Aged‐senescent cells contribute to impaired heart regeneration

Abstract Aging leads to increased cellular senescence and is associated with decreased potency of tissue‐specific stem/progenitor cells. Here, we have done an extensive analysis of cardiac progenitor cells (CPCs) isolated from human subjects with cardiovascular disease, aged 32–86 years. In aged subjects (>70 years old), over half of CPCs are senescent (p16INK4A, SA‐β‐gal, DNA damage γH2AX, telomere length, senescence‐associated secretory phenotype [SASP]), unable to replicate, differentiate, regenerate or restore cardiac function following transplantation into the infarcted heart. SASP factors secreted by senescent CPCs renders otherwise healthy CPCs to senescence. Elimination of senescent CPCs using senolytics abrogates the SASP and its debilitative effect in vitro. Global elimination of senescent cells in aged mice (INK‐ATTAC or wild‐type mice treated with D + Q senolytics) in vivo activates resident CPCs and increased the number of small Ki67‐, EdU‐positive cardiomyocytes. Therapeutic approaches that eliminate senescent cells may alleviate cardiac deterioration with aging and restore the regenerative capacity of the heart.

as negative controls for all flow cytometry procedures. All antibodies were applied at 1:10 diluted in incubation media for 30 min at 4°C with agitation. Data were analysed using FlowJo® software (FlowJo LLC).
SA-β-gal pos CPC sorting was performed by fluorescent activated cell sorting (FACS) using an ImaGene Green C 12 FDG lacZ gene expression kit (Life Technologies, USA) according to the manufacturer's instructions and protocols previously reported (Debacq-Chainiaux, Erusalimsky et al. 2009). Briefly, freshly isolated CPCs were resuspended in 1ml of 33µM C 12 FDG β-gal diluted in pre-warmed DMEM/F12 media and incubated at 37°C for 30 minutes with agitation. The cell suspension was then sorted based upon SA-β-gal pos fluorescence using FACS.

Immunocytochemistry
Human CPCs were either freshly isolated or grown in culture and then directly cytospun onto poly-lysine-coated slides using a Shandon Cytospin 4 Cytocentrifuge (ThermoFisher Scientific). Slides were immediately fixed using Shandon Cell-Fixx (ThermoFisher Scientific). After fixation, cells were allowed to air dry before proceeding with immunostaining. To prepare cells for immunostaining, slides were incubated with 95% ethanol for 15 min at room temperature, washed with PBS, and in the case of intracellular staining, incubated in 0.1% Triton X-100:PBS at room temperature for 10 minutes. After washing with 0.1% tween:PBS and a 1 hr incubation in 10% donkey serum, cells were incubated overnight at 4°C with primary antibodies to c-kit, CD45, CD31, CD34, p16 INK4A , γH2AX, or Ki67 (Supplementary Table 2) all applied at 1:50 in 0.1% Tween:PBS. Slides were then washed and incubated with corresponding Alexafluor TM secondary antibodies (Life Technologies) for 1 hour at 37°C. The nuclear DNA of the cells was counterstained with 4',6diamidino-2-phenylindole (DAPI) (Sigma-Aldrich) at 1µg/ml and mounted using Vectasheild mounting media (Vector labs). The cells were viewed and imaged using an ApoTome fluorescent microscope (Zeiss) or A1 confocal microscope (Nikon). A minimum of 20 random fields of view at x20 magnification were used to determine the percentage of positively stained cells, expressed as a percent of total nuclei.
SA-β-gal (SABG) staining was performed using a SA-β-gal staining kit (Cell Signaling Technology) according to the manufacturer's instructions and protocols previously reported (Debacq-Chainiaux, Erusalimsky et al. 2009).. In brief, human CPCs were either freshly isolated and directly cytospun onto poly-lysine-coated slides (ThermoFisher Scientific) or grown in culture and then fixed in situ using 2% paraformaldehyde: PBS (vol/vol) (Sigma-Aldrich) for 5 minutes at room temperature. Following fixation, cells were washed with PBS before being incubated in SABG activity solution (pH 6.0) at 37°C overnight. The enzymatic reaction was stopped by washing slides with ice-cold PBS and SABG staining was fixed with ice-cold methanol for 30s before mounting/visualising. Senescent CPCs were identified as blue-stained cells using light microscopy. A minimum of 10 images were taken at x4 magnification from random fields and the percentage of SA-β-gal cells were expressed as percentage of total nuclei.

Q-FISH analysis
Freshly isolated human CPCs were cytospun and fixed as described above.

Proliferation, clonogenicity, cardiosphere formation and cardiomyocyte differentiation assays in vitro
A BrdU incorporation assay (Roche) was used to determine cell proliferation, as previously described (Ellison, Torella et al. 2011).. In brief, 2.5 x 10 3 CPCs were plated in 24 well plates and serum-starved for 6 h in DMEM/F12 medium. Medium was then replaced with growth medium and BrdU was added, 1μg/ml every 8 hours. Cells were fixed after 24 hours and BrdU incorporation was assessed using a BrdU detection kit (Roche) according to the manufacturer's instructions (Supplementary Table 2). Nuclei were counterstained with DAPI (Sigma-Aldrich). Cells were viewed and imaged using an ApoTome fluorescent microscope (Zeiss). A minimum of 10 random fields of view at x20 magnification were analysed and number of BrdU-positive cells expressed as a percent of total nuclei.
Human CPCs (P2-P3) were serially diluted and deposited into CTS CELLstart pre-coated 96well plates (Corning Inc., USA) at 1 cell per well, as described previously (Vicinanza, Aquila et al. 2017). Individual cells were cultured in growth medium which was refreshed every 3 days. Wells containing CPC colonies were scored by bright-field microscopy after 7 days of culture to determine clonal efficiency. The clonogenicity of the human CPCs was expressed as the percent number of clones generated from the total number of single cells plated. A minimum of 3 plates were quantified per subject.
Medium was replaced every 3 days and cultures were fixed and analysed at 14 days.
Cardiosphere formation was quantified from a minimum of 10 images per subject taken at x4 magnification from random fields. The number of cardiospheres were expressed per cm 2 .
Cardiosphere size was determined from the same images by measuring the diameter of the spheres using ImageJ software (Fiji, USA). A minimum of 20 spheres per donor were measured. Cardiomyocyte differentiation was measured by immunostaining of differentiated cultures using Nkx2.5 (1:20, overnight 4°C) and α-sarcomeric actin (1:50, 1 hour 37°C) Table 2). Secondary antibodies were Alexa Fluor 488, or 594 (Life Technologies). Nuclei were counterstained with DAPI. A minimum of 5 images at x20 magnification from random fields were taken per subject for quantification of Nkx2.5 pos cells per total nuclei and fluorescent intensity of α-sarcomeric actin expression.

Senescence induction, conditioned media, and multiplex SASP protein analysis
To induce senescence pharmacologically, cycling-competent CPCs were exposed to Rosiglitazone (0.1 µM; Calbiochem) or Doxorubicin (0.2 µM; Calbiochem) at ~70% confluence for 24 h. Medium was removed and cell cultures washed three times with PBS before replacing with fresh growth medium. Culture medium was refreshed every 3 days and CPCs were maintained for up to 35 days. Cells were analysed for SA-β-gal (as described above) and p16 INK4A expression over this time course by quantifying the total number of positive cells per total nuclei from a minimum of 5 random fields at 10x and 20x magnification respectively. Conditioned media (CM) were prepared by pre-washing cyclingcompetent or senescent CPC cultures (5x10 5 cells) three times with PBS, then exposing them to RMPI 1640 containing 1 mM sodium pyruvate, 2 mM glutamine, MEM (minimum essential medium), vitamins, MEM non-essential amino acids, and antibiotics (ThermoFisher Scientific) for 24 h. CM was filtered through a 0.2 µM filter and stored at -20°C until ready for use or analysis. Luminex® xMAP technology was used to quantify SASP factors in CM.
Multiplexing analysis was performed using the Luminex 100 system (Luminex) by Eve Technologies using human multiplex kits (Millipore).

PKH26 cell labelling
Human cycling-CPCs, doxorubicin (Dox)-induced Senescent-CPCs or c-kit neg cells (5x10 5 ) were harvested and washed in serum-free DMEM/F12 medium before being resuspended in 25µl diluent C (Sigma-Aldrich). An equal volume of PKH26 labelling solution (4µM; Sigma-Aldrich) was added to the cell suspension and mixed thoroughly. Cells were incubated for 2 minutes at room temperature in the dark, before adding an equal volume of DMEM/F12 + 10% FCS to halt the labelling reaction. Cells were then centrifuged and washed three times in PBS before transferring to a new Eppendorf tube and re-suspending in 15µl PBS for transplantation in vivo. Initial PKH26 labelling was validated using flow cytometry compared to unlabelled cells. PKH26 labelled cycling CPCs and c-kit neg cells were also propagated in vitro to monitor label retention over culture passage (P1-P9). To quantify PKH26 pos labelling, cells were cytospun and counterstained with DAPI. The total number of positive cells per total nuclei were quantified from a minimum of 5 random fields at x20 magnification.

Acute myocardial infarction model
The Prior to echocardiography, mice were weighed and anaesthetised in an anaesthetic chamber at 0.5ml/min O 2 and 4% Isoflurane. The depth of anaesthesia was assessed by a change in the breathing pattern of the mouse and the absence of the pinch withdrawal reflex. Once under deep anaesthesia, mice were placed onto an imaging station (Vevo, Netherlands). Anaesthesia was regulated to maintain heart rate between 400-500 beats per minute monitored by ECG recordings, and body temperature was maintained at 37 ± 0.5  C using a temperature probe.
Mice were depilated and ultrasound gel was applied onto the vicinity of the chest (Aquasonic 100, Germany). Parasternal long and short axis images of the heart were taken using the to 2% isoflurane at 0.5ml/min for the duration of the surgical procedure. Proceeding depilation, the thoracic area was sterilised and a midline incision along the skin overlying the sternum was performed to expose the muscle. A left-sided thoracotomy between the fourth and fifth rib was performed to expose the heart. Following the separation of the pericardial sac, the coronary artery was ligated using 6-0 silk suture (Ethicon, US) using the left atrium and aorta as entry and exit landmarks, respectively. The force of the suture was determined by the appearance of a color change in the left ventricle. Following the permanent ligation, the heart was assessed for atrial fibrillation for a 10 minute interval and was preceded by the closure of the thorax and the skin using 5-0 silk suture (Ethicon, US). In sham-treated mice, a permanent ligature was not tied around the LAD coronary artery. Immediately following MI, mice were injected intramyocardial with PKH26 labelled cycling-CPCs, Senescent-CPCs, or c-kit neg cells (all 5x10 5 ) delivered in a total volume of 15µl PBS, delivered across two sites at the border zone. Directly following MI+cell injection, mice were implanted with an osmotic pump (Alzet ® , USA) loaded with a 0.2M solution of BrdU (MP Biomedicals), releasing the thymidine analogue for 14 days. Anaesthesia was then removed and mice were taken off mechanical ventilation after observing an independent breathing pattern dyssynchronous to that of the ventilator. Methadone analgesic was administered, Comfortan® (1mg/kg; Dechra) i.m., and mice were kept in heat boxes overnight in a constant warm environment (27±1 o C and a relative humidity of 50±5%). 24 and 48 hours following surgery, body weight and signs of pain were checked and post analgesic was administered as and when necessary. Following surgery, mice were housed singly caged with access to water and chow ad libitum.
Echocardiography measurements were taken at baseline (BL), 7 days, and 28 days post-MI, after which mice were sacrificed.

Senolytic drug treatment, viability, TUNEL staining in vitro
Cycling-competent human CPCs and Senescent-CPCs, induced by Doxorubicin ( Cell viability was measured by crystal violet (CV) staining at day 0 then following 3 days of drug exposure. Briefly, cells were washed twice with PBS, fixed in methanol on ice for 10 minutes, and stained with 0.5% crystal violet (Sigma-Aldrich) in methanol for 15 minutes at room temperature. Cells were washed with deionized water and staining intensity was measured using a light microscope, with a minimum of 5 random fields of view at 4x magnification taken per well. The captured images were analyzed using ImageJ software (Fiji, USA).
Cell apoptosis was detected using the terminal deoxynucleotidyl transferase (TdT)-mediated dNTP nick end labelling (TUNEL) assay (Trevigen) according to the manufacturer's instructions. Cells were cultured in 24-well plates as described previously and then treated with either vehicle or D+Q for 16 hours. To perform TUNEL staining, cells were incubated with proteinase K for 15 min at room temperature then a labelling buffer containing TdT enzyme, TdT dNTP, and Co 2+ was applied for 1h at 37°C. Positive controls were incubated with TACS nuclease, while negative controls omitted the TdT enzyme from the labelling mix. After labelling, cells were washed in stop buffer for 5 min and were then incubated with Strep-Fluorescein in 0.05% Tween:PBS for 20 min at room temperature. After washing, cells were counterstained with DAPI (Sigma-Aldrich) and mounted using Vectasheild mounting media (Dako). The cells were viewed and imaged using an ApoTome fluorescent microscope (Zeiss). A minimum of 5 random fields of view at x20 magnification were analysed and number of TdT-positive cells expressed as a percent of total nuclei.

Co-culture assays
CPCs were plated in the bottom well of 0.4µm pore Costar Transwell 24-well plates (Corning) and, once attached, were induced to senescence with Doxorubicin as described above and maintained for 30 days. Cycling-competent CPCs were plated at low density (200 cells/well) onto the Transwell membrane 1 day prior to co-culture. Cycling-competent CPCs were then combined with either senescent CPCs or cultured alone (control wells) and maintained for 7 days. At this time point, cultures were either fixed for analysis, conditioned media were collected, or senolytics were applied to transwells for 3 days to selectively clear senescent cells (as described above). After senolytic clearance, a sub-set of wells were fixed and stained for SA-β-gal (as described previously) to confirm effective clearance. Cyclingcompetent CPCs were maintained for a further 7 days, at which point conditioned media were collected for luminex analysis (as described above) and cultures fixed for analysis.

INK-ATTAC mice and drug treatments
The following experimental procedures were conducted in accordance with Mayo Clinic Institutional Animal Care and Use Committee (IACUC) guidelines. Both stocks of INK-ATTAC and C57BL/6 wild type mice were bred and aged at Mayo Clinic. The generation and characterization of INK-ATTAC mice has been described previously (Baker, Wijshake et al. 2011, Xu, Palmer et al. 2015 This dosing regime was chosen because it has been shown to be effective at clearing senescent cells in chronologically aged INK-ATTAC mice in previous studies (Xu, Palmer et al. 2015, Roos, Zhang et al. 2016. Another group of INK-ATTAC mice aged 3 months (n=10) and 22 months (n=10) were randomly assigned to treatment groups and injected intraperitoneally (i.p.) with either vehicle or AP20187 for 2 consecutive days every 2 weeks for 2 months as shown in Supplementary Fig 10c. These mice were injected (i.p) with EdU (123 mgkg -1 ) 2 hours prior to sacrifice.

Tissue collection, immunohistochemistry, and confocal imaging
Mice were sacrificed at either 4 days or 28 days after MI with the hearts arrested in diastole with cadmium chloride solution (Sigma), removed, and embedded in OCT compound INK-ATTAC and wild-type C57BL/6 mice were sacrificed 4 days after the last dose of the last course of treatment with AP20187 or senolytic drugs. Hearts were explanted and cut into two parts. One half was fixed in 4% formaldehyde for 24 hrs and embedded in paraffin. The other half was snap-frozen in liquid nitrogen for subsequent RT-qPCR analyses as described below. 10μm transverse heart sections were cut on a microtome (Leica) and mounted onto microscope slides. Antigen retrieval was achieved using Target

Real-time quantitative polymerase chain reaction (RT-qPCR)
To measure human gene expression, total cellular RNA was extracted and reverse transcription was performed as described previously . RT-qPCR reactions were run with SYBR Green PCR Master Mix (Bio-Rad) and primers (IDT) (Supplementary analyzed using Graphpad software. From mouse hearts, total RNA was extracted using Trizol (Thermo Fisher Scientific) and reverse transcription was performed using an M-MLV reverse transcriptase kit as described previously

Statistical analysis
Data are reported as mean ± SEM or mean ± SD. The data displayed normal variance. The experiments were not randomized, except for the in vivo animal studies as described above.
The investigators were blinded to allocation during experiments and outcome assessment.
Significance between 2 groups was determined by Student's t-test and in multiple comparisons by the analysis of variance (ANOVA) using GraphPad Prism (GraphPad Software  Viability of human CPCs exposed to different senolytic drugs for 3 days. The red line denotes cell viability on day 0 prior to senolytic exposure. (c) Dox-induced Sen-CPCs were exposed to different senolytics for 3 days and clearance was quantified using SA-β-gal staining relative to untreated control. (d) D+Q exposure resulted in an increased number of apoptotic cells in Sen-CPC cultures after 16h. All data are Means±SEM, n=2 independent experiments *P<0.0001 by two-way ANOVA.