.
F U L L T E X T S O U R C E : Aging Cell
Abstract
Extracellular vesicles (EVs) have emerged as important regulators of inter‐cellular and inter‐organ communication, in part via the transfer of their cargo to recipient cells. Although circulating EVs have been previously studied as biomarkers of aging, how circulating EVs change with age and the underlying mechanisms that contribute to these changes are poorly understood. Here, we demonstrate that aging has a profound effect on the circulating EV pool, as evidenced by changes in concentration, size, and cargo. Aging also alters particle function; treatment of cells with EV fractions isolated from old plasma reduces macrophage responses to lipopolysaccharide, increases phagocytosis, and reduces endothelial cell responses to vascular endothelial growth factor compared to cells treated with young EV fractions. Depletion studies indicate that CD63+ particles mediate these effects. Treatment of macrophages with EV‐like particles revealed that old particles increased the expression of EV miRNAs in recipient cells. Transfection of cells with microRNA mimics recapitulated some of the effects seen with old EV‐like particles. Investigation into the underlying mechanisms using bone marrow transplant studies revealed circulating cell age does not substantially affect the expression of aging‐associated circulating EV miRNAs in old mice. Instead, we show that cellular senescence contributes to changes in particle cargo and function. Notably, senolytic treatment of old mice shifted plasma particle cargo and function toward that of a younger phenotype. Collectively, these results demonstrate that senescent cells contribute to changes in plasma EVs with age and suggest a new mechanism by which senescent cells can affect cellular functions throughout the body.
1 INTRODUCTION
Aging is associated with a decline in the function of a number of organ systems which ultimately contributes to increased risk of developing age‐related diseases (Lopez‐Otin, Blasco, Partridge, Serrano, & Kroemer, 2013). Although much attention has been given to how aging affects tissue structure and function, less is known about how aging affects circulating factors. Investigation of circulating factors has been primarily focused on biomarker discovery for disease diagnosis and risk stratification. However, more recent studies have demonstrated that circulating factors that change with age can affect tissue homeostasis. In a landmark study, Conboy and colleagues demonstrated that exposure of aged mice to young circulating factors improved skeletal muscle progenitor cell function (Conboy et al., 2005). Since this initial discovery, there have been a number of studies demonstrating the ability of young circulating factors to improve the function of aged tissues (Conboy & Rando, 2012). However, there are also changes in the aged host's circulation which adversely affect tissue function/repair. For example, Rebo and colleagues demonstrated that a single blood exchange between young and old mice led to rapid inhibition of skeletal muscle regeneration in young mice (Rebo et al., 2016). Therefore, there is a need to better define the factors that change in the aged circulation to understand the pathophysiology of aging.
Extracellular vesicle (EV) is a broad term used to describe membrane encapsulated vesicles that range in size from 30 to 1,000 nm and arise from different modes of secretion. Extracellular vesicles are secreted by cell types throughout the body into the extracellular environment or bodily fluids such as the blood and urine where they carry a number of factors including microRNAs (miRNAs) and proteins (Alibhai, Tobin, Yeganeh, Weisel, & Li, 2018; Yanez‐Mo et al., 2015). Changes in plasma EV content reflect changes in cellular function in a number of disease states. For example, changes in circulating EV cargo have been observed in diabetes, postmyocardial infarction, and in cancer (Jansen, Nickenig, & Werner, 2017; Schwarzenbach, 2015). Extracellular vesicles also play a role in inter‐cellular communication as they are capable of transferring their cargo to influence recipient cell function. Given their potential for regulating cellular function, EVs have been suggested to be key mediators of aging (Robbins, 2017; Takasugi, 2018). Despite this interest, how plasma EVs change with age and the underlying mechanisms that contribute to these changes are poorly understood.
Here, we investigate the changes that occur in plasma EVs during aging. We examine how aging affects plasma EV concentration, size, cargo, and function. Using bone marrow transplant experiments, we investigate the role of aging circulating cells in regulating plasma EV miRNA expression. Furthermore, using in vitro and in vivo models of senescence as well as senolytic treatment of aged mice we examine the contribution of senescent cells to plasma EVs. Our study suggests a key role of cellular senescence in regulating circulating EV miRNA cargo and function in aged mice.
2 RESULTS
2.1 Aging affects plasma EV concentration and size
Plasma EVs were isolated from young (3 month) and old (18‐21 month) mice using size exclusion chromatography (SEC) which efficiently separates plasma particles from plasma proteins (Figure 1a). Based on the peak particle count and separation from plasma protein, fractions 7–10 were collected for experiments. Examination of isolated particles by transmission electron microscopy (TEM) revealed particles with diverse sizes (<300 nm in diameter) and morphologies, including the expected “cup‐shaped” morphology in both preparations (Figure 1b). Quantification of particle size from TEM images revealed a greater number of smaller particles in old versus young EV fractions (Figure 1c). To further quantify changes in particle size and concentration, we used nanoparticle tracking analysis (NTA) which measures the rate of particle Brownian motion in solution to determine size and uses the number of particles tracked to determine concentration. Nanoparticle tracking analysis revealed a significant reduction in plasma particle concentration and smaller mean particle size in old versus young EV fractions (Figure S1a–c). A decline in particle concentration with aging has been previously shown with human plasma using a precipitation‐based method (Eitan et al., 2017); thus, we next assessed whether old murine plasma also shows reduced particle count using this approach. Lower particle count in old versus young plasma was also observed using the precipitation method; however, the amount of co‐isolated protein was substantially higher compared to fractions collected using SEC, suggesting reduced purity (Figure S1e–i). Thus, we used SEC‐purified vesicles for this study.
Figure 1. Aging alters plasma EV‐like particle concentration and size. (a) Size exclusion chromatography isolation of plasma EV enriched fractions from young (top) and old (bottom) plasma. (b) Representative transmission electron microscopy (TEM) images of particles isolated from young and old plasma. © Quantification of particle size distribution in 25 nm bins from TEM images, n = 3/group. (d) Representative Western blots images and (e) quantification, n = 4/group, *p < .05. All values are mean ± SEM.
As NTA cannot distinguish EVs from other particles such as lipoproteins, we characterized isolated plasma particles by flow cytometry and Western blotting. Surprisingly, examination of CD63, an EV marker, using flow cytometry revealed significantly increased fluorescence in old versus young EV fractions (Figure S1d). Western blotting similarly demonstrated significantly greater levels of TSG101, CD81, and CD63 in old versus young EV fractions, suggesting increased EV levels in old plasma (Figure 1d,e). Assessment of plasma contaminants revealed that old EV fractions had significantly less APOA1, APOB‐100, and APOB‐48 compared with young fractions (Figure 1d,e). Reduced lipoprotein content may be responsible for the lower particle counts by NTA as chylomicrons which carry APOB represent a substantial number of plasma particles (Sodar et al., 2016). Albumin levels were similar between young and old EV fractions (Figure S1i). Importantly, neither EV fractions contained organelle contaminants indicated by a lack of calnexin signal. Collectively, these data demonstrate that although circulating particle count is lower; EV levels are significantly elevated in old plasma. As characterization of our EV fractions indicate that SEC leads to co‐isolation of plasma factors, the collected fractions used are referred to as EV enriched fractions and particles as EV‐like particles throughout this study.
2.2 Old plasma EVs alter cellular responses to stimuli
Next, we investigated whether aging affects the function of isolated plasma EV enriched fractions. The functional effects of young and old EV fractions were tested using two cell types that interact with circulating EVs, macrophages and endothelial cells. First, we examined how EV fractions affected the expression of activation markers in unstimulated macrophages including polarization markers arginase 1 (Arg1), mannose Receptor C‐Type 1 (MRC1), transforming growth factor β‐1 (Tgfβ1), and inducible nitric oxide synthase (iNOS) as well as cytokine expression. Treatment of young unstimulated macrophages with young or old EV fractions increased the expression of MRC1 and Tgfβ1 compared to PBS‐treated cells (Figure 2a and Figure S2). Although old EV fractions reduced IL1β expression compared to young EVs, the overall trend of basal gene expression was similar between young and old EV‐treated cells and many of these cytokines had low expression in unstimulated cells. Thus, we next investigated whether young or old EV enriched fractions can affect cellular responses to activation. Following stimulation of young macrophages with LPS, old EV‐treated cells had increased Arg1, IL10, MRC1, and Tgfβ1 as well as reduced IL‐6 and iNOS expression compared to PBS‐treated cells (Figure 2b). In contrast, young macrophages treated with young EVs had significantly reduced expression of IL1β and IL12B compared to PBS‐treated cells (Figure 2b and Figure S2). Next, we examined whether EV fractions can also affect old macrophage responses to LPS. Interestingly, the effects on old cells were different than those observed in young cells. In unstimulated cells, old EV fractions reduced the expression of IL‐1β, IL‐12B, and IL‐10 compared to PBS‐treated cells (Figure S3). Young EV fractions also reduced expression of IL‐1β, but did not affect IL‐12B, IL‐10, MRC1 or Tgfβ1 expression. In LPS‐stimulated cells old EVs reduced expression of IL‐1β, IL‐12, IL‐6, and IL‐10 as well as increased Arg1 expression compared to PBS‐treated cells (Figure S3). Young EVs did not affect gene expression in old LPS‐stimulated cells (Figure S3).
.../...
.
