Re: Extracellular vesicles.. SASP의 명암!! 2020

[문형철]

건강한 PRP 치료법으로... 

건생병사를 경험시키기!!!

Mech Ageing Dev

. 2020 Jul;189:111263. doi: 10.1016/j.mad.2020.111263

The bright and dark side of extracellular vesicles in the senescence-associated secretory phenotype

Ryan Wallis 1, Hannah Mizen 1, Cleo L Bishop 1,*

• Author information
• Article notes
• Copyright and License information

PMCID: PMC7347005  PMID: 32461143

Highlights

• •
• •
• •
• •

주요 내용

• •세포외 소포체(EVs)는 노화 관련 분비 형질(SASP) 내 주요 매개체입니다.
• •노화 유도 후 EV 생산량이 증가한다는 것이 입증되었습니다.
• •노화 유도 후 EV의 화물(단백질, 핵산, 지질 등)에 변화가 발생한다는 것이 입증되었습니다.
• •EVs는 SASP의 유익한 측면(Bright)과 유해한 측면(Dark) 모두에 기여한다는 것이 입증되었습니다.

Keywords: Ageing, Senescence, Extracellular vesicles, EVs, Inflammation, SASP, miRNA

Abstract

Senescence is a state of proliferative arrest which has been described as a protective mechanism against the malignant transformation of cells. However, senescent cells have also been demonstrated to accumulate with age and to contribute to a variety of age-related pathologies. These pathological effects have been attributed to the acquisition of an enhanced secretory profile geared towards inflammatory molecules and tissue remodelling agents – known as the senescence-associated secretory phenotype (SASP). Whilst the SASP has long been considered to be comprised predominantly of soluble mediators, growing evidence has recently emerged for the role of extracellular vesicles (EVs) as key players within the secretome of senescent cells. This review is intended to consolidate recent evidence for the roles of senescent cell-derived EVs to both the beneficial (Bright) and detrimental (Dark) effects of the SASP.

초록

노화는 세포의 증식 중단 상태로,

세포의 악성 변환을 방지하는 보호 메커니즘으로 설명되어 왔습니다.

그러나

노화 세포는 연령과 함께 축적되며 다양한 연령 관련 질환에 기여하는 것으로 밝혀졌습니다.

이러한 병리적 효과는

염증 분자와 조직 재모델링 인자에 초점을 맞춘 강화된 분비 프로파일의 획득에 기인한다고 알려져 있습니다.

이는 노화 관련 분비 형질(SASP)로 알려져 있습니다.

SASP는 주로 용해성 매개체로 구성되어 있다고 오랫동안 여겨져 왔지만,

최근 연구에서 노화 세포의 분비체 내 핵심 역할을 하는 세포외 소포(EVs)의 역할이 밝혀지고 있습니다.

이 리뷰는 노화 세포 유래 EVs가

SASP의 유익한(Bright) 및 유해한(Dark) 효과에 미치는 역할을

최근 연구 결과를 바탕으로 정리합니다.

1. Introduction

1.1. Senescence and the senescence-associated secretory phenotype (SASP)

Senescence is a stable state of proliferative arrest which occurs in cells exposed to a variety of damaging stressors including telomere shortening, DNA-damaging agents, irradiation, hydrogen peroxide (H2O2), tumour suppressor loss, autophagy impairment and oncogene-expression‎ (Sharpless and Sherr, 2015; Gorgoulis et al., 2019). Senescence is widely considered to be a protective mechanism, which acts to prevent the malignant transformation of cells by instigating a mostly irreversible cell-cycle arrest in response to potentially oncogenic stimuli (Munoz-Espin and Serrano, 2014). A key hallmark of senescence is the acquisition of an enhanced secretory profile through which senescent cells modulate their local microenvironment – the senescence-associated secretory phenotype (SASP) (Coppe et al., 2008; Coppé et al., 2010). One key physiological role of the SASP is the removal of senescent cells, via immune recruitment, following senescence induction (Xue et al., 2007; Kang et al., 2011; Munoz-Espin and Serrano, 2014; Yun et al., 2015). Another, is the contribution of the SASP to optimal wound healing following injury (Demaria et al., 2014; Baker et al., 2016). However, whilst these physiological roles may be considered the “bright” side of the SASP it has also been demonstrated to have “dark” pathological effects during ageing (Coppé et al., 2010). Senescent cells have been demonstrated to accumulate with age (possibly due to immune impairment) and, through the SASP, are implicated in a variety of age-related pathologies including osteoarthritis, atherosclerosis, diabetes, kidney dysfunction, neurodegeneration and vision loss (Baker et al., 2016; Childs et al., 2016; Baar et al., 2017; Ogrodnik et al., 2017; Xu et al., 2018). A key pathogenesis which links these diseases is the maintenance of a persistent state of chronic sterile inflammation (known as inflammageing) to which the SASP has been proposed as a major contributor (Franceschi and Campisi, 2014). Furthermore, depending on the setting, the SASP has been demonstrated to be both pro- and anti- tumorigenic (Bavik et al., 2006; Coppe et al., 2008; Coppé et al., 2010; Acosta et al., 2013). These diverse roles are driven by heterogeneity in the composition of the SASP, which occurs as a result of both cell type and senescence trigger, giving the SASP an intensely context dependent role (Gorgoulis et al., 2019). Furthermore, the SASP has been demonstrated to be a dynamic phenotype, developing during the course of senescence induction (Hoare et al., 2016). These changes are driven by the transcription factors NF-κB and C/EBPβ, which have been identified as key determinants of SASP composition (Acosta et al., 2008; Kuilman et al., 2008; Rodier et al., 2009; Chien et al., 2011). To date, the functions of the SASP have been predominantly attributed to soluble proteins (Coppé et al., 2010). However, emerging evidence suggests extracellular vesicles (EVs) may play a key role in the complex and varied roles of the SASP (Takasugi, 2018).

1. 서론

1.1. 노화 및 노화 관련 분비 형질 (SASP)

노화는

텔로미어 단축, DNA 손상 물질, 방사선 조사,

과산화수소(H2O2), 종양 억제 유전자의 상실,

자가포식 장애 및 종양 유전자 발현 등

다양한 유해한 스트레스 요인에 노출된 세포에서 발생하는

증식 정지 상태입니다 (Sharpless 및 Sherr, 2015; Gorgoulis 외, 2019).

telomere shortening,

DNA-damaging agents,

irradiation,

hydrogen peroxide (H2O2),

tumour suppressor loss,

autophagy impairment and oncogene-expression‎

노화는

잠재적으로 종양 유발 자극에 대응하여

주로 역전 불가능한 세포 주기 정지를 유발함으로써

세포의 악성 변형을 방지하는 보호 메커니즘으로 널리 인식됩니다(Munoz-Espin and Serrano, 2014).

노화의 주요 특징 중 하나는

노화 세포가 국소 미세환경을 조절하기 위해 강화된 분비 프로필을 획득하는 것으로,

이를 노화 관련 분비 형질(SASP)이라고 합니다(Coppe et al., 2008; Coppé et al., 2010).

SASP의 주요 생리적 역할 중 하나는

노화 유도 후 면역 세포의 모집을 통해 노화 세포를 제거하는 것입니다

(Xue et al., 2007; Kang et al., 2011; Munoz-Espin and Serrano, 2014; Yun et al., 2015).

또 다른 역할은

상처 치유 후 최적의 치유에 기여하는 것입니다(Demaria et al., 2014; Baker et al., 2016).

그러나 이러한 생리적 역할이 SASP의 '밝은' 측면으로 간주될 수 있지만,

노화 과정에서 '어두운' 병리적 효과를 갖는다는 것도 입증되었습니다(Coppé et al., 2010).

노화 세포는

연령과 함께 축적되며(면역 기능 저하 때문일 수 있음)

SASP를 통해 골관절염, 동맥경화증, 당뇨병, 신장 기능 장애, 신경퇴행성 질환, 시력 상실 등

다양한 노화 관련 질환과 연관되어 있습니다

(Baker et al., 2016; Childs et al., 2016; Baar et al., 2017; Ogrodnik et al., 2017; Xu et al., 2018).

이러한 질환을 연결하는 주요 병리학적 메커니즘은

만성 무균성 염증의 지속적 상태(inflammageing으로 알려져 있음)이며,

SASP는 이 상태의 주요 기여 요인으로 제안되었습니다(Franceschi and Campisi, 2014).

또한

환경에 따라 SASP는

종양 발생 촉진 및 억제 양면성을 보입니다

(Bavik et al., 2006; Coppe et al., 2008; Coppé et al., 2010; Acosta et al., 2013).

세포 노화는 세포 주기 탈출의 안정적인 형태입니다. 노화된 세포는 더 이상 분열하지 않지만, 대사적으로 활성 상태를 유지하며 다양한 단백질을 분비합니다. 이 단백질들은 노화 관련 분비 형질(SASP)로 통칭됩니다. 노화 관련 세포 주기 탈출은 아마도 항종양 메커니즘으로 진화했을 가능성이 있지만, SASP는 항종양 및 종양 유발 잠재력을 모두 포함합니다. 다루는 내용: 이 리뷰에서는 노화 세포의 발견과 암 및 노화와의 관계를 간략히 논의합니다. 또한 노화 자극이 시작될 때 SASP의 발현과 조절 메커니즘을 설명합니다. SASP의 암 유발성과 암 억제성 특성에 초점을 맞춥니다. 마지막으로, 치료로 유도된 노화와 선택적 SASP 억제를 결합한 암 치료법의 잠재적 이점에 대해 추론합니다. 전문가 의견: 특정 맥락에서 종양 촉진적 및 종양 억제적 SASP 인자의 추가적인 식별과 특성화는 암의 성공적인 치료를 위한 맞춤형 치료법 개발에 필수적일 것입니다.

이러한 다양한 역할은

세포 유형과 노화 유발 요인에 따라 달라지는 SASP의 구성 이질성에서 기인하며,

이는 SASP에 강하게 맥락에 의존적인 역할을 부여합니다(Gorgoulis et al., 2019).

또한

SASP는 노화 유도 과정에서 발달하는

동적 표현형으로 입증되었습니다(Hoare et al., 2016).

이러한 변화는

SASP 구성의 핵심 결정 요인으로 식별된

전사 인자 NF-κB와 C/EBPβ에 의해 주도됩니다

(Acosta et al., 2008; Kuilman et al., 2008; Rodier et al., 2009; Chien et al., 2011).

현재까지 SASP의 기능은

주로 용해성 단백질에 귀인되어 왔습니다(Coppé et al., 2010).

그러나 최근 연구 결과는

세포외 소포(EVs)가

SASP의 복잡하고 다양한 역할에 핵심적인 역할을 할 수 있음을 시사합니다(Takasugi, 2018).

2. Biogenesis and sub-classes of extracellular vesicles (EVs)

Following their discovery in 1983, EVs were primarily thought to be a route for the removal of harmful cellular waste (Pan and Johnstone, 1983; Rashed et al., 2017). However, increasingly, their role in cellular communication is becoming appreciated, particularly since the discovery that EVs can deliver functional RNA to recipient cells (Valadi et al., 2007; Edgar et al., 2016). Generally defined as lipid bilayer bound vesicles secreted by cells, EVs consist of a highly heterogeneous population that differs in size, function and biogenesis. This heterogeneity is reflected in the varied nomenclature that has been used to describe EV sub-types. However, efforts have recently been made to reach consensus when defining EV populations (Lotvall et al., 2014; Théry et al., 2018).

Exosomes represent the smallest and most widely investigated subset of EVs and are defined by their size (50−150 nm) and endosomal origin (Colombo et al., 2014; Edgar et al., 2016). The budding of endosomal membranes leads to formation of multivesicular bodies (MVBs) containing intraluminal vesicles (ILVs) (Fig. 1) (Gould and Raposo, 2013). The endosomal sorting complex required for transport (ESCRT) is key to this process, mediating the recruitment of cargos, budding and eventual cleavage processes of the endosomal membrane (van Niel et al., 2018). The ESCRT machinery is subdivided into four domains which are utilised in a stepwise manner during MVB formation. Initially, ESCRT-0 and -I subunits, such as TSG101 (tumour susceptibility gene 101), and accessory proteins, such as ALIX (ALG-2-interacting protein X) and syntenin, cluster transmembrane and cytoplasmic cargos at microdomains on the MVB membrane. This is followed by recruitment of ESCRT-II and -III and associated proteins which perform membrane budding and scission to release intraluminal vesicles (ILVs) (Colombo et al., 2013). Components involved in MVB formation, such as TSG101 and ALIX, are considered markers of exosomal biogenesis (Tamai et al., 2010; Baietti et al., 2012).

2. 세포외 소포(EVs)의 생성과 하위 분류

1983년 발견 이후,

EVs는 주로 유해한 세포 폐기물을 제거하는 경로로 여겨져 왔습니다(Pan and Johnstone, 1983; Rashed et al., 2017).

그러나 최근에는

세포 간 통신에서의 역할이 점점 더 인정받고 있으며,

특히 EVs가 수신 세포에 기능적 RNA를 전달할 수 있다는 발견 이후(Valadi et al., 2007; Edgar et al., 2016)

더욱 주목받고 있습니다.

세포에 의해 분비되는 지질 이중층으로 둘러싸인 소포로 일반적으로 정의되는 EV는

크기, 기능, 생합성 과정에 따라 매우 이질적인 집합체를 이룹니다.

이 이질성은 EV 하위 유형을 설명하기 위해 사용된 다양한 명명법에 반영되어 있습니다.

그러나 최근 EV 집합체를 정의할 때 합의에 도달하기 위한 노력이 이루어지고 있습니다(Lotvall et al., 2014; Théry et al., 2018).

엑소좀은

엑소좀의 가장 작고 가장 널리 연구된 하위 집합으로,

크기(50~150 nm)와 내소체 기원(Colombo et al., 2014; Edgar et al., 2016)으로 정의됩니다.

내소체 막의 분열은

내소체 내 소포(ILVs)를 포함하는 다중 소포체(MVBs)의 형성을 유발합니다(그림 1) (Gould and Raposo, 2013).

내소체 수송에 필요한 내소체 분류 복합체(ESCRT)는

이 과정의 핵심으로, 화물의 모집, 분열 및 최종적으로 내소체 막의 분열 과정을 매개합니다(van Niel et al., 2018).

ESCRT 기계장치는 MVB 형성 과정에서 단계적으로 활용되는 네 가지 도메인으로 세분화됩니다.

초기 단계에서 ESCRT-0 및 -I 서브유닛(예: TSG101(종양 감수성 유전자 101))과 보조 단백질(예: ALIX(ALG-2 상호작용 단백질 X) 및 syntenin)은 MVB 막의 미세 영역에서 막을 통과하는 화물과 세포질 화물을 집합시킵니다. 이어 ESCRT-II 및 -III와 연관된 단백질이 모집되어 막 분열과 절단을 통해 내막 소포(ILVs)를 방출합니다(Colombo et al., 2013). MVB 형성에 관여하는 구성 요소인 TSG101과 ALIX는 엑소좀 생성의 표지자로 간주됩니다(Tamai et al., 2010; Baietti et al., 2012).

Fig. 1.

Open in a new tab

Biogenesis pathway of Extracellular Vesicles (EVs).

Microvesicles are formed from outward budding of the plasma membrane of cells encapsulating various cargoes including RNA and proteins from the cytoplasm. Exosome biogenesis pathway: 1. The early endosome repeatedly undergoes inward budding forming intraluminal vesicles (ILVs) using either ESCRT dependent or independent mechanisms. 2. This results in a multivesicular body (MVB). 3. The MVB then fuses with the plasma membrane and ILVs are released into the extracellular space where they classified as exosomes. 4. Alternatively, MVBs fuse with the lysosome allowing ILVs to be broken down.

An ESCRT-independent biogenesis pathway also exists, which may utilise the clustering of cone shaped lipids to trigger spontaneous membrane budding (Tkach and Théry, 2016). This alternative pathway has been demonstrated to produce CD63 enriched ILVs following ESCRT knockdown (Stuffers et al., 2009). CD63 is a tetraspanin which, alongside CD9 and CD81, also features in exosome biogenesis, clustering with other membrane proteins and contributing to the sorting of exosomal cargoes (Buschow et al., 2009; Chairoungdua et al., 2010; Charrin et al., 2014). ILV formation may also be driven through the production of ceramide by neutral sphingomyelinase, which facilitates the required endosomal membrane curvature (van Niel et al., 2018). Importantly, this process may be perturbed through use of the inhibitor GW4869, which has proven a useful experimental tool (Trajkovic et al., 2008). Whilst the process of ILV biogenesis is complex and not yet fully elucidated, ILV formation leads to classification of endosomes as MVBs. MVBs have several possible fates, one of which is fusion with lysosomes for enzymatic degradation and, indeed, in cells with impaired lysosomal function, exosome production has been demonstrated to increase (Eitan et al., 2016; Schorey and Harding, 2016) — potentially acting as an alternative mechanism through which to dispose of harmful or defective proteins. For ILVs to be released from a cell, thus becoming exosomes, the MVB must avoid trafficking to the lysosome in favour of the plasma membrane. The cargo and composition of ILVs plays a key role in their eventual trafficking fate, with the formation of a complex between ALIX, syntenin and syndecans (heparin sulphate proteoglycans) leading exclusively to the release of ILVs as exosomes (Baietti et al., 2012). Furthermore, differential trafficking fates have also been attributed to the cholesterol levels within MVBs, with ILVs containing high cholesterol levels preferentially released as exosomes (Möbius et al., 2003).

The RAB superfamily of small GTPases are also crucial for MVB trafficking to the plasma membrane, with the ubiquitination status of RAB7 believed to be essential in determining lysosome or plasma membrane trafficking (Baietti et al., 2012; Song et al., 2016). RAB27a and RAB27b are also key mediators of exosome release, regulating MVB fusion with the plasma membrane (Ostrowski et al., 2010). Other RAB family members (including RAB11, RAB35 and their effectors) have also been reported to perform similar roles in cells that do not constitutively express RAB27. A comprehensive review of the role of RABs in exosome biogenesis may be found here (Blanc and Vidal, 2018). Once secreted, exosomes can interact with recipient cells through numerous mechanisms, including endocytosis and direct membrane fusion, leading to the intracellular delivery of luminal EV cargoes (Thery et al., 2009; Mulcahy et al., 2014; van Niel et al., 2018).

Other EV sub-populations of note include microvesicles, which bud directly from the plasma membrane (Fig. 1). Whilst not as extensively studied as exosomes, it has been shown that the machinery which drives the biogenesis of exosomes, including ESCRT proteins and tetraspanins, are also crucial for the formation of microvesicles. Due to this, it is difficult to confidently identify isolated EVs as exosomes (Gould and Raposo, 2013; Lotvall et al., 2014; Witwer and Théry, 2019). Therefore, we have decided to use the general term extracellular vesicles (EVs) rather than a specific sub-type of vesicle throughout this review (Witwer et al., 2013).

미세소체는 세포질 내의 RNA와 단백질 등 다양한 화물을 포함하여 세포막의 외부로 돌출되는 과정으로 형성됩니다.

엑소좀 생합성 경로:

1. 초기 엔도좀은 ESCRT 의존적 또는 비의존적 메커니즘을 통해 내부 돌출을 반복적으로 겪으며 내막 소체(ILVs)를 형성합니다. 2

. 이 과정은 다중 소체(MVB)를 생성합니다.

3. MVB는 세포막과 융합되어 ILVs가 세포외 공간으로 방출되며, 이들이 엑소소스로 분류됩니다.

4. 대안적으로, MVB는 리소좀과 융합되어 ILVs가 분해됩니다.

ESCRT 독립적 생합성 경로도 존재하며, 이는 원뿔 모양의 지질 집합을 통해 자발적인 막 돌출을 유발할 수 있습니다(Tkach and Théry, 2016). 이 대체 경로는 ESCRT 억제 후 CD63 풍부한 ILVs를 생성하는 것으로 입증되었습니다(Stuffers et al., 2009). CD63은 CD9 및 CD81과 함께 엑소좀 생성에 관여하는 테트라스판인 단백질로, 다른 막 단백질과 집합하여 엑소좀 화물의 분류에 기여합니다(Buschow et al., 2009; Chairoungdua et al., 2010; Charrin et al., 2014). ILV 형성은 중성 스핑고미엘리나아제에 의한 세라마이드 생산을 통해 촉진될 수 있으며, 이는 필요한 엔도소체 막 곡률을 촉진합니다(van Niel et al., 2018). 중요하게도, 이 과정은 억제제 GW4869를 사용함으로써 방해받을 수 있으며, 이는 유용한 실험 도구로 입증되었습니다(Trajkovic et al., 2008). ILV 생합성 과정은 복잡하며 아직 완전히 규명되지 않았지만, ILV 형성은 엔도소체를 MVBs로 분류하는 근거가 됩니다. MVBs는 여러 가지 가능한 운명을 가질 수 있으며, 그 중 하나는 리소좀과의 융합을 통해 효소적 분해입니다. 실제로 리소좀 기능이 손상된 세포에서 엑소좀 생산이 증가한다는 것이 입증되었습니다(Eitan et al., 2016; Schorey and Harding, 2016) — 이는 유해하거나 결함 있는 단백질을 제거하는 대체 메커니즘으로 작용할 수 있습니다. ILVs가 세포에서 방출되어 엑소좀이 되려면 MVB는 리소좀으로의 운반을 피하고 세포막으로 이동해야 합니다. ILVs의 화물과 구성은 최종 운반 운명에 핵심적인 역할을 하며, ALIX, 시텐틴 및 시덴칸(헤파린 황산 프로테오글리칸) 사이의 복합체 형성은 ILVs가 엑소좀으로만 방출되도록 합니다(Baietti et al., 2012). 또한 MVB 내 콜레스테롤 수준에 따라 운반 운명이 달라지며, 고콜레스테롤을 함유한 ILVs는 주로 엑소좀으로 방출됩니다(Möbius et al., 2003).

RAB 초가족의 소형 GTPase는 MVB의 세포막으로의 수송에 필수적이며, RAB7의 유비퀴틴화 상태가 리소좀 또는 세포막으로의 수송을 결정하는 데 필수적이라고 믿어집니다(Baietti et al., 2012; Song et al., 2016). RAB27a와 RAB27b는 엑소좀 방출의 주요 매개체로, MVB와 세포막의 융합을 조절합니다(Ostrowski et al., 2010). RAB 가족의 다른 구성원(RAB11, RAB35 및 그 효능 인자 포함)도 RAB27을 지속적으로 발현하지 않는 세포에서 유사한 역할을 수행한다는 보고가 있습니다. RAB의 엑소좀 생성에 대한 포괄적인 검토는 여기에서 확인할 수 있습니다(Blanc and Vidal, 2018). 분비된 엑소좀은 내포작용 및 직접적인 막 융합을 포함한 다양한 메커니즘을 통해 수신 세포와 상호작용하며, 이로 인해 내막 EV 화물이 세포 내로 전달됩니다(Thery et al., 2009; Mulcahy et al., 2014; van Niel et al., 2018).

기타 주목할 만한 EV 하위 집단에는 플라즈마 막에서 직접 분비되는 마이크로베시클이 포함됩니다(그림 1). 엑소좀만큼 광범위하게 연구되지는 않았지만, 엑소좀 생성에 관여하는 기계적 장치(ESCRT 단백질 및 테트라스판인)가 마이크로베시클 형성에 필수적이라는 것이 밝혀졌습니다. 이 때문에 고립된 EV를 엑소좀으로 확실히 식별하는 것은 어렵습니다(Gould and Raposo, 2013; Lotvall et al., 2014; Witwer and Théry, 2019). 따라서 본 리뷰에서는 특정 소포 하위 유형 대신 일반적인 용어인 세포외 소포(EVs)를 사용하기로 결정했습니다(Witwer et al., 2013).

3. Changes in extracellular vesicle production and secretion in senescence

Whilst the enhanced secretome of senescent cells has been extensively investigated, changes in EV secretion by senescent cells has only recently received significant research interest (Coppé et al., 2010; Takasugi, 2018). Lehmann et al. (2008) were the first to investigate the association between EVs and senescence, demonstrating that irradiation-induced senescence caused an increase in EV production from human prostate cancer cells (Lehmann et al., 2008). This finding, supported previous work conducted by Yu and colleagues (2006), who demonstrated that an increase in EV production from irradiated non–small cell lung cancer cells was dependent on the p53 targets tumour suppressor-activated pathway 6 (TSAP6) and vacuolar protein sorting-associated protein 32 (VPS32; a key part of the ESCRT complex), although the investigators did not look at senescence specifically (Yu et al., 2006). Furthermore, Lespagnol et al. (2008) demonstrated a p53-dependent increase in EV secretion following DNA damage, which was abrogated in TSAP6-null mice (Lespagnol et al., 2008). Interestingly, p53 was also implicated in the upregulation of RAB27b (a key component of EV secretory pathways) in senescent human SVts8 fibroblasts (Fujii et al., 2006). Following these early investigations, it has recently been demonstrated that normal TIG-3 human fibroblasts produce a greater number of EVs in response to senescence inducing-stimuli including replicative senescence, oncogenic RAS expression‎ and the chemotherapeutic doxorubicin (Takasugi et al., 2017). Subsequent investigations have demonstrated an increase in EV production in models of therapy-induced (Kavanagh et al., 2017), H2O2-induced (Terlecki-Zaniewicz et al., 2018), irradiation-induced (Basisty et al., 2020), oncogene-induced (Borghesan et al., 2019), and replicative senescence (Mensà et al., 2020; Riquelme et al., 2020). Therefore, it appears that, much like the soluble secretome, EV production increases in multiple cell types following a variety of different senescence-inducing stimuli.

The mechanisms which underpin the increased production of EVs in senescence have not been extensively studied. However, recent work by Choi, Kil and Cho, demonstrated that senescent dermal fibroblasts upregulate neutral sphingomyelinase (a key enzyme in ILV biogenesis) which led to a reduction in EV secretion when inhibited (van Niel et al., 2018; Choi et al., 2020). Furthermore, the investigators proposed a link between the dysfunctional lysosomal activity of senescent cells and the increased rate of EV production, as artificially lowering lysosomal pH in senescent fibroblasts perturbed EV production, whilst increasing pH in proliferating fibroblasts increased EV production (Choi et al., 2020). Together, this research implicates dysfunctional lysosomal activity as a major factor affecting the rate of EV production by senescent cells.

4. Changes in extracellular vesicle cargo in senescence

EV cargo is derived from the cell of origin, and, as such, can provide information about the composition of that cell, at the time of EV production (Saleem and Abdel-Mageed, 2015; McBride et al., 2017). However, as described above, EVs are also enriched for a variety of cargoes, some of which are involved in their biogenesis (van Niel et al., 2018). Nevertheless, as EVs derive their cargo directly from a producing cell, changes in cellular content following senescence induction may alter the composition and luminal cargo of EVs derived from those cells. The changes in EV composition have now been studied extensively in a variety of cell types, following a variety of senescence triggers (Kadota et al., 2018; Takasugi, 2018). These are discussed below in the context of protein, nucleic acid and lipid constituents.

4.1. Proteins

Soluble proteins have been the most widely investigated factors within the SASP with constituents including inflammatory cytokines (e.g. interleukin-1, (IL-1), IL-6), chemokines (e.g. IL-8), growth factors (e.g. epidermal growth factor (EGF)) and proteases (e.g. matrix metalloproteinase-1 (MMP1)) (Coppé et al., 2010). However, recently, characterisation of EVs and their functional protein cargoes have garnered greater attention and revealed a distinct role as regulatory units (Takasugi, 2018; Basisty et al., 2020). Takasugi et al, (2017) performed proteomic analysis of EVs following senescence induction using doxorubicin (Takasugi et al., 2017). The investigators demonstrated that EVs isolated from senescent cells promoted the proliferation of several types of cancer cell, whilst senescent cell derived conditioned media depleted of EVs attenuated a pro-proliferative effect on MCF-7 breast cancer cells. The tyrosine kinase receptor ephrin receptor A2 (EphA2) was identified as being upregulated in EVs from doxorubicin-induced senescent cells and established as a key functional cargo for the pro-proliferative effects of the EVs. Strikingly, this role of EV EphA2 could not be recapitulated using the soluble form of the receptor. This phenomenon, which suggests that EV derived cargo may be more potent than the soluble form of the same protein, has previously been reported for the soluble factors interferon gamma (IFN-γ) (Cossetti et al., 2014) and transforming growth factor beta (TGF-β) (Webber et al., 2015). This not only emphasises the distinction between EVs and the soluble SASP but also raises the intriguing possibility that EVs may play a role in enhancing the signalling of soluble factors.

EVs have also been derived from cancer cells which have undergone therapy-induced senescence (TIS) (Kavanagh et al., 2017). Investigators identified changes in key proteins involved in cell proliferation, ATP depletion, apoptosis and the SASP. It was suggested that EVs may represent a route through which TIS cancer cells dispose of these proteins and thus contribute to cancer cell survival and resistance to chemotherapy.

One of the key roles of the SASP is the ability to confer paracrine senescence on neighbouring cells (Acosta et al., 2013). A recent study by Borghesan et al., demonstrated that EVs derived from oncogenic RAS-induced fibroblasts can contribute to this effect (Borghesan et al., 2019). Inhibition of EV production was sufficient to attenuate the paracrine senescence effect of the SASP. Furthermore, the investigators demonstrated that EV paracrine senescence was partially mediated through interferon-induced transmembrane protein 3 (IFTM3). These findings collectively highlight the potential of EVs as mediators of cellular communication between senescent cells and the local microenvironment.

Recently, efforts have been made to fully characterise the secretome of senescent cells by generating a so called “Atlas” of SASP and EV factors in different cell types and senescence models (Basisty et al., 2020). Investigators demonstrated little overlap in protein composition between these two components of the senescent secretome. Therefore, from both a functional and compositional perspective, EVs appear to have a distinct role as regulatory mediators beyond the traditional soluble factors which have previously been considered as the main players within the SASP.

4.2. Nucleic acids

Functional transfer of nuclei acids between cells via EVs was first demonstrated in 2007 when Valadi et al. demonstrated RNA transfer from mouse to human mast cells via EVs (Valadi et al., 2007). This finding led to an increase in research interest surrounding EVs as a mechanism for cellular communication, with miRNA in senescent cell derived EVs first being investigated in 2015 (Weiner-Gorzel et al., 2015). The potential of EVs to deliver nucleic acid cargoes, such as upregulated miRNAs, from senescent cells, to the cytoplasm of proliferative cells, adds an additional facet to the SASP, as recently demonstrated with an assortment of miRNAs triggering senescence induction in previously proliferating cells (Overhoff et al., 2014; Luo et al., 2015; Xu et al., 2019). miRNAs are the most widely investigated nucleic acid cargo of EVs and are, therefore, the most extensively discussed in this review. However, a variety of additional nucleic acids including DNA, Telomeric repeat-containing RNA (TERRA) and mitochondrial DNA (mtDNA) have been identified in senescent cell derived EVs.

4.2.1. miRNA

Early research investigating the association between senescence, EVs and miRNAs demonstrated that senescence induction via overexpression‎ of miR-433 in A2780 ovarian cancer cells also resulted in the packaging of miR-433 into EVs. Through this miRNA cargo, the EVs were implicated as potential contributors to a paracrine senescence response in recipient cells (Weiner-Gorzel et al., 2015). Recently, investigations by Fulzele et al., demonstrated that the serum of aged mice contained EVs with increased levels of miR-34a. Upregulation of miR-34a was also detected in muscle derived EVs present in the serum of aged mice and in EVs derived from human myotubes undergoing oxidative stress from H2O2. Furthermore, EVs isolated from H2O2 treated mouse myoblasts, or mouse myoblasts overexpressing miR-34a, were found to contain increased levels of miR-34a. The latter two of these EV populations caused a paracrine senescence effect when cultured with bone marrow stem cells in vitro, compared to EVs from untreated mouse myoblasts. Of note, miR-34a has been shown by others to indirectly regulate p53 through the degradation of the p53 suppressor Sirtuin 1 (SIRT1), which can result in senescence induction (Yamakuchi and Lowenstein, 2009). Fulzele et al. further demonstrated that EVs could be detected in multiple tissues, including the fore- and hind-limb, following tail-vein injection of EVs derived from miR-34a overexpressing mouse myoblasts into healthy mice. Additionally, bone marrow cells isolated from the limbs of untreated mice were cultured ex vivo with either EVs from mouse myoblasts overexpressing miR-34a or EVs isolated from control cells. Finally, the authors determined that treatment with EVs from mouse myoblasts overexpressing miR-34a resulted in a reduction in SIRT1 at both the mRNA and protein level in bone marrow cells, although the functional consequences of this reduction were not explored (Fulzele et al., 2019). In combination with SIRT1, DNA methyltransferase 1 (DNMT1) acts to ensure genomic integrity and exert pro-longevity effects. Mensa et. al., explored this by screening replicatively senescent human umbilical vein endothelial cells (HUVECs) and their small EVs for miRNAs that target SIRT1 and/ or DNMT1. Several miRNAs were identified including miR-21-5p and miR-217, which the authors demonstrated were upregulated in replicatively senescent human aortic endothelial cells (HAECs) and their EVs. This was also observed in a model of drug-induced senescence established in both HUVECs and HAECs. Senescent EVs were able to transfer miR-21-5p and miR-217 to recipient cells, which resulted in reduced expression‎ of both SIRT1 and DNMT1, leading to a decrease in cell proliferation and an increase in senescent cell markers, including p16, IL-6 and IL-8 mRNA (Mensà et al., 2020). Therefore, EV derived miRNA cargo from aged cells may constitute a potential “dark” pathological role of EVs by facilitating the propagation of senescence between tissues, raising the possibility that similar interactions could be driven by a diverse set of miRNAs and cell types in a context dependent manner.

Wound healing has often been reported as one of the “bright” beneficial effects of the SASP (Demaria et al., 2014) and, interestingly, senescent human dermal fibroblast derived EVs containing miR-23a-3p have been shown to play a role in wound healing through the transfer of this miRNA to non-senescent keratinocytes, resulting in more efficient wound healing (Terlecki-Zaniewicz et al., 2019). The miRNA profile of EVs derived from H2O2-induced senescent human fibroblasts was altered when compared to EVs from quiescent fibroblasts and, intriguingly, developed with time after senescence induction. Furthermore, the authors identified miRNAs that are specifically packaged into EVs for secretion from the senescent cell, including miR-15b-5p and miR-29c-3p, and miRNAs that are specifically retained by senescent cells, such as miR-409-5p and miR-323a-3p. The top 20 most secreted miRNAs were predicted to target transcription factors that are known pro-apoptotic mediators, suggesting a role for EV-packaged miRNAs as anti-apoptotic members of the SASP. The authors confirmed this potential role by demonstrating that EVs from H2O2-induced senescent human fibroblasts reduced apoptosis in recipient fibroblast cells undergoing acute stress from high H2O2 levels, compared to EVs from quiescent control fibroblasts (Terlecki-Zaniewicz et al., 2018). Therefore, EV carried miRNAs represent underappreciated yet important players within both the “bright” and “dark” sides of the SASP.

4.2.2. Telomeric repeat-containing RNA (TERRA) and telomeric DNA

Telomere shortening – a key facet of replicative senescence – increases the expression‎ of Telomeric repeat-containing RNA (TERRA) within the cell cytoplasm (Cusanelli et al., 2013; Takasugi, 2018); TERRA has been identified in EVs and has been demonstrated to induce inflammatory gene expression‎ in recipient cells (Wang et al., 2015b; Wang and Lieberman, 2016). However, a link to increased TERRA in senescent EVs has not yet been established. Contrasting the effects of TERRA, cytoplasmic telomeric DNA has been shown to have an anti-inflammatory role, as reviewed previously (Bonafè et al., 2020). The authors speculate that telomeric DNA fragments could be packaged into EVs and exert anti-inflammatory benefits on recipient cells in order to delay age related pathologies.

4.2.3. DNA

The presence of cytoplasmic chromatin DNA fragments (CCFs) within senescent cells has been recently established (Adams et al., 2013), with CCFs potentially acting as indirect regulators of the SASP through the cGAS-STING pathway. In brief, cGAS directly binds CCFs and subsequently synthesises cyclic GMP-AMP (cGAMP), which in turn activates the endoplasmic reticulum adaptor STING, leading to activation of crucial SASP transcription factors including NF-κB. (Dou et al., 2017; Glück et al., 2017). Chromosomal DNA secretion from cells within EVs has been demonstrated to increase with senescence, and appears to maintain cellular homeostasis by removing potentially harmful aberrant DNA, thus avoiding a ROS dependent DNA damage response (Takasugi et al., 2017). This principle was validated by the inhibition of EV section through both siRNA knockdown of ALIX or RAB27a, or through use of the EV secretion inhibitor GW4869, which led to an increased expression‎ of the DNA damage marker γ-H2AX and phosphorylation of key DNA damage response kinases – ATM and ATR (Takahashi et al., 2017; Takasugi et al., 2017). Inhibition of CCFs excretion in EVs also resulted in cells becoming prone to apoptosis. This pathway implicates EV secretion as a homeostatic mechanism, returning to the concept of “waste disposal” first suggested decades earlier (Johnstone, 1992). Interestingly, CCF containing vesicles could elicit a DNA damage response in recipient TIG-3 fibroblasts suggesting that, whilst this mechanism may be beneficial to the producing cell, it may have detrimental consequences in the surrounding microenvironment (Takahashi et al., 2017). This is reminiscent of the SASP as a whole and suggests that senescent cell derived EVs and their cargo may contribute to both sides of the bright and dark dichotomy of the SASP.

4.2.4. Mitochondrial DNA (mtDNA)

Along with cellular senescence, mitochondrial dysfunction has been proposed as one of nine key “pillars” of ageing (López-Otín et al., 2013). Oxidative stress is a key trigger of senescence and excessive reactive oxygen species (ROS) production has been implicated in the protein damage considered to be a hallmark of senescence and ageing (Kaushik and Cuervo, 2015; Höhn et al., 2017). Accumulation of aberrant cellular products due to mitochondrial dysfunction underpins the “garbage catastrophe” theory of ageing, whereby the production of cellular waste products outstrips disposal or recycling mechanisms, thus leading to impairment of homeostasis (Terman et al., 2010; Baixauli et al., 2014; Picca et al., 2019). As discussed above, EVs have long been considered a potential route for cellular waste disposal and it has been proposed that EVs may serve as a reserve mechanism to meet demand where excessive waste production overcomes lysosomal capacity, particularly during oxidative stress (Soubannier et al., 2012; Picca et al., 2019). This is supported by the observable increase in EV secretion following H2O2 induced senescence (Terlecki-Zaniewicz et al., 2018). Furthermore, mitochondrial DNA (mtDNA) has been identified in EVs isolated from astrocytes (Guescini and Genedani, 2010) and myoblasts (Guescini and Guidolin, 2010) suggesting that mitochondrial components may be removed via a vesicular route. Supporting this concept is evidence that oxidative stress can cause astrocytes (Hayakawa et al., 2016) and mesenchymal stem cells (Phinney et al., 2015) to release EVs containing mitochondrial particles (Picca et al., 2019). Therefore, the reprogramming of cellular metabolism in response to oxidative stress-induced senescence may involve the use of EVs to dispose of aberrant mitochondrial components. Parallels to this may be drawn in therapy resistant metastatic breast cancer cells which were demonstrated to contain a complete mitochondrial genome (Soubannier et al., 2012). Furthermore, these cells were able to produce EVs containing mtDNA in response to oxidative stress, contributing to therapeutic resistance (Soubannier et al., 2012). The consequences of these EVs to cells in the surrounding microenvironment is not known and is worthy of further investigation. Overall, EVs represent an underexplored dimension to the metabolic reprogramming which occurs following senescence induction.

4.3. Lipids

A relatively underexplored facet of senescent cell derived EVs is their lipid profile. Several recent publications have identified changes in the lipid profiles of oncogene-induced senescent fibroblast derived EVs. Enrichment was demonstrated in hydroxylated sphingomyelin, lyso- and ether-linked phospholipids, when compared to the producing cell, indicating that the lipid composition of EVs may differ from the parental cells more than previously appreciated (Buratta et al., 2017). Furthermore, a subsequent study identified that EVs from oncogene-induced senescent fibroblasts showed increased levels of monounsaturated and decreased level of saturated fatty acids, as compared to control cells (Sagini et al., 2018). Intriguingly, EVs from bone-derived mesenchymal stem cells have been demonstrated to contribute to paracrine senescence when loaded with very long chain C24:1 ceramide (Khayrullin et al., 2019). The lipid composition of senescent cell derived EVs, therefore, potentially represents an underexplored mechanism through which senescent cells contribute to the paracrine effects of the SASP.

5. Role of extracellular vesicles in ageing

5.1. Changes in EV production with aging

Senescent cells have previously been demonstrated to accumulate with age and at sites of age-related diseases (Nielsen et al., 1999; Krishnamurthy et al., 2004, 2006). Given that, as described above, senescent cells have an increased rate of EV secretion, it might be predicted that EV production would increase with ageing. However, contradictory data has emerged regarding the change in EV production with age. An initial study concluded that, although the soluble SASP component IL-6 increased in aged volunteers, EV concentration did not alter (Alberro et al., 2016). Another suggested that fewer EVs could be isolated from the plasma samples of elderly patients. However, the authors also noted that circulating EVs were increased in individuals that smoked or had higher a body mass index (BMI), both of which are correlated with higher levels of senescent cells in tissues (Nyunoya et al., 2006; Eitan et al., 2017). By contrast, circulating EVs have also been demonstrated to increase in aged subjects (Im et al., 2018) and in mouse models of normal ageing (Alibhai et al., 2020; Qureshi et al., 2020). Aged chondrocytes isolated from arthritic cartilage have also been shown to produce an increased number of EVs. (Jeon et al., 2019). One possible explanation for these conflicting data is a reported increase in clearance of EVs in ageing, which may mask increased production in certain settings (Eitan et al., 2017). Another consideration, is the use of different methods of EV characterisation, with one study demonstrating that whilst the number of circulating particles detected by nanoparticle tracking analysis (NTA) was higher in young (3 month) mice, EV markers were enriched in old (18–21 month) mice as determined by western blot (Alibhai et al., 2020). This demonstrates the importance of utilising multiple analysis approaches when characterising EVs (Lötvall et al., 2014). Further complexity also comes from the suggestion that EV concentration increases with certain age-related diseases (Tan et al., 2017). Finally, EV concentration could be affected by confounding factors including medication used in the treatment of age-related pathologies, as it has been shown that the regular use of aspirin in elderly patients reduced the number of platelet derived EVs without altering the overall number of circulating EVs (Goetzl et al., 2016).

5.2. Changes in EV cargo with age

Whilst the level of EV production in age remains an open question, it is clear from numerous studies that EV composition changes with age. miR-31 has been demonstrated to increase and Galectin-3 determined to decrease, in EVs isolated from elderly patients. These changes resulted in both the impairment of osteogenic differentiation of mesenchymal stem cells (MSCs) and compromised tissue regeneration (Weilner and Keider, 2016; Weilner and Schraml, 2016). Furthermore, EVs from the bone marrow of aged mice have also been demonstrated to reduce osteogenic differentiation, in this case through increased miR-183-5p expression‎, leading to a reduction in bone mineral density and bone function (Davis et al., 2017). The mitochondrial content of immune cell derived EVs was also recently shown to decline with age (Picca et al., 2019; Zhang et al., 2020). Macrophage-derived serum EVs from aged (>65 years) individuals were also shown to contain significantly higher levels of IL-6 and IL-12 mRNA than matched EVs from young (21–45 years) volunteers (Mitsuhashi et al., 2013). The miRNA cargo of EVs has also been implicated as a potential biomarker of pathological aging. Ipson et al., identified 8 miRNAs that are increased in EVs from frail older individuals compared to both robust older and young patients (Ipson et al., 2018). Furthermore, expression‎ of miR-21 has been previously recognised to increase in pathological ageing (Olivieri et al., 2012) and was recently demonstrated to be one of several miRNAs that were upregulated in EVs in both aged mice and an irradiation induced global model of senescence (Alibhai et al., 2020). miR-21-5p has also recently been proposed as a potential marker of senescence and inflammageing (Mensà et al., 2020).

Changes in EV cargoes have also been described in pathologies associated with age. EVs from senescent coronary artery endothelial cells and from patients with acute coronary artery syndrome-induced premature endothelial dysfunction have been demonstrated to propagate a paracrine senescence phenotype (Abbas et al., 2017). Aortic smooth muscle cell calcification may also be facilitated by EVs from senescent endothelial cells, and from the plasma of elderly patients, representing another pathology closely linked to age (Alique et al., 2018). EVs isolated from senescent chondrocytes from patients with osteoarthritis (OA) were also able to induce a paracrine senescence effect in proliferative chondrocytes and inhibited cartilage formation, indicating a role in disease severity (Jeon et al., 2019). Senolytic clearance of senescent cells reduced EV production in OA patients and also altered miRNA cargo composition. Senolytic treatment in aged mice induced primarily immunological changes in EV proteins, and age-related synovial EVs were demonstrated to transfer arthritic disease to young animals, demonstrating a pathological role in OA (Jeon et al., 2019). EVs have also been implicated in the distribution of damaged proteins in neurodegenerative conditions including Alzheimer’s, Huntington’s and Parkinson’s disease (Soria et al., 2017; D’Anca et al., 2019). Furthermore, EV encapsulated miR-21-5p was also found to be upregulated in a range of patients with age-related diseases compared to healthy individuals of the same age or healthy young individuals (Olivieri et al., 2017). Therefore, changes in EV composition have been identified throughout both age and age-related diseases (Urbanelli et al., 2016; Kadota et al., 2018).

5.3. Use of EVs as theraputics in age-related pathologies

EVs are emerging as attractive novel therapeutic agents in a variety of settings and have, to date, proceeded as far as phase II clinical trials (Besse et al., 2016). Strikingly, extracellular nicotinamide phosphoribosyl transferase (eNAMPT) containing EVs were recently demonstrated to extend the health-span and increase the physical activity of elderly mice (Yoshida et al., 2019). A variety of different other therapeutic approaches have also been proposed, including the replacement of cell-based therapies with EV treatments and the use of EVs as drug delivery vectors (Ha et al., 2016; Boulestreau et al., 2020). Of particular interest is the use of EVs as a means of RNA interference technology, through loading with a miRNA or siRNA “payload” which could then be delivered to a recipient cell (Hagiwara et al., 2014). EVs present significant advantages over other more established delivery systems, such as liposomes, due to a low toxicity, presence of membrane proteins and a lack of immunogenic effects (Lener et al., 2015; Ohno et al., 2016). Furthermore, efforts have been made to develop EVs which are specifically targeted to cells or tissues through the display of antibody fragments or functional peptides (Ohno et al., 2013; Longatti et al., 2018; Tian et al., 2018). This has the potential to maximise efficacy whilst limiting off-target effects (Ha et al., 2016). Here, we discuss the use of EV based theraeptuics in ageing.

The loss of tissue regeneration represents a universal characteristic of ageing (Gorgoulis et al., 2019). This is driven by a process of stem cell exhaustion which has been proposed (along with senescence) as one of the hallmarks of ageing (López-Otín et al., 2013). Stem cell based therapies have an established regenerative potential within the tissue microenvironment but are limited by adverse effects such as tumour formation (Ma et al., 2020). Harnessing the secretome of stem cells has been proposed as an alternative therapeutic approach which may overcome this shortcoming and utilising stem cell derived EVs (SC-EVs) has become the subject of significant research interest (Boulestreau et al., 2020). SC-EVs have shown promising results in a number of settings, including the regeneration of liver tissue from a fibrotic state in mice (Li et al., 2013; Hyun et al., 2015), improvement of functional recovery after stroke in rats (Xin et al., 2013), restriction of pulmonary hypertension under hypoxic conditions in mice (Lee et al., 2012), protection of cardiomyocytes from apoptosis in ischemic myocardium (Wang et al., 2015a) and acceleration of cutaneous wound healing (Golchin et al., 2018; Chen et al., 2019). SC-EVs have also been modified with functional peptides to target EVs to specific tissues, of particular note is the use of rabies viral glycoprotein (RVG) tagged SC-EVs to target the central nervous system, for the functional improvement of mice in a model of Alzheimer’s disease (Cui et al., 2019). In other studies, RVG tagged EVs have demonstrated the functional delivery of siRNAs to the brain in several models of neurodegenerative disease, thus highlighting the potential of EVs as nucleic acid delivery vectors (Alvarez-Erviti et al., 2011; Cooper et al., 2014). Following these promising preclinical models the International Society for Extracellular Vesicles (ISEV) have released a positional paper for consideration as EV based therapies move into clinical trials (Lener et al., 2015).

Finally, whilst not a true EV based approach, Muñoz-Espín et al., have developed a synthetic nanoparticle based technology which mirrors many of the advantages of EV based therapeutics (Muñoz-Espín et al., 2018). Here, 100 nm silica beads are coated with a layer of galacto-oligosaccharides (GalNP) and upon uptake by recipient cells, these beads are trafficked to the lysosome and subsequently released by exocytosis. However, in senescent cells, high levels of lysosomal β-galactosidase activity digest the GalNP coating leading to a release of any luminal cargo. Investigators used these beads to selectively clear senescent cells through use of chemotherapeutic cargos in vivo whilst also reducing the toxicities typically associated with those agents (Muñoz-Espín et al., 2018). Together, these studies highlight the exciting potential of both biological and synthetic nanoparticle based therapeutics in ageing.

6. Conclusion

The SASP is a key hallmark of senescence with both beneficial and deleterious effects in homeostasis and pathology. These numerous and varied roles have thus far predominantly focused on the soluble factors produced including cytokines, chemokines, growth factors and tissue remodelling agents (Coppe et al., 2008; Coppé et al., 2010). However, more recent studies have begun to highlight the role of extracellular vesicles within the SASP (Takasugi, 2018). EVs represent a mechanism through which a wide range of different cargoes may engage in intercellular signalling and, in particular, brings the possibility of the functional transfer of nucleic acids between senescent cells and their microenvironment (Terlecki-Zaniewicz et al., 2018; Jeon et al., 2019). As with the soluble SASP, the role of EVs in senescence and ageing appears to have both beneficial and detrimental effects depending on the cellular context. These are outlined in Fig. 2. However, despite these recent advances, the senescence field must now reflect on the importance of research methodology when investigating EVs, an issue which has recently become the subject of focus from the International Society for Extracellular Vesicles (ISEV) (Lötvall et al., 2014; Théry et al., 2018). In particular, care needs to be employed in separating the soluble and vesicular fractions of the SASP, given the propensity for co-isolation of soluble proteins with EVs when employing the widely used ultracentrifugation method of isolation (Foers et al., 2018). Therefore, in order to confidently attribute effects to EVs derived from senescent cells, efforts must be made to maximise isolation rigor and stringency. Nevertheless, EV production and composition have been extensively demonstrated to change in multiple models of senescence. Whilst more work is required to fully elucidate their role in age and age-related pathologies, EVs represent an alternate method for senescent cells to modulate their local micro-environment beyond the traditional, soluble SASP.

Fig. 2.

Open in a new tab

The Bright and Dark Side of Extracellular Vesicles in Senescence and Ageing.

The senescence associated secretory phenotype (SASP) has been demonstrated to have beneficial (Bright, yellow) and detrimental (Dark, grey) roles within both normal homeostatic processes and pathology. These diverse and varied effects are driven by heterogeneity in the composition of the SASP, which is influenced by factors including cell type, senescence trigger and disease state. The SASP is also a variable phenotype, developing over the course of senescence induction, leading to significant alterations in constituent factors. Extracellular vesicles (EVs, pink) have recently emerged as key mediators within the SASP and have also been demonstrated to have a wide range of roles in senescence and ageing. As with the more traditional “soluble” SASP, this vesicular secretome also mediates a heterogeneous set of effects that may be considered “bright” or “dark” depending on the specific cellular context. This is driven by a diverse set of EV cargos including proteins, nucleic acids and lipids, which underpin the functional roles of EVs within senescence and ageing. However, as emerging players within the SASP, further work is required to elucidate the contribution of EVs to both the homeostatic and pathological effects of senescent cells. In particular, advancements in isolation and characterisation methodologies will likely lead to the establishment of further heterogeneity in EV sub-populations, adding further complexity to the role of these vesicles within the secretome of senescent cells. EphA2 (Ephrin-A2), C24:1 ceramide (Long chain C24:1 ceramide), IFITM3 (Interferon-induced transmembrane protein 3), TERRA (Telomeric repeat-containing RNA), CCFs (Cytoplasmic chromatin fragments), miRNAs (microRNAs), mtDNA (Mitochondrial DNA) eNAMPT (extracellular nicotinamide phosphoribosyl transferase) Figure created with BioRender. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

Acknowledgements

Ryan Wallis is funded by the Prof. Derek Willoughby Foundation and The Medical College of Saint Bartholomew’s Hospital Trust. Hannah Mizen is funded by the BBSRC (BB/M009513/1) andUnilever (MA-2016-01914N).

References

1. Abbas M. Endothelial Microparticles from acute coronary syndrome patients induce premature coronary artery endothelial cell aging and thrombogenicity: role of the Ang II/AT1 receptor/NADPH oxidase-mediated activation of MAPKs and PI3-kinase pathways. Circulation. 2017;135(3):280–296. doi: 10.1161/CIRCULATIONAHA.116.017513. [DOI] [PubMed] [Google Scholar]
2. Acosta J.C. Chemokine Signaling via the CXCR2 Receptor Reinforces Senescence. Cell. 2008;133(6):1006–1018. doi: 10.1016/j.cell.2008.03.038. [DOI] [PubMed] [Google Scholar]
3. Acosta J.C. A complex secretory program orchestrated by the inflammasome controls paracrine senescence. Nat. Cell Biol. 2013 doi: 10.1038/ncb2784. [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Adams P.D. Lysosome-mediated processing of chromatin in senescence. J. Cell Biol. 2013;202(1):129–143. doi: 10.1083/jcb.201212110. [DOI] [PMC free article] [PubMed] [Google Scholar]
5. Alberro A. Inflammaging and frailty status do not result in an increased extracellular vesicle concentration in circulation. Int. J. Mol. Sci. 2016;17(7) doi: 10.3390/ijms17071168. MDPI AG. [DOI] [PMC free article] [PubMed] [Google Scholar]
6. Alibhai F.J. Aging Cell. John Wiley & Sons, Ltd; 2020. Cellular Senescence Contributes to Age‐Dependent Changes in Circulating Extracellular Vesicle Cargo and Function. 00, p. e13103. [DOI] [PMC free article] [PubMed] [Google Scholar]
7. Alique M. Senescent microvesicles: a novel advance in molecular mechanisms of atherosclerotic calcification. Int. J. Mol. Sci. 2018;19(7) doi: 10.3390/ijms19072003. Multidisciplinary Digital Publishing Institute (MDPI) [DOI] [PMC free article] [PubMed] [Google Scholar]
8. Alvarez-Erviti L. Delivery of siRNA to the mouse brain by systemic injection of targeted exosomes. Nat. Biotechnol. 2011;29(4):341–345. doi: 10.1038/nbt.1807. [DOI] [PubMed] [Google Scholar]
9. Baar M.P. Targeted Apoptosis of Senescent Cells Restores Tissue Homeostasis in Response to Chemotoxicity and Aging. Cell. Cell Press. 2017;169(1):132–147. doi: 10.1016/j.cell.2017.02.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Baietti M.F. Syndecan–syntenin–ALIX regulates the biogenesis of exosomes. Nat. Cell Biol. 2012;14(7):677–685. doi: 10.1038/ncb2502. [DOI] [PubMed] [Google Scholar]
11. Baixauli F., López-Otín C., Mittelbrunn M. Exosomes and autophagy: coordinated mechanisms for the maintenance of cellular fitness. Front. Immunol. 2014;5(AUG):403. doi: 10.3389/fimmu.2014.00403. [DOI] [PMC free article] [PubMed] [Google Scholar]
12. Baker D.J. Naturally occurring p16Ink4a-positive cells shorten healthy lifespan. Nature. 2016;530(7589):184–189. doi: 10.1038/nature16932http://www.nature.com/nature/journal/v530/n7589/abs/nature16932.html#supplementary-information. Nature Publishing Group, a division of Macmillan Publishers Limited. All Rights Reserved. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Basisty N. A proteomic atlas of senescence-associated secretomes for aging biomarker development. PLoS Biol. 2020;18(1) doi: 10.1371/journal.pbio.3000599. Edited by M. Serrano, p. e3000599. [DOI] [PMC free article] [PubMed] [Google Scholar]