs to the nucleoside analogs as gemcitabine.33 Additionally, flow cytometry analysis showed that SSK1 induced senescent cells into annexin V and propidium iodide double-positive cells, and western blot result showed SSK1 could activate caspase 3, which indicated that SSK1 killed senescent cells by inducing apoptosis (Fig. 1g; Supplementary information, Fig. S2b). These results suggested that our prodrug SSK1 was activated by lysosomal β-gal and selectively killed senescent cells through the activation of p38 MAPK and induction of apoptosis.
 
 
 
SSK1 kills senescent cells in a broader manner
 
We then tested the specificity of SSK1 for mouse and human senescent cells. First, we used SSK1 to treat mouse embryonic fibroblasts (MEFs) in which senescence was induced by ionizing radiation, oncogene (KrasG12V) overexpression, or genotoxic stress (etoposide treatment). These senescent cells induced by various stimuli were selectively killed by SSK1, while non-senescent cells were largely unaffected (Fig. 2a; Supplementary information, Table S2). Second, we found that SSK1 could selectively kill replication-induced senescent lung fibroblasts from adult mice compared with non-senescent counterparts (Fig. 2b). Third, to understand if there is any efficacy difference of SSK1 between species, we treated senescent human embryonic fibroblasts (HEFs) induced by different stimuli with SSK1. The results showed that SSK1 selectively cleared human senescent cells induced by different stresses in a dose-dependent manner, without an obvious effect on non-senescent counterparts (Fig. 2c; Supplementary information, Fig. S2e, f and Table S2). We also treated human umbilical vein endothelial cells (HUVECs) and human preadipocytes with SSK1, and found that senescent cells were selectively killed by SSK1 when compared with their non-senescent counterparts (Fig. 2d; Supplementary information, Figs. S2g–i and S3e). Collectively, our results showed that SSK1 selectively killed both mouse and human senescent cells in a cell type- and senescence stimulus-independent manner.
 
 
41422_2020_314_Fig2_HTML.png
 
 
 
Fig. 2: SSK1 kills senescent cells in a broader manner.
a Quantification of cell viability of non-senescent and senescent mouse embryonic fibroblasts (MEFs) induced by etoposide, irradiation (10 Gy) or oncogene (KrasG12V) incubated with increasing concentrations of SSK1 for 3 days (n = 3). Significance analysis is shown in Supplementary information, Table S2. b Quantification of cell viability of non-senescent mouse lung fibroblasts from 3-month-old mice and replication-induced senescent counterparts incubated with increasing concentrations of SSK1 for 3 days (n = 4). c Quantification of cell viability of non-senescent and senescent human embryonic fibroblasts (HEFs) induced by replication, etoposide, H2O2 (200 µM), irradiation (10 Gy) or oncogene (KrasG12V) incubated with SSK1 for 3 days (n = 3). Significance analysis is shown in Supplementary information, Table S2. d Quantification of cell viability of non-senescent and replication-induced senescent HUVECs incubated with increasing concentrations of SSK1 for 3 days (n = 4). ‘n’ represents the number of biological replicates. Data are presented as means ± SEM. Unpaired two-tailed t-test for (b) and (d), two-way ANOVA test for (a) and ©, **P < 0.01, ****P < 0.0001.
 
 
 
We further compared the specificity of SSK1 with other three reported senolytic drugs (ABT263, dasatinib plus quercetin, and fisetin) on HEFs, human preadipocytes and HUVECs, which were previously used to study senolytics.11,12,17 In our study, SSK1 selectively killed senescent HEFs induced by different stimuli (Supplementary information, Fig. S3a and Table S2). ABT263 could also specifically kill senescent HEFs induced by different stresses but in a narrow concentration range (Supplementary information, Fig. S3b and Table S2). On the other hand, the combination of dasatinib and quercetin (D + Q) showed limited selectivity as it killed both senescent and non-senescent HEFs (Supplementary information, Fig. S3c and Table S2), and fisetin only killed certain senescent HEFs at high concentration (Supplementary information, Fig. S3d and Table S2). SSK1 also specifically killed senescent human preadipocytes (Supplementary information, Fig. S3e), while ABT263 and fisetin had limited elimination effect on this cell type (Supplementary information, Fig. S3f, h). D + Q could kill senescent human preadipocytes at high concentration with an affection of non-senescent cells (Supplementary information, Fig. S3g). Additionally, the experiments on HUVECs showed that SSK1 selectively eliminated senescent cells with a broader concentration spectrum compared with ABT263 (Supplementary information, Fig. S3i, j), whereas D + Q showed no selectivity to senescent and non-senescent cells (Supplementary information, Fig. S3k). These results demonstrated that SSK1 has higher specificity and potency to eliminate senescent cells.
 
 
 
SSK1 decreases senescent cells in lung-injured mice
 
Next, we examined the effect of SSK1 on senescent cells in vivo. We employed two independent in vivo models of senescence: stress-induced senescence and naturally occurring senescence. First, we induced DNA strand break and cellular senescence in mouse lungs by intratracheal instillation of bleomycin.34,35 SA-β-gal-positive senescent cells were easily induced in these mice with lung injury (Supplementary information, Fig. S4a, b), a commonly used model to study elimination of senescent cells.36,37,38 We treated these lung- injured mice with vehicle or SSK1 by intraperitoneal injection (Fig. 3a) and found that SSK1 significantly reduced the percentage of SA-β-gal-positive cells in lung by 3.8-fold compared with that in vehicle-treated lung-injured mice (Fig. 3b). RT-qPCR results revealed that SSK1 also decreased senescence-associated markers compared with vehicle treatment (Fig. 3c). In consistent with this result, our RNA-seq results indicated that the expression of senescence-related genes was significantly decreased in lung tissues from SSK1-treated group (Fig. 3d). We further confirmed that SSK1 reduced p21-positive cells by immunofluorescence staining (Fig. 3e). Taken together, our results indicated that SSK1 could eliminate SA-β-gal-positive cells and decrease senescence-associated markers in vivo.
 
 
41422_2020_314_Fig3_HTML.png
 
 
Fig. 3: SSK1 eliminates SA-β-gal-positive senescent cells in a bleomycin-induced lung injury model.
a Experimental design for bleomycin-induced lung injury. Mice (3–6-month-old) were subjected to transtracheal injection of bleomycin (1.5 mg/kg) or saline (sham surgery, Sham) 5 days before drug treatment. Lung-injured mice were intraperitoneally injected with SSK1 (0.5 mg/kg) or vehicle (DMSO) consecutive two days every week for four weeks; sham surgery mice were treated with DMSO in the same way. b Representative images (left) and quantification (right) of SA-β-gal staining of lungs from bleomycin-injured mice after vehicle or SSK1 treatment (vehicle-treated, n = 6; SSK1-treated, n = 5). Scale bars, 200 µm. c Expression of senescence-associated genes by RT-qPCR in lungs of sham surgery and bleomycin-injured mice treated with vehicle or SSK1 (Sham, n = 9; vehicle-treated, n = 9; SSK1-treated, n = 10). d Heatmap of senescence-related gene expression of lungs from sham surgery and bleomycin-injured mice treated with vehicle or SSK1 (Sham, n = 6; vehicle-treated, n = 6; SSK1-treated, n = 6). e Representative images (left) and quantification (right) of p21 immunofluorescence of lungs from bleomycin-injured mice after vehicle or SSK1 treatment (vehicle-treated, n = 11; SSK1-treated, n = 14). White arrowheads show p21-positive cells. Scale bars, 50 µm. Each data point represents an individual mouse. ‘n’ represents the number of mice. Data are presented as means ± SEM. Unpaired two-tailed t-test, *P < 0.05, **P < 0.01.
 
 
 
We further found that lung fibrosis and the expression of fibrosis genes, including Col1a1, Col3a1 and Fn1, an outcome mediated by the accumulation of senescence,36,39 were reduced by SSK1 treatment (Fig. 3d; Supplementary information, Fig. S4c, d). We then evaluated the change of systemic health parameters following SSK1 treatment. We found that the treatment of SSK1 was beneficial to the body weight recovery in bleomycin-induced lung-injured mice (Supplementary information, Fig. S4e). Especially, exercise capacity and muscle strength measured by treadmill and grip strength respectively were significantly increased in SSK1-treated mice compared with vehicle-treated mice (Supplementary information, Fig. S4f, g). Additionally, SSK1 treatment showed a tendency to increase the survival rate (Supplementary information, Fig. S4h). These results suggested that SSK1-induced reduction of SA-β-gal-positive senescent cells provides a potential method to alleviate senescence-related lung injury.
 
 
 
SSK1 attenuates the senescence-associated secretory phenotype (SASP) and age-associated signatures in aged mice
 
To further examine the effect of SSK1 on naturally occurring senescent cells in aged mice, we treated 20-month-old C57BL/6 mice with SSK1, gemcitabine or vehicle intermittently for 8 weeks (Fig. 4a). Senescent cells accumulated in different tissues of aged mice,4,40,41 which was consistent with our observation that SA-β-gal-positive cells increased in aged livers, kidneys and lungs (Supplementary information, Fig. S5a–c). We found that SSK1 reduced SA-β-gal-positive cells in liver, kidney and lung tissues of aged mice compared with vehicle and gemcitabine (Fig. 4b, c; Supplementary information, Fig. S5d). SSK1 treatment also decreased the expression of other senescence markers, including the CDK inhibitors p16 and p21 in aged mice as indicated by RT-qPCR analysis compared with vehicle and gemcitabine treatment (Fig. 4d, e). Additionally, SSK1 treatment in aged mice could down-regulate the gene signatures associated with senescence as shown by gene set enrichment analysis (GSEA) in both livers and kidneys (Fig. 4f, g). These results indicated that SSK1 reduced naturally accumulated senescent cells and decreased senescence markers in mice.
 
 
 
 
 
.../...
 
 
 
 
 
 
 
 
 
 
 
.

 


Edited by Engadin, 27 April 2020 - 07:06 PM.






Also tagged with one or more of these keywords: apoptosis, senescence

1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users