Mesenchymal stem cell‐derived extracellular vesicles reduce senescence and extend health span in mouse models of aging

Abstract Aging drives progressive loss of the ability of tissues to recover from stress, partly through loss of somatic stem cell function and increased senescent burden. We demonstrate that bone marrow‐derived mesenchymal stem cells (BM‐MSCs) rapidly senescence and become dysfunctional in culture. Injection of BM‐MSCs from young mice prolonged life span and health span, and conditioned media (CM) from young BM‐MSCs rescued the function of aged stem cells and senescent fibroblasts. Extracellular vesicles (EVs) from young BM‐MSC CM extended life span of Ercc1 −/− mice similarly to injection of young BM‐MSCs. Finally, treatment with EVs from MSCs generated from human ES cells reduced senescence in culture and in vivo, and improved health span. Thus, MSC EVs represent an effective and safe approach for conferring the therapeutic effects of adult stem cells, avoiding the risks of tumor development and donor cell rejection. These results demonstrate that MSC‐derived EVs are highly effective senotherapeutics, slowing the progression of aging, and diseases driven by cellular senescence.


| INTRODUC TI ON
With aging comes an inevitable and progressive loss of the ability of tissues to recover from stress. Consequently, the incidence of chronic degenerative diseases increases exponentially beginning at age 65 and is accompanied by an elevated risk for neurodegenerative diseases, cardiovascular disease, diabetes, osteoarthritis, cancers, and osteoporosis. More than 90% of people over 65 years of age have at least one chronic disease while 75% have two or more comorbidities. Thus, it is imperative to find a way to therapeutically target the cellular processes underlying aging in order to compress the period of functional decline in old age. Such a therapeutic approach would simultaneously prevent, delay, or alleviate multiple diseases of old age.
Senescence is a cell fate that involves loss of proliferative potential of normally replication-competent cells, with associated resistance to cell death through apoptosis, and generally increased metabolic activity. Some senescent cells develop a senescence-associated secretory phenotype (SASP) involving increased secretion of pro-inflammatory cytokines and chemokines, tissue-damaging proteases, and factors that can impact stem and progenitor cell function and growth factors (Tchkonia et al., 2013).

Clearance of senescent cells in the INK-ATTAC and p16-3MR
mouse genetic models or treating mice with novel senolytics extended health span Zhu et al., 2015) and ameliorated the symptoms of numerous pathologies Jeon et al., 2017;Ogrodnik et al., 2017;Roos et al., 2016;Schafer et al., 2017). Thus, the increase in cellular senescence that occurs with aging plays a major role in driving life-limiting, age-related diseases LeBrasseur et al., 2015;Palmer et al., 2015;Tchkonia et al., 2013).
A characteristic of aging is the loss of regenerative capacity, which leads to an impaired ability to respond to stress and therefore increased morbidity and mortality. This has led to the hypothesis that aging is partly driven by the loss of functional adult stem cells necessary for maintenance of tissue homeostasis. Indeed, mice greater than two years of age have a significant reduction in the number and proliferative capacity of various types of adult stem cells.
We previously demonstrated that muscle-derived stem/progenitor cells (MDSPC) are adversely affected upon aging (Lavasani et al., 2012). MDSPCs isolated from old and Ercc1 −/∆ progeroid mice have reduced proliferative capacity and impaired differentiative potential, and this dysfunction directly contributes to agerelated degeneration given that transplantation of young MDSPCs extended health span and life span in ERCC1-deficient progeroid mouse models. Transplanted MDSPCs did not differentiate or migrate from the site of injection, suggesting that the therapeutic effect of MDSPCs was mediated by secreted factors acting systemically (Lavasani et al., 2012). Concordantly, co-culture of young MDSPCs with old MDSPCs resulted in renewal of old MDSPC proliferative and differentiative potential, yet the identification of factors responsible for the rejuvenation of aged MDSPCs remained elusive.
Here, we identified BM-MSCs from young animals, and lineagedirected hESC-derived BM-MSC surrogates, as a novel source of EVs with senotherapeutic activity. We demonstrate that transplantation of BM-MSCs from young, but not old mice, prolonged life span and health span in ERCC1-deficient mice. Further, conditioned media (CM) from young BM-MSCs rescued the function of aged, senescent stem cells and senescent murine embryonic fibroblasts (MEFs) in culture. Moreover, the senotherapeutic activity of CM co-purified with extracellular vesicles (EVs) that were released by young, but not old MSCs and MDSPCs. Importantly, IP injection of EVs from BM-MSCs from young mice extended the life span of ERCC1-deficient mice.
Similarly, treatment with EVs isolated from human embryonic stem cell-derived MSCs (hESC-MSC) was capable of significantly reducing the expression of markers of senescence in cultured senescent fibroblasts as well as naturally aged wild-type and Ercc1 −/∆ mice, and improving measures of healthspan in vivo. These novel results identified EVs as key factors released by young, functional stem cells that can rescue cellular senescence and stem cell dysfunction in culture and reduce senescent cell burden in vivo. Thus, functional stem cellderived EVs represent a novel therapeutic to reduce the senescent cell burden and extend health span.

| Aged murine bone marrow-derived mesenchymal stem cells are dysfunctional
We previously demonstrated MDSPC functional decline with natural and accelerated aging in ERCC1-deficient mice in regard to proliferation and differentiation. To determine if other types of adult stem cells have similar dysfunction, we examined the effect of age on murine bone marrow-derived MSCs. BM-MSC were isolated from K E Y W O R D S aging, extracellular vesicles, mesenchymal stem cells, senescence, stem cells young (3-20 weeks) and old (2 years) WT, and young Ercc1 −/− mice  (Figure 1e). Quantification of SA-ß-gal positive cells at passage 5 showed that both old WT and ERCC1-deficient BM-MSCs senesce more rapidly in culture, with more than 65% of cells staining positive for SA-ß-gal compared to less than 10% with BM-MSC derived from young WT mice (Figure 1f and 1g). RNA-seq analysis of young and old MSCs identified several mRNAs and miRNAs that were differentially expressed and Ingenuity Pathway Analysis (IPA) of the differentially expressed mRNAs identified key pathways affected by age ( Figure S1a). The majority of these pathways also were identified by IPA analysis of the differentially expressed miRNAs ( Figure S1b). MSCs from different mouse models were differentiated to adipocytes using adipogenic media for 21 days. The lipid content was quantified using Nile Red stain and the values normalized to young WT mouse values. Bar graph shows mean ±SD from 3 independent experiments. Significance was determined by one-way ANOVA (p = 0.0024, F(2,11) = 11.01) with Tukey's multiple comparisons test; p value for specific comparisons shown in figure. (d) BM-MSCs from naturally aged and Ercc1 −/− mice have impaired osteogenic potential. MSC from the different mouse models was differentiated to osteocytes using osteogenic media for 21 days. Calcium matrix was stained using Alizarin Red S. Photographs show a differentiation representative of 3 independent experiments. (e) Proliferative potential of old WT and Ercc1 −/− BM-MSCs is reduced compared to WT BM-MSC controls. Significance was determined by one-way ANOVA (p = 0.0038, F(2,8) = 12.11) with Tukey's multiple comparisons test; p values for specific comparisons shown in figure. (f) Quantitation of SAβ-gal staining, and representative brightfield (X-gal) microscopy (g) of BM-MSCs from naturally aged and Ercc1 −/− mice, which undergo senescence earlier than BM-MSCs from young mice. The percentage of senescent cells was manually determined by counting the percent of cells positive for senescence-associated β-galactosidase at passage 5. Bar graph shows mean ±SD from 3 replicate representing 3 independent experiments. Significance was determined by one-way ANOVA (p < 0.0001, F(2,9) = 394.3) with Tukey's multiple comparisons test; p values for specific comparisons shown in figure

| Extracellular vesicles in CM from young MSCs can reduce cellular senescence and improve stem cell function
Numerous studies have documented that the accumulation of senescent cells with age drives age-related pathologies. Compounds that specifically ablate senescent cells (senolytics) or suppress the senescent phenotype (senomorphics) can extend health span in mouse models of accelerated and natural aging (Niedernhofer & Robbins, 2018). To identify senotherapeutic agents, we previously developed a screen for compounds able to suppress senescence in BM-MSCs   These results are consistent with young stem cells secreting factor(s) able to suppress the senescent phenotype and thus functioning as a senomorphic(s).
To determine whether EVs were among the factor(s) secreted by young, but not old MSCs that confer rescue of senescence in MEFs and BM-MSCs, we purified the EV-enriched high molecular weight ( Further, EVs from young MSCs grown at 3% O 2 or briefly exposed to oxidative stress (20% O 2 ) had similar senomorphic activity when applied to aged MSC cultures (Figure 4b). We previously demonstrated that CM from young, but not old MDSPCs can rescue the ability of Ercc1 −/− MDSPCs to proliferate and to differentiate into myoblasts.
EVs isolated from the conditioned media of young MDSPCs were able to reduce senescence and improve differentiation of MDSPCs isolated from the Zmpste24 −/− mouse model of Hutchinson-Gilford progeria syndrome (HGPS), which also rapidly undergo senescence in culture (Figure 4c).

| EVs from hESC-MSCs have senomorphic activity in vitro and regulate expression of genes implicated in aging
Expansion of murine BM-MSC in culture results in higher-passage cells that secrete EVs with diminishing senomorphic potency, limiting the quantity of functional EVs that can be purified for multiple

| EVs from human embryonic stem cell-derived MSCs suppress senescence in vivo and extend health span
To evaluate the penetrance of the senotherapeutic activity of AC83 EVs in vivo, 2-year-old wild-type C57/Bl6 mice (108 weeks) were treated with two IP injections of 5 × 10 9 AC83 EVs, at days 0 and 7, and then sacrificed at day 10 ( Figure S5a). RT-PCR revealed significant decreases in expression of the senescence markers p16 INK4a and p21 Cip1 , and the SASP factors, IL-6, and IL-1β in multiple tissues of treated mice ( Figure 6a). Interestingly, and in concordance with in vitro data, AC83 EVs were also capable of suppressing expression of p53, PTEN, MYC, and IGF-1R in brain, kidney, and lung of treated naturally aged wild-type mice ( Figure S5b), suggesting a similar therapeutic mechanism of action in vitro and in different tissues in vivo.
To further document the suppression of senescence in vivo, AC83 EVs were tested in Ercc1 −/∆ mice carrying a p16 INK4a -luciferase reporter (p16 luc/+ ; Ercc1 −/∆ ) (Robinson et al., 2018). Previous experiments have established that p16 INK4a -luciferase expression increases with accelerated aging in the same tissues as naturally aged mice (Burd et al., 2013). Two IP injections of 10 9 AC83 EVs into Ercc1  Figure S6a) and animals monitored for changes in multiple parameters of health. Progeroid Ercc1 −/∆ mice treated with AC83 EVs scored consistent with a significant improvement in aggregate (percent) symptoms compared to control mice (Figure 6c) and demonstrated a significant reduction in age-related weight loss ( Figure S6). Taken together, these in vivo results suggest that adult stem cell EVs can suppress markers of senescence and extend health span, similarly to the results observed with senotherapeutic compounds.

| DISCUSS ION
Numerous studies suggest that the number and function of diverse   (Zhang, Yeo, et al., 2018), cellular proliferation, enhancement of extracellular matrix and immune function (Zhang, Chuah, et al., 2018), and stem cell regenerative capacity (Chew et al., 2019;Tan et al., 2014;Zhang et al., 2016;Zhang, Chuah, et al., 2018). The mechanism by which EVs suppress senescence in culture and in vivo is unclear, but likely involves the convergent activity of specific miRNAs.
These observations do not preclude the involvement of lipids, proteins, and metabolites in the MSC exosomes (Pathan et al., 2019; as contributors to their senotherapeutic effect, and the magnitude and therapeutic mechanism of EVs may vary depending on the progenitor cell source, as well as their target tissues. For example, preadipocyte-derived EVs carrying eNAMPT have been demonstrated to increase NAD+levels (Yoshida et al., 2019).
Previously, we have used heterochronic parabiosis to demonstrate that circulating factors increase or suppress senescence (Yousefzadeh et al., 2020), results consistent with a role for circu-

| Human embryonic stem cell-derived MSC EV isolation
The hESC-MSC EVs were prepared as previously described . Briefly, immortalized hES-MSCs were grown in a chemically defined medium for 3 days and CM was harvested and pre-cleared with a 0.22 µm syringe filter. The CM was concentrated

| Clonogenic capacity assay
The clonogenic capacity of BM cells was measured at passage 0. Six days after seeding passage 0 cells, the cells were fixed (PFA 2%) and stained with Crystal Violet 0.05% (Sigma-Aldrich) for 30 min and clones counted manually.

| Senescence assay
Cell senescence was assayed by measurement of senescence specific β-galactosidase activity, using both conventional X-gal and fluorescent C 12 FDG as SAβ-galactosidase substrates. For the chromogenic senescence assay, cells were seeded at 5000 cells/cm 2 and treated for 48 hrs with the desired treatment. The cells were fixed (PFA 2%) for 5 min and stained overnight with X-gal (2 mg/ml) staining buffer pH 6.0 (Teknova) at 37ºC. Nuclei were labeled with DAPI 1 µg/ml (Life Technologies) and β-gal positive cells were counted manually. The C 12 FDG-based senescence assay was performed as described Fuhrmann-Stroissnigg et al., 2019).

| Treatment of mice with EVs
Experimental procedures involving animals were performed in strict observance of the regulatory standards outlined in the U.S.

Department of Health and Human Services Guide for the Care and
Use of Laboratory Animals and with the approval of the Institutional Animal Care and Use Committees at the University of Minnesota and Scripps Florida. Wild-type mice were administered 5 × 10 9 AC83 EVs by IP injection, twice as indicated ( Figure S6a). Ercc1 −/− mice were given two IP injections of 10 9 EVs at indicated ages. For health span, Ercc1 −/Δ mice were weighed twice a week and monitored for the onset of age-related symptoms, including dystonia, trembling, ataxia, priapism and urinary incontinence, hind-limb muscle wasting, lethargy, and kyphosis.

| Relative gene expression by RT-PCR
For the senescence markers p16 INK4a , p21 Cip−1 , IL-6, and IL-1β, a panel of published and validated primers for each gene was used to reliably and reproducibly detect changes in gene expression; the primer sequences were previously described (source data file Table S1  RT-PCR data were analyzed using the delta-delta Ct method (Livak & Schmittgen, 2001).

| miRNA Interactome
The miRNAs previously identified in hESC-MSC EVs (Chen et al., 2010) were queried for miRNA-target interaction using the miR-Net collection of validated miRNA targets (Fan & Xia, 2018). An interactome was compiled from raw miRNA-target data using the Cytoscape bioinformatics software platform (Shannon et al., 2003). where df = degrees of freedom and tstat = t statistic. All p values for comparisons are shown in graphs.

ACK N OWLED G M ENTS
The work was supported by NIH grants PO1AG043376 and RO1AR065445 to PDR, LJN, and JH, P01AG062412 and U19AG056278 to PDR and LJN, R56AG4059675 to PDR, and R01AG063543 to LJN.

CO N FLI C T O F I NTE R E S T
SKL is a founder of Paracrine Therapeutics, which develops EVs for therapeutic applications. PDR and LJN are co-founders of NRTK Biosciences, which is developing approaches to reduce the senes-

DATA AVA I L A B I L I T Y S TAT E M E N T
The authors declare that all the data used to generate the findings within this study are furnished within the paper, in the supplementary information, or are available from the corresponding author upon request. A link to the source data file is included with this publication (https://doi.org/10.5281/zenodo.4317941).