Host pre‐conditioning improves human adipose–derived stem cell transplantation in ageing rats after myocardial infarction: Role of NLRP3 inflammasome

Abstract Functional decline of stem cell transplantation in ageing hosts is well documented. The mechanism for this is poorly understood, although it is known that advancing age does not provide an optimal milieu for exogenous stem cells to survive, engraft and differentiate. We showed that n‐butylidenephthalide improved human adipose–derived stem cell (hADSC) engraftment via attenuating the production of reactive oxygen species (ROS). It remained unclear whether pre‐treated hosts with n‐butylidenephthalide can rejuvenate the ageing heart and improve hADSC engraftment by regulating the ROS/NLRP3 inflammasome‐mediated cardiac fibrosis after myocardial infarction. One hour after coronary ligation, hADSCs were transplanted into the hearts of young and ageing Wistar rats that were pre‐treated with or without n‐butylidenephthalide for 3 days. At day 3 after infarction, myocardial infarction was associated with an increase in ROS levels and NLRP3 inflammasome activity with age. hADSC transplant effectively provided a significant decrease in ROS levels, NLRP3 inflammasome activity, IL‐1β levels and cardiac fibrosis in either young or old infarcted rats. However, the beneficial effects of hADSCs were greater in young compared with old rats in terms of NLRP3 inflammasome activity. The infarcted ageing rats pre‐conditioned by n‐butylidenephthalide improved engraftment and differentiation of hADSCs and additionally attenuated cardiac fibrosis compared with hADSCs alone. The anti‐inflammation effects of n‐butylidenephthalide were reversed by SIN‐1. In conclusions, the increased NLRP3 inflammasome activity plays the pathogenesis of ageing‐related functional hADSC decline in the ageing hosts. n‐butylidenephthalide‐pre‐treated ageing hosts reversibly ameliorate the harsh microenvironments, improve stem cell engraftment and attenuate cardiac fibrosis after myocardial infarction.


| INTRODUC TI ON
Ageing patients have been shown to have dramatically worse cardiovascular outcomes compared with young patients. 1 The advanced age of most patients who experience a myocardial infarction (MI) hinders the healing ability of the infarcted region of the heart. 1 Previous studies have demonstrated an age-dependent reduction in survival and greater cardiac fibrosis following MI. 1 Importantly, the changes seen in the older hosts could be improved with therapy.
Previous studies on whole muscle graft transplantation in mice have shown that the ability to regenerate muscles mainly depends on the host's age and that successful muscle engraftment was seen in young hosts, with failure of graft regeneration in old hosts. 2 Other studies have also shown optimal regeneration in muscle grafts in young hosts. 3 Although the intrinsic regenerative potential of ageing hosts appears to be largely intact, critical factors required for regeneration remained unclear. Therefore, further investigations into the molecular and cellular events post-MI in older patients may improve the therapeutic outcomes in this population.
Cardiac remodelling after MI is a complex inflammatory process involving numerous signalling pathways. 4 An exaggerated inflammatory response is believed to be responsible for the increased morbidity and mortality following MI. Inflammasomes are innate immune signalling pathways and cytosolic multiprotein complexes that are present mainly in macrophages. 5 Nucleotide-binding oligomerization domain-like receptor with a Pyrin domain 3 (NLRP3) inflammasome requires two steps for activation: signal 1 for priming and signal 2 for activation. Priming requires an NF-κB-activating stimulus, such as lipopolysaccharides binding to Toll-like receptor 4, which induces elevated expression of NLRP3 and IL-1β. NF-κB is inactive in the cytosol due to interactions with its inhibitory protein, I-κB. Once activated, NF-κB dissociates from I-κB, resulting in nuclear translocation of the heterodimer p50/p65 where it subsequently binds to decameric consensus sequences in target gene promoter/enhancer regions. After priming, canonical NLRP3 inflammasome activation requires reactive oxygen species (ROS) to activate NLRP3 and lead to the formation of the NLRP3 inflammasome complex. 6 ROS scavenger treatment has been shown to block the activation of NLRP3 in response to a range of agonists. 7 The structures of all inflammasomes are similar, and they are usually formed by an NLRP3, an effector component (caspase-1), an adaptor component (ASC) and a substrate component (such as pro-inflammatory cytokines pro-IL-1β/pro-IL-18). The inflammasomes respond to danger signals such as damage-associated molecular pattern molecules (DAMPs) which are released by stressed or injured cells such as urate crystals, β-amyloid, extracellular ATP and cell debris. DAMPs have been associated with endogenous sterile inflammation after MI, 8 and NLRP3 inhibition via small interfering RNA has been shown to prevent cardiac cell death and the activation of inflammasomes, thereby ameliorating myocardial remodelling. 9,10 Thus, NLRP3 inflammasome inhibition might be a new strategy for sterile MI.
Recently, we showed that human adipose-derived stem cell (hADSC) transplantation is a promising new therapy to improve cardiac fibrosis after MI. 11 Poor graft survival has a significant detrimental effect on the success of cell transplantation. Over 99% of mesenchymal stem cells injected into the left ventricular (LV) myocardium of mice die within four days after injection. 12 An essential challenge would be to determine the signals controlling reduced stem cell survival. Injured myocardium represents harsh environment for the transplanted cells including inflammatory cytokines, acidosis, increased calcium and oxidative stress, hypoxia, or unbalanced supply of nutrition. IL-1β has been shown to play a role in acute inflammation and graft death after direct intramuscular cell transplantation to the heart. 13 Blockade of IL-1β by neutralizing antibody improved stem cell survival and resulted in an increase in the total number of donor-derived cells. 13 Moreover, in mice lacking the IL-1 receptor, markedly decreased adverse remodelling of infarcted hearts has been observed, possibly through a reduction in collagen deposition caused by the lack of IL-1 signalling. 14 Therefore, targeting IL-1 as a way to modulate pathological heart remodelling may be clinically relevant. 15 Angelica sinensis is a well-known traditional Chinese herbal medicine that has been shown to have a broad spectrum of biological activities, including anti-inflammatory properties and the ability to regulate immunity in hepatic fibrosis and cardiovascular diseases. 16 Butylidenephthalide (BP) is the major component of Angelica sinensis. 17 Very recently, we have shown that BP can attenuate superoxide production after MI. 11 Naoxintong, an extract from 16 various kinds of Chinese traditional herbal Medicines including Angelica sinensis, has been shown to provide cardioprotective effect by inhibiting the NLRP3 inflammasome activity. 18 However, whether BP alone inhibited the NLRP3 inflammasome activity remained unknown. Previous studies have shown that ageing repair can be modulated by regulation of many signalling pathways including ROS and NLRP3 inflammasome activity. 19 Donor and recipient ageing has been shown to impair cell engraftment 20 ; however, the host environment was more important in determining the regenerative efficacy than was the age of the donor. 20 The regenerative potential of ageing stem cells can be restored by exposure to serum from young hosts, implying the importance of the host environment in regulating regeneration. 20 Patients who suffer from MI and heart failure are generally older.
Thus, it is important to delineate the role of BP in rejuvenating ageing hosts as defined by morphological and functional characteristics and expression of molecular markers. In this study, we assessed (a) whether there were differential age-related responses of NLRP3 inflammasomes to MI, (b) whether there were different responses between young and ageing infarcted hosts in hADSC administration, and (c) whether pre-treatment with BP in ageing hosts can improve the age-related fibrosis by regulating NLRP3 inflammasomes in a rat MI model.

| ME THODS
The cell culture and animal experiment were approved and conducted in accordance with local institutional guidelines for the care and use of laboratory animals at the China Medical University
These homogeneous hADSCs did not contain haematopoietic lineages or endothelial cells. In addition, they expressed the mesenchymal stem cell marker CD90 but not haematopoietic markers CD31 or CD45, and they were confirmed to be >95% CD90 + and <2% CD31 + /CD45 + .

| Characterization of hADSC surface phenotype
The cells (1 × 10 5 per sample) were treated with the following spe-

| Experiment 1
Myocardial infarction was induced by ligation of the anterior descending artery in young (2-month-old) and old (24-month-old) male Wistar as described previously. 11 To avoid the possibility that the measures were due to senescent decompensation, not ageing compensation, we did not use very old rats. For surgery, haemodynamic measurements and killing, the rats were intraperitoneally anaesthetized with Zoletil (20 mg/kg bodyweight) and xylazine (9 mg/kg). Anaesthesia was tested by reflexes of the hind feet before and during the procedures, monitoring the respiratory patterns, and responses to manipulations during the procedures. The animals were ventilated with 95% O 2 and 5% CO 2 using a ventilator (Harvard Apparatus 486).
One hour after ligation, rats were assigned into groups: young rats (young), young rats receiving hADSC transplantation (young/ ADSC), old rats (old), old rats receiving hADSC transplantation (old/ADSC), and old rats receiving hADSC transplantation and BP pre-treatment (old (BP)/ADSC). For cell transplantation, the hADSCs were detached from the plate, suspended in 30 μL of PBS (1 × 10 6 cells) and transplanted at three injection sites into viable myocardium bordering the infarction using a syringe with a 30-gauge needle. BP (Alfa Aesar) was dissolved in dimethylsulphoxide (Sigma). The oral dose of BP (150 mg/kg/d) was according to previous studies, 21 starting 3 days prior to and ending 3 days after MI. The heart was excised at days 3 or 28 after MI as early and late stages of MI.

| Experiment 2
Given both hADSC and BP provided anti-inflammation effect, we performed additional experiments to dissect the role of BP in regulating host microenvironment without hADSC implantation. To confirm the importance of the ROS signalling in BP-induced NLRP3 inflammasomes, we employed 3-morpholinosydnonimine (SIN-1, a peroxynitrite generator). Immediately after induction of MI by coronary ligation in old male Wistar adult rats (24 months old), infarcted rats were randomized to vehicle, pre-treatment with BP, or a combination of pre-treatment with BP and SIN-1. BP (150 mg/kg/d) was treated 3 days prior to coronary ligation. The dose of SIN-1 (0.1 mg/ kg, single oral dose immediately after coronary ligation) has been shown to be effective in biological activity. 22 The heart was excised one hour after MI. At the end of the study, the ROS levels were measured in all of the hearts (n = 5 in each group) and Western analysis of NF-κB, NLRP3 and IL-1β proteins at the border zone (<2 mm outside the infarct).

| Echocardiogram
Echocardiography was performed at baseline before and then again at 3 and 28 days after surgery. For a detailed information, please refer to the Supplementary material online.

| Haemodynamics and infarct size measurements
At the end of the study, the size of the infarcts and haemodynamic parameters was measured (online Supplementary material). Only rats with clinically important MIs (>30%) were selected for analysis.

| Western blot analysis of p65 NF-κB, NLRP3 and IL-1β
Samples were obtained from the border zone either at day 3 in Experiment 1 or 1 hour in Experiment 2 after MI. Antibodies to p65 NF-κB (Santa Cruz Biotechnology), NLRP3 (Santa-Cruz Biotechnology, sc-34408), cleaved IL-1β (Cell Signaling Technology) and β-actin (Santa Cruz Biotechnology) were used. Western blotting procedures were described previously. 11 Experiments were replicated three times and results expressed as the mean value.

| Quantitative PCR (qPCR) of human Alu, p65
NF-κB, NLRP3 and IL-1β qPCR was performed from samples obtained from the border zone at day 3 as previously described. 11 The gene expression patterns for human Alu, p65 NF-κB, NLRP3 and IL-1β were assessed. The PCR primer sequences are shown in the online data supplement.

| Immunohistochemical analysis of human mitochondria and sarcomeric α-actinin
Immunohistochemistry staining for human mitochondria antibody, sarcomeric α-actinin antibody and 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI) was used to identify and transdifferentiate the transplanted cells at the border zone on day 28. For a detailed information, please refer to the Supplementary material online.

| In situ detection of superoxide anion
To evaluate the production of myocardial intracellular superoxide using in situ dihydroethidium (DHE, 1 µmol/L, Invitrogen Molecular Probes) fluorescence, 5-µm paraffin-embedded tissue sections were incubated with DHE in PBS (10 mmol/L) in a dark, humidified container at room temperature for 30 minutes. Superoxide radical generation in the tissues was assessed by the presence of a red fluorescent signal, and image density was presented as arbitrary units/mm 2 field. 23 DHE fluorescence was digitally recorded using an Olympus microscope. To minimize interference by non-specific DHE oxidation products, a red fluorescence was detected with excitation line 405 nm. 24 The fluorescence intensity of nuclei from each section was measured and was corrected for background fluorescence in nonnuclear regions with Image-Pro Plus software (Media Cybernetics). Four sections per rat were studied.

| Morphometry of cardiac fibrosis
Picrosirius staining and aniline blue staining were used to stain cardiac fibrosis from the remote zone (>2 mm within the infarct) at day 28 after MI. For a detailed information, please refer to the Supplementary material online.

| Laboratory measurements
Serum IL-1β levels at day 3 after MI were assessed using ELISA kits (R&D Systems, Minneapolis, MN, USA) and expressed as pg/mL. Collagen histology was confirmed by hydroxyproline assay using a method adapted from that reported by Stegemann and Stalder. 25 Remote zone samples were immediately frozen in liquid nitrogen and stored at −80°C until the hydroxyproline content per weight of tissue was measured.
Myocardial superoxide production at the remote zone was measured using lucigenin (5 μmol/L bis-N-methylacridinium nitrate, Sigma) enhanced chemiluminescence as previously described. 23 The chemiluminescence in the sample was continuously measured for a total of 300 seconds. The specific chemiluminescence signal was calculated after subtracting background activity and was expressed as counts/min/mg weight (cpm/mg).

| Statistical analysis
The results were presented as mean ± SD. Statistical analysis was performed using SPSS version 19 (SPSS). Differences among the groups of rats were tested using ANOVA. If an F value was significant for the main effect or the interaction, between-group differences were tested using Tukey's multiple range post hoc test. The significant level was assumed at a P value of < 0.05.

| Effect of ageing hosts and BP-preconditioned hosts on hADSC engraftment
To quantify the total number of transplanted hADSCs (surviving and proliferating), the human origin of these transplanted cells was The number of mitochondria-positive cells in old hosts pre-treated with BP was 38 ± 9, significantly higher than that without (20 ± 8, P < .05). These data suggest that BP-primed hosts increase the survival of hADSCs in post-infarct old hearts. qPCR analysis of human Alu sampled from the border zone was performed on day 3 after infarction. The results revealed the presence of human Alu DNA sequences in the myocardium of all hADSC recipients, albeit at low levels ( Figure 1B). The fractions of human Alu genomic DNA relative to total myocardium DNA were significantly higher in the BP-pre-treated group compared to the hADSCs alone.

| Effect of ageing hosts and BP-preconditioned hosts on ROS, NF-κB, NLRP3 inflammasome and IL-1β after hADSC transplantation
Superoxide production was assessed by lucigenin-enhanced chemiluminescence and DHE staining. Compared with young vehicle, the magnitude of increased ROS level was higher in the ageing infarcted hearts ( Figure 2A). Thus, although ageing hearts have similar trends in responses to injury, the responses in ageing hearts were overreactive. Besides, the increased ROS levels were reversed after adding hADSC-treated infarcted group compared with vehicle in ageing rats. BP pre-treatment further decreased ROS levels to 31 ± 11% of old/ADSC (P < .05). DHE staining mirrored the results of lucigeninenhanced chemiluminescence ( Figure 2B). The experiment 2 showed that the attenuated effect of BP on superoxide production was abolished after adding SIN-1 ( Figure 2C).
Then, we evaluated the protein and mRNA expression of NLRP3 inflammasome components by Western blot and qPCR. Compared with young vehicle-treated rats, the protein levels of p65 NF-κB ( Figure 3A), NLRP3 ( Figure 3B) and IL-1β ( Figure 3C) were significantly increased compared with those in the ageing infarcted hearts, indicating that the NLRP3 inflammasome was significantly overactivated in the ageing infarcted hearts. hADSC transplant effectively provided a significant decrease in p65 NF-κB, NLRP3 inflammasome activity, myocardial IL-1β levels in either young or old infarcted rats. The beneficial effects of hADSCs were greater in young (55 ± 8%) compared with old (39 ± 7%, P < .05) rats in terms of inhibiting NLRP3 inflammasome activity. BP primed hosts resulted in a decrease in p65 NF-κB, NLRP3 and IL-1β levels compared with that in hADSCs alone ( Figure 3A-C, all P < .05).
Serum IL-1β levels were verified by ELISA assay (Figure 3D), which was consistent with the Western blotting results. p65 NF-κB, NLRP3 and IL-1β mRNA were significantly increased in old infarcted rats compared to young infarcted rats ( Figure 3E-G). hADSC transplantation showed significant reduction in p65 NF-κB, NLRP3 and IL-1β mRNA in either young or ageing hosts. Besides, pre-conditioning with BP provided additional reduction in these inflammation-related mRNAs.
The protein levels of p65 NF-κB, NLRP3 and IL-1β can be reduced after administering BP as shown in Experiment 2 ( Figure 3H-J), which can be reversed after adding SIN-1.

| Part 2: chronic stage (day 28)
No tumours were seen during necropsy at the end of the study in any of the animals. No significant differences in mortality among the infarcted groups were found throughout the study (Table 1). In addition, there were no significant differences in relative heart weights The values of the right ventricular weight/bodyweight ratio, the lung weight/bodyweight ratio +dp/dt and −dp/dt were significantly improved after administering BP compared with those old infarcted rats treated with hADSCs alone. LV end-systolic pressure, LV enddiastolic pressure and infarct size did not differ among the infarcted groups.

| Effect of ageing hosts and BP-preconditioned hosts on the long-term transdifferentiation
To assess transdifferentiation of hADSC-derived ventricular cardiomyocytes, cardiomyocyte phenotype was confirmed by the coexpression of human mitochondria and sarcomeric α-actinin antibodies. In some engrafted cells, co-localization of human cell marker and cardiomyocyte-specific marker was observed (Figure 4).

| Effects of ageing hosts and BP-preconditioned hosts on cardiac function
To determine whether BP-pre-treated hosts generated an enhanced therapeutic effect of stem cells, we evaluated the effects of hADSC transplantation on cardiac function in post-MI rat hearts (Table 2, Figure 5). LV fractional fraction was significantly decreased in the old infarcted rats compared with young infarcted rats. Compared with young infarcted rats, ageing rats had significant increase in LV end-systolic dimension and LV end-diastolic dimension. Fractional fraction was significantly improved in the hADSC-treated groups with a significant improvement of fractional fraction in the young group compared with the old group. Fractional fraction was further improved in the old rats pre-treated with BP compared with those in hADSCs alone.

| Effects of ageing hosts and BP-preconditioned hosts on remote myocardial fibrosis
Sirius Red staining showed LV fibrosis at the remote zones of the tissue sections ( Figure 6A). Old infarcted rats treated with vehicle had significantly larger areas of intense focal fibrosis compared with young infarcted rats (9.15 ± 0.66% vs. 6.87 ± 0.43%, P < .05).
Compared with hADSCs alone in old infarcted rats, treatment with BP-pre-treated rats further attenuated fibrosis. Hydroxyproline analysis also showed a marked increase in collagen content in the old rats compared to the young rats after MI, and this was significantly attenuated by either hADSC or BP pre-treatment ( Figure 6B). F I G U R E 3 Experiment 1. The changes in NLRP3 inflammasome at day 3 from the border zone. The protein levels of NF-κB (A), NLRP3 (B) and IL-1β (C) assessed by Western blot were significantly increased in old than young rats after infarction. Relative abundance was obtained against that of β-actin. To further confirm serum IL-1β levels, ELISA was performed (D). The mRNA changes in NF-κB (E), NLRP3 (F) and IL-1β (G) were similar to those of proteins.
Results are mean ± SD of 3 independent experiments (n = 5 each group). Experiment 2. Pre-treated hosts with BP provided reduction of these cytokines, which can be reversed by adding SIN-1 (H-J). The number of animals in each group is indicated in parentheses. *P < .05 compared with young; † P < .05 compared with the infarcted groups treated with young/ADSC and old; ‡ P < .05 compared with old/ADSC; § P < .05 compared with the infarcted groups treated with old and old/BP/SIN-1 Note: Values are mean ± SD. Abbreviations as in Table 1.

| Effect of ageing on cardiac NLRP3 inflammasome changes in infarcted rats
Our results showed that more structural and functional changes were observed in ageing rats compared with young rats assessed by inflammation activity at day 3 and cardiac fibrosis at day 28 after MI. We evaluate the ROS/NLRP3 signalling pathway in the ageing rat heart and to explore the potential therapeutic role of this signalling pathway in cardiac rejuvenation and tissue regeneration after myocardial injury. Mitochondrial overproduction of ROS has been shown to contribute to cellular senescence. 26 ROS overproduction ultimately leads to formation of the highly reactive product superoxide anions, whose accumulation and diffusion foster inflammation. Even though various hypotheses have been proposed to explain the ageing process, ROS are still regarded to be the main cause of ageing. 27 Previous studies have shown that an increase in ROS generation is a key factor in regulating the secretion of NLRP3 inflammasome-activated IL-1β in the innate immune system. 28 We found that NLRP3 levels after MI in old rats were significantly increased than those in young rats. Thus, interventions that reduce NLRP3 levels in the old rat heart may confer beneficial effects.
Previous studies have shown that ablation of NLRP3 inflammasome protected mice from age-related increases in systemic inflammation. 29

| Effect of hADSC on engraftment and transdifferentiation in old infarcted rats
Our results showed that hADSC administration in either young or old rats is effective in eliminating inflammation activity at day 3 and cardiac fibrosis at day 28 after MI. However, the extent of the improvement is greater in young compared with old. Cell therapy has shown promise to ameliorate heart failure following MI. However, experimental results of cell therapy for MI have not been as promising in older patients as in younger patients. 30 We have quantitatively determined that recipient age has a dramatic influence on the seeding efficiency of hADSCs in the heart of old rats which is only 40% that measured in young rats ( Figure 1A). Our data suggest that hADSC engraftment and differentiation change as a function of age-related mechanisms. Environmental influences on stem cell function were shown in a heterochronic parabiosis experiment, in which the circulatory system of a young mouse was directly connected to that of an old mouse. 31 The results demonstrated that exposing ageing stem cells to systemic factors derived from the young mouse increased their proliferative capacity and ability to contribute to muscle regeneration after injury. 31 Our results were consistent with the findings of Suzuki et al, 13 showing that the survival of the grafted cells might be reduced by the acute inflammatory response mediated by IL-1β. A previous study reported that skeletal myoblasts expressing an IL-1 inhibitor had a more than sixfold better survival rate after transplantation into infarcted myocardium than control cells at 3 weeks, and that this was associated with reduced fibrosis. 15

| Effect of host pre-conditioning by BP on hADSC engraftment
Our results revealed that the ageing infarcted rats pre-conditioned by BP showed further reduced inflammation activity at day 3 and cardiac fibrosis at day 28 after hADSC administration compared with without BP pre-conditioning. The detrimental effects of inflammaging on implanted stem cells can be pharmacologically re- versed. In the current study, hADSC delivery into BP-pre-treated old hosts decreased active IL-1β by 34% compared with hADSCs alone ( Figure 3C). IL-1β has been shown to play a role in acute inflammation and graft death after direct intramuscular cell transplantation to the heart. 13 Our results showed in experiment 2 that BP pre-treatment before stem cell implantation provided less inflam-  35 Providing the proper environment for stem cell survival and host tissue integration is crucial in myocardial repair strategies.

F I G U R E 7
Schematic representation illustrates the involvements of BP-mediated NLRP3 inflammasomes in cardiac fibrosis in postinfarcted rats. Activation of NLRP3 inflammasome, composed of NLRP3, ASC and pro-caspase-1, is tightly regulated by two-step signals. The first "priming" signal, such as MI-mediated DAMP, enhances the expression of inflammasome components and target proteins via activation of transcription factor NF-κB. The second "activation" signal activates NLRP3 which recruits the ASC scaffolding protein and procaspase-1 allowing for homodimerization and autocatalytic activation of caspase-1. Active caspase-1 cleaves pro-IL-1β into the active isoform. BP inhibits ROS production, which in turn suppresses two-step signals of NLRP3 inflammasome activation. Inhibition of these signalling pathways by their respective inhibitors is indicated by the vertical lines. TLR, Toll-like receptor; DAMPs, danger-associated molecular patterns

| Clinical implications
Considering the high global rate of age-associated heart disease along with the rapidly increasing elderly population, the question of whether cardioprotection by hADSCs is maintained in ageing cohorts is crucial. Yet, the vast majority of research fuelling our understanding of the mechanisms is conducted in young animals. An older age has been reported to be a powerful independent predictor of mortality and morbidity in patients after a MI. 35

| CON CLUS IONS
In the present investigation, we identify ageing-associated NLRP3 inflammasome as the cause of stem cell implantation failure and provide mechanistic insights into its reversal. This study provides evidence that BP can rescue an important cardioprotective pathway, and this may lead to the development of new preventive and therapeutic strategies for age-related cell therapy.

ACK N OWLED G EM ENTS
The data that support the findings of this study are available from the corresponding author upon reasonable request.

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

AUTH O R S' CO NTR I B UTI O N S
TML and HJH designed and performed most of the experiments, analysed and interpreted the data, and wrote the manuscript. TWC, MHC, CHC, PCL and SZL assisted during the acquisition, analysis, and interpretation of data and revised the manuscript. CHC assisted with data analysis and revision of the manuscript. TML is responsible for the integrity of the work as a whole. All authors reviewed and approved the final version of the manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.