Re: 흉선 노화... 어떻게?

[문형철]

Aging Cell

. 2022 Jul 12;21(8):e13671. doi: 10.1111/acel.13671

Age‐related thymic involution: Mechanisms and functional impact

Zhanfeng Liang 1,2,3,4, Xue Dong 1,2, Zhaoqi Zhang 1,2, Qian Zhang 1,2, Yong Zhao 1,2,3,4,✉

• Author information
• Article notes
• Copyright and License information

PMCID: PMC9381902  PMID: 35822239

Abstract

The thymus is the primary immune organ responsible for generating self‐tolerant and immunocompetent T cells. However, the thymus gradually involutes during early life resulting in declined naïve T‐cell production, a process known as age‐related thymic involution. Thymic involution has many negative impacts on immune function including reduced pathogen resistance, high autoimmunity incidence, and attenuated tumor immunosurveillance. Age‐related thymic involution leads to a gradual reduction in thymic cellularity and thymic stromal microenvironment disruption, including loss of definite cortical‐medullary junctions, reduction of cortical thymic epithelial cells and medullary thymic epithelial cells, fibroblast expansion, and an increase in perivascular space. The compromised thymic microenvironment in aged individuals substantially disturbs thymocyte development and differentiation. Age‐related thymic involution is regulated by many transcription factors, micro RNAs, growth factors, cytokines, and other factors. In this review, we summarize the current understanding of age‐related thymic involution mechanisms and effects.

초록

흉선은 

자기 내성 및 면역 기능이 정상적인 T 세포를 생성하는 주요 면역 기관입니다. 

그러나 

흉선은 생애 초기 단계에서 점차 퇴화되며, 

이 과정에서 미성숙 T 세포의 생산이 감소하는 현상을 

나이 관련 흉선 퇴화라고 합니다. 

흉선 퇴화는 

면역 기능에 부정적인 영향을 미치며, 

병원체 저항력 감소, 

자가면역 질환 발생률 증가, 

종양 면역 감시 기능 약화 등을 포함합니다. 

연령 관련 흉선 퇴화는

티무스 세포 수의 점진적 감소와 티무스 간질 미세환경의 파괴를 초래합니다.

이는 명확한 피질-수질 경계의 상실,

피질 티무스 상피 세포와 수질 티무스 상피 세포의 감소,

섬유모세포의 확장,

혈관 주위 공간의 증가 등을 포함합니다.

노화된 개체에서 손상된 티무스 미세환경은

티무스 세포의 발달과 분화를 크게 방해합니다.

연령 관련 흉선 퇴화는

많은 전사 인자, 마이크로 RNA, 성장 인자, 사이토킨 및 기타 요인에 의해 조절됩니다.

이 리뷰에서는

연령 관련 흉선 퇴화의 메커니즘과 영향에 대한 현재의 이해를 요약합니다.

Keywords: aging, T cells, thymic epithelial cells, thymic involution, thymus

---

Age‐related thymic involution has many negative impacts on immune function including reduced pathogen resistance, high autoimmunity incidence, and attenuated tumor immunosurveillance. This review summarizes the current understanding of how age impacts thymic development and function, as well as the mechanisms underlying age‐related thymic involution, particularly TEC transcriptional profile changes during thymic aging.

연령 관련 흉선 퇴화는 

면역 기능에 부정적인 영향을 미치며, 

이는 병원체 저항력 감소, 자가면역 질환 발생률 증가, 종양 면역 감시 기능 약화 등을 포함합니다. 

이 리뷰는 

연령이 흉선 발달 및 기능에 미치는 영향에 대한 현재의 이해를 요약하며, 

특히 흉선 노화 과정에서 TEC(흉선 상피 세포)의 전사 프로파일 변화와 같은

 연령 관련 흉선 퇴화의 기전도 살펴봅니다.

---

AbbreviationsCR

caloric restriction

cTECs

cortical thymic epithelial cells

DN

double negative

DP

double positive

EMT

epithelial–mesenchymal transition

ETPs

early T‐lineage progenitors

FGF21

fibroblast growth factor 21

GH

growth hormone

HSCs

hematopoietic stem cells

LPCs

lymphohematopoietic progenitor cells

MHCII

major histocompatibility complex class II

mTECs

medullary thymic epithelial cells

OSM

oncostatin M

PPARγ

peroxisome proliferator‐activated receptor gamma

PVS

perivascular space

SCF

stem cell factor

SP

single positive

TCR

T‐cell receptor

TECs

thymic epithelial cells

Tgfbr1

transforming growth factor beta receptor I

TRECs

T‐cell receptor excision circles

1. INTRODUCTION

Age‐related thymic involution is one of the most ubiquitous changes during immune system senescence. Most vertebrates experience thymic involution, which presents as a decrease in thymic epithelial cells (TECs) and accumulation of adipose tissue within the thymus. Human thymic involution is thought to begin as early as 1 year of age (Hale, 2004; Murray et al., 2003; Palmer et al., 2018). The murine thymus reaches its peak mass at 4 weeks, followed by a gradual diminution (Sutherland et al., 2005). Recently, Wu et al. examined medullary thymic epithelial cell (mTEC) transcriptomes from mice at 2, 6, and 10 weeks of age and showed that age‐related TEC degeneration occurs as early as 6 weeks after birth, as evidenced by reduced cell cycle‐related gene expression‎ and increased inflammatory response‐related gene expression‎ (Wu et al., 2018). Because the thymus involutes as early as 4–6 weeks in mice and 1 year in humans, ages when most other organs do not show any signs of aging, thymic involution may be an evolutionarily conserved process (Chaudhry et al., 2016). Some scientists have speculated that thymic involution has an undiscovered biological purpose and have divided thymic involution into an early phase (growth‐dependent thymic involution) and a late phase (age‐dependent thymic involution) (Aw & Palmer, 2012). The early phase (growth‐dependent thymic involution) may be due to a bioenergetic trade‐off (Boehm & Swann, 2013). In early life, it is essential to maintain high thymic activity and produce broad T‐cell receptor (TCR) diversity to protect against infections (Aw & Palmer, 2012). Once the T‐cell repertoire is established, it may be beneficial for the organism to reduce thymic activity and redistribute energy to other organs. In contrast, the late phase (age‐dependent thymic involution) may be similar to the aging process of other organs (Aw & Palmer, 2012). In this review, we mainly discuss age‐dependent thymic involution.

Thymic involution leads to a decline in new naïve T‐cell production and a collapse in peripheral TCR repertoire, resulting in impaired immune function (Figure 1). Thymic involution is associated with increased susceptibility to many diseases, including cancer, infection, and autoimmunity (Fahy et al., 2019; Goronzy & Weyand, 2003). It is well known that aging is associated with increased incidence of infectious diseases and neoplastic diseases, which is commonly attributed to systematic immunosenescence and a gradual accumulation of genetic mutations (Gavazzi & Krause, 2002; Pawelec, 2017). Recently, using a delicate mathematical model, Palmer et al. showed that age‐related decline in T‐cell production caused by thymic involution is a major risk factor for many cancers and infectious diseases in humans (Palmer et al., 2018). A connection between age‐related thymic demise and autoimmunity has been shown in many studies. For example, Hosaka et al. demonstrated that thymus transplantation could correct autoimmune disease in aging MRL/+ mice that exhibit dramatic thymic involution (Hosaka et al., 1996). In addition, some studies suggest that thymic aging might be involved in rheumatoid arthritis progression in humans (Goronzy & Weyand, 2003). Thus, a more complete understanding of the mechanisms and impacts of age‐related thymic involution will help us to better understand and prevent immunosenescence‐associated diseases during aging. In this review, we summarize the current knowledge on age‐related thymic involution mechanisms and effects.

1. 서론

연령 관련 흉선 위축은

면역 체계 노화 과정에서 가장 널리 관찰되는 변화 중 하나입니다.

대부분의 척추동물은

흉선 위축을 경험하며,

이는 흉선 상피 세포(TECs)의 감소와 흉선 내 지방 조직의 축적으로 나타납니다.

인간 티무스 퇴화는

1세부터 시작된다고 추정됩니다(Hale, 2004; Murray et al., 2003; Palmer et al., 2018).

쥐의 티무스는

4주에 최대 질량을 달성한 후 점차 감소합니다(Sutherland et al., 2005).

최근 Wu 등(Wu et al., 2018)은 2, 6, 10주령 마우스의 골수 티모스 상피 세포(mTEC) 전사체를 분석하여 출생 후 6주부터 연령 관련 TEC 퇴화가 발생함을 보여주었습니다. 이는 세포 주기 관련 유전자 발현 감소와 염증 반응 관련 유전자 발현 증가로 입증되었습니다. 티무스는 쥐에서 4–6주, 인간에서 1년 만에 퇴화하며, 이 시기는 다른 대부분의 장기가 노화의 징후를 보이지 않는 시기입니다.

따라서

티무스 퇴화는

진화적으로 보존된 과정일 수 있습니다(Chaudhry et al., 2016).

일부 과학자들은

흉선 퇴축이 아직 알려지지 않은 생물학적 목적을 가질 수 있다고 추측하며,

흉선 퇴축을 초기 단계(성장 의존적 흉선 퇴축)와 후기 단계(연령 의존적 흉선 퇴축)로 나누었습니다(Aw & Palmer, 2012).

초기 단계(성장 의존성 흉선 위축)는

생체 에너지 균형의 결과일 수 있습니다(Boehm & Swann, 2013).

생애 초기에는

감염으로부터 보호하기 위해 높은 흉선 활동과

광범위한 T세포 수용체(TCR) 다양성을 유지하는 것이 필수적입니다(Aw & Palmer, 2012).

T 세포 레퍼토리가 확립되면,

유기체는 티모스 활동을 감소시키고

에너지를 다른 기관으로 재배분하는 것이 유리할 수 있습니다.

반면,

후기 단계(연령 의존적 티모스 퇴화)는

다른 기관의 노화 과정과 유사할 수 있습니다(Aw & Palmer, 2012).

이 리뷰에서는

주로 연령 의존적 티모스 퇴화에 대해 논의합니다.

흉선 퇴화는

새로운 미성숙 T 세포의 생산 감소와 말초 TCR 레퍼토리의 붕괴를 초래하여

면역 기능 저하를 유발합니다(그림 1).

흉선 퇴화는

암, 감염, 자가면역 질환 등 다양한 질환의 발병 위험 증가와 연관되어 있습니다(Fahy et al., 2019; Goronzy & Weyand, 2003).

노화는

감염성 질환과 종양성 질환의 발생률 증가와 연관되어 있으며,

이는 일반적으로 체계적인 면역 노화와 유전적 변이의 점진적 축적에 기인한다고 알려져 있습니다

(Gavazzi & Krause, 2002; Pawelec, 2017).

최근 Palmer 등(Palmer et al., 2018)은

정교한 수학적 모델을 통해

흉선 위축에 의한 T세포 생산의 연령 관련 감소가

인간에서 많은 암과 감염성 질환의 주요 위험 요인임을 보여주었습니다.

연령 관련 흉선 소실과

자가면역 질환 간의 연관성은

많은 연구에서 입증되었습니다.

예를 들어, Hosaka 등(Hosaka et al., 1996)은

흉선 위축이 극심한 노화 MRL/+ 마우스에서

흉선 이식으로 자가면역 질환을 교정할 수 있음을 보여주었습니다.

또한 일부 연구는

인간에서 류마티스 관절염 진행에

흉선 노화가 관여할 수 있음을 제안했습니다(Goronzy & Weyand, 2003).

따라서,

연령 관련 흉선 위축의 메커니즘과 영향을 더 완전히 이해하는 것은

노화 과정에서 면역 노화 관련 질환을 이해하고 예방하는 데 도움이 될 것입니다.

이 리뷰에서는

연령 관련 흉선 위축의 메커니즘과 영향에 대한 현재의 지식을 요약합니다.

FIGURE 1.

Open in a new tab

Effects of age on thymic development and function. Age‐related thymic involution leads to a gradual reduction in thymic cellularity and thymic stromal microenvironment disruption, including the loss of definite cortical‐medullary junctions, a reduction in cTECs and mTECs, fibroblast expansion, an increase in perivascular space (PVS), and more. The disrupted thymic stromal microenvironment disturbs thymocyte development causing decreased ETP and DP frequency, increased DN frequency, and abnormal CD3+ DN cell accumulation. The young thymus is able to produce functionally competent T cells expressing a broad TCR repertoire, whereas the aged thymus produces fewer naïve T cells with a restricted TCR repertoire

2. THE IMPACT OF AGE ON THYMIC DEVELOPMENT AND FUNCTION

2.1. The impact of age on thymic stromal cells

Aged‐related thymic involution reduces thymic cellularity in mice by 50% at 16 weeks in comparison with its adult peak at 4 weeks, eventually leading to less than 5% thymic cellularity (Baran‐Gale et al., 2020; Dooley & Liston, 2012; Gray et al., 2006; Sutherland et al., 2005). In humans, thymus size reduction begins as early as 1 year of age, and it continues to decline at a rate of approximately 3% per year until middle age before slowing down to less than 1% per year (George & Ritter, 1996). Morphological analysis has shown that cortical and medullary thymic epithelial region structure becomes increasingly less reticular and less globular with age in mice, and the definite cortical‐medullary junction is also gradually lost with age (Aw et al., 2008; Baran‐Gale et al., 2020) (Figure 1). Furthermore, aging is concomitant with thymic epithelial space contraction and perivascular space (PVS) augmentation in humans (Steinmann et al., 1985) (Figure 1). The aged thymus displays obvious TEC reduction, fibroblast and adipocyte expansion, and senescent cell accumulation (Aw et al., 2008; Gray et al., 2006; Palmer, 2013). Compared with 4 week old mice, TEC cellularity is reduced by about 50% at 16 weeks and over 80% at 50 weeks (Baran‐Gale et al., 2020). In particular, the mTEC population decreases gradually with age, leading to a decline in mTEC/cortical thymic epithelial cell (cTEC) ratio in aged mice (Gray et al., 2006). Although it is widely accepted that TEC number significantly decreases with thymic aging, a recent publication indicated that thymic aging leads to the contraction of cTEC complex cell projections, but has no effect on TEC cell number in mice (Venables et al., 2019). The authors of this article speculated that the use of mechanical/enzymatic methods to isolate TECs in previous studies may have led to a gross underestimate of total TEC number, and they believed that the genetic labeling approach used in their study could overcome this (Venables et al., 2019).

Proliferation of both CD45− non‐TECs stromal cells and TECs decreases dramatically with thymic aging in mice (Gray et al., 2006). Impaired TEC proliferation in aged mice was recently further demonstrated using transcriptome analysis (Cowan et al., 2019; Ki et al., 2014). Thymic aging is accompanied by a decline in TEC marker expression‎, including EpCAM, keratin, CD205, and Ulex Europaeus Agglutinin 1 (Aw et al., 2008). During thymic aging, the ratio of MHCIIhi TECs to MHCIIlo TECs clearly decreases, reflecting a reduced TEC antigen presentation ability in aged mice (Gray et al., 2006). However, it is notable that some recent studies have shown that the emergence of MHCIIlo TEC subsets during thymic development has other specific roles, such as supporting invariant NKT (iNKT) cell development in the thymus (Kozai et al., 2017; Lucas et al., 2020). Tissue‐restricted antigen expression‎ also diminishes with age, representing a potential mechanism for age‐related increase in autoimmune diseases (Baran‐Gale et al., 2020; A. Griffith et al., 2015; Griffith et al., 2012). Aging also impairs TEC secretion ability, as demonstrated by diminished production of the thymopoietic cytokine IL‐7 in mice (Aspinall & Andrew, 2000; Ortman et al., 2002). IL‐7 administration in older mice and in the rhesus macaque increases thymic output (Aspinall et al., 2007; Pido‐Lopez et al., 2002).

Advances in bulk RNA‐seq and single‐cell RNA‐seq (scRNA‐seq) technology have allowed us to more comprehensively investigate TEC subpopulation changes and transcriptional profile changes during thymic aging. A recent scRNA‐seq study compared TEC subsets in young and old mice. Most mTECs considerably diminished and most cTECs dramatically increased in percentage upon aging, which is consistent with previous reports (Yue et al., 2019). The mTEC progenitor subsets also reduced with age; however, there was a much higher frequency of bipotent TEC progenitors in the aged thymus compared to young mice (Yue et al., 2019). It is worth noting that the mTEC and cTEC subsets in this study were divided roughly based on t‐distributed stochastic neighbor embedding analysis, and the precise nature of these subsets needs to be further elucidated experimentally. A more recent study subdivided TECs from 1, 4, 16, 32, and 52 week old mice into 9 different subtypes using scRNA‐seq analysis. In this study, the authors showed that the proportion of both perinatal cTECs and mature mTECs were significantly reduced with aging, in contrast to the proportion of mature cTECs and intertypical TECs, which increased with aging (Baran‐Gale et al., 2020). By using scRNA‐seq and lineage tracing mouse models, the authors demonstrated that intertypical TECs represent a TEC progenitor state and that aging compromises intertypical TEC differentiation into mature mTECs (Baran‐Gale et al., 2020). They further analyzed the transcriptional signatures of mature cTECs, mature mTECs, and intertypical TECs during aging and found that an inflamm‐aging transcriptional signature was restricted to mature cTECs and mature mTECs, rather than intertypical TECs (Baran‐Gale et al., 2020).

By comparing thymic stromal cell population transcriptomes from 1‐, 3‐, and 6‐month‐old mice, Ki et al. found that the expression‎ of E2F3 transcriptional targets and cell cycle‐associated genes decreased with early thymic aging in cTECs and mTECs (Ki et al., 2014). A similar study showed that the decline in E2F3 transcriptional targets and cell cycle‐associated genes occurs as early as 6 weeks in mice (Wu et al., 2018). E2F3 is a transcription factor that regulates cell proliferation and many cell cycle‐associated genes (Humbert et al., 2000); thus, reduced E2F3 activity results in decreased TEC cell‐cycle progression in aged mice. Cell cycle‐related gene downregulation during thymic aging was further confirmed by another study that showed a decline in myc targets and ribosomal genes with thymic aging in mice (Cowan et al., 2019). By using a FoxN1MycTg mouse model, in which myc is overexpressed in TECs, the authors further demonstrated that myc mainly promotes ribosomal gene expression‎ in TECs, which are distinct from cyclin D1 regulated genes (Cowan et al., 2019). These bulk RNA‐seq and scRNA‐seq results provide an overview of TEC transcriptional and cell subset changes during thymic aging. Some representative cTEC and mTEC transcripts that are downregulated during aging are summarized in Table 1.

2. 연령이 흉선 발달 및 기능에 미치는 영향

2.1. 연령이 흉선 간질 세포에 미치는 영향

노화 관련 흉선 퇴화는

16주령 마우스에서 성체 피크(4주령)에 비해 흉선 세포 수가 50% 감소하며,

결국 5% 미만의 흉선 세포 수로 감소합니다

(Baran-Gale 등, 2020; Dooley & Liston, 2012; Gray et al., 2006; Sutherland et al., 2005).

인간에서는 흉선 크기의 감소가 1세부터 시작되어

중년까지 약 3%씩 감소한 후,

연간 1% 미만으로 감소 속도가 느려집니다(George & Ritter, 1996).

형태학적 분석 결과, 쥐에서 티무스 상피 조직의 피질부와 수질부 구조는 연령에 따라 점차 망상 구조가 감소하고 구형 구조가 줄어들며, 명확한 피질-수질 경계도 연령에 따라 점차 사라집니다(Aw 등, 2008; Baran-Gale 등, 2020) (그림 1). 또한 인간에서 노화는 흉선 상피 공간의 수축과 혈관 주위 공간(PVS)의 증가와 동반됩니다(Steinmann et al., 1985) (그림 1). 노화된 흉선은 TEC 감소, 섬유모세포 및 지방세포의 확장, 노화 세포의 축적을 보여줍니다(Aw et al., 2008; Gray et al., 2006; Palmer, 2013). 4주령 마우스와 비교했을 때, TEC 세포 수는 16주령에서 약 50%, 50주령에서 80% 이상 감소합니다(Baran-Gale et al., 2020). 특히 mTEC 인구 수는 연령에 따라 점차 감소하여 노화 마우스에서 mTEC/피질 티무스 상피 세포(cTEC) 비율이 감소합니다(Gray et al., 2006). 티무스 노화에 따라 TEC 수가 유의미하게 감소한다는 것은 널리 인정되고 있지만, 최근 연구에서는 티무스 노화가 cTEC 복합체 세포 투사체의 수축을 유발하지만 쥐의 TEC 세포 수에는 영향을 미치지 않는다는 결과가 보고되었습니다(Venables et al., 2019). 이 논문의 저자들은 이전 연구에서 TEC를 분리하기 위해 사용된 기계적/효소적 방법이 총 TEC 수를 크게 과소평가했을 수 있다고 추측했으며, 그들의 연구에서 사용된 유전적 표지 접근법이 이를 극복할 수 있다고 믿었습니다(Venables et al., 2019).

쥐에서 흉선 노화 시 CD45− 비-TEC 간질 세포와 TEC의 증식이 급격히 감소합니다(Gray et al., 2006). 노화된 쥐에서 TEC 증식의 장애는 최근 전사체 분석을 통해 추가로 입증되었습니다(Cowan et al., 2019; Ki et al., 2014). 흉선 노화는 EpCAM, 케라틴, CD205, Ulex Europaeus Agglutinin 1(Aw et al., 2008)과 같은 TEC 표지자 발현 감소와 동반됩니다. 흉선 노화 과정에서 MHCIIhi TEC와 MHCIIlo TEC의 비율이 명확히 감소하며, 이는 노화 마우스에서 TEC의 항원 제시 능력이 감소함을 반영합니다(Gray et al., 2006). 그러나 최근 일부 연구에서는 티무스 발달 과정에서 MHCIIlo TEC 하위 집단의 출현이 티무스 내 불변 NKT(iNKT) 세포 발달을 지원하는 등 다른 특정 역할을 한다는 점이 주목됩니다(Kozai et al., 2017; Lucas et al., 2020). 조직 특이적 항원 발현도 연령과 함께 감소하며, 이는 연령 관련 자가면역 질환의 증가를 설명하는 잠재적 메커니즘으로 제시되었습니다(Baran-Gale et al., 2020; A. Griffith et al., 2015; Griffith et al., 2012). 노화는 TEC의 분비 능력을 손상시키며, 이는 쥐에서 흉선 생성 사이토킨 IL-7의 생산 감소로 입증되었습니다(Aspinall & Andrew, 2000; Ortman et al., 2002). 노화된 쥐와 레서스 원숭이에 IL-7을 투여하면 흉선 출력이 증가합니다(Aspinall et al., 2007; Pido-Lopez et al., 2002).

대량 RNA-seq 및 단일 세포 RNA-seq(scRNA-seq) 기술의 발전은 흉선 노화 과정에서 TEC 하위 집단 변화와 전사 프로파일 변화를 더 포괄적으로 조사할 수 있게 했습니다. 최근 scRNA-seq 연구는 젊은 쥐와 노화된 쥐의 TEC 하위 집단을 비교했습니다. 대부분의 mTEC은 노화에 따라 비율이 크게 감소했으며, 대부분의 cTEC은 급격히 증가했습니다. 이는 이전 보고와 일치합니다(Yue et al., 2019). mTEC 전구체 하위집합도 연령에 따라 감소했지만, 노화 흉선에서는 젊은 쥐에 비해 양분화 가능 TEC 전구체의 빈도가 훨씬 높았습니다(Yue et al., 2019). 이 연구에서 mTEC와 cTEC 하위집합은 t-분포 확률적 이웃 임베딩 분석을 기반으로 대략적으로 구분되었으며, 이러한 하위집합의 정확한 특성은 실험적으로 추가로 규명되어야 합니다. 최근 연구에서는 1, 4, 16, 32, 52주령 마우스의 TECs를 scRNA-seq 분석을 통해 9개의 다른 하위 유형으로 세분화했습니다. 이 연구에서 저자들은 출생 후 초기 cTEC와 성숙한 mTEC의 비율이 노화에 따라 유의미하게 감소한 반면, 성숙한 cTEC와 인터타입 TECs의 비율은 노화에 따라 증가했음을 보여주었습니다(Baran-Gale 등, 2020). scRNA-seq 및 계통 추적 마우스 모델을 사용해 저자들은 인터타입 TECs가 TEC 전구체 상태를 나타내며, 노화가 인터타입 TEC의 성숙한 mTEC로의 분화를 방해한다는 것을 입증했습니다(Baran-Gale et al., 2020). 연구진은 노화 과정에서 성숙한 cTECs, 성숙한 mTECs, 및 인터타입 TECs의 전사체 서명을 분석했으며, 염증-노화 전사체 서명이 인터타입 TECs가 아닌 성숙한 cTECs와 성숙한 mTECs에 제한적으로 나타났음을 발견했습니다(Baran-Gale et al., 2020).

Ki 등(Ki et al., 2014)은 1개월, 3개월, 6개월령 마우스의 흉선 간질 세포 군집 전사체를 비교하여 cTECs와 mTECs에서 E2F3 전사체 표적 유전자 및 세포 주기 관련 유전자의 발현이 초기 흉선 노화 과정에서 감소함을 발견했습니다. 유사한 연구에서 E2F3 전사 표적 유전자와 세포 주기 관련 유전자의 감소는 쥐에서 6주 만에 발생한다는 것이 밝혀졌습니다(Wu et al., 2018). E2F3는 세포 증식을 조절하는 전사 인자로, 많은 세포 주기 관련 유전자를 조절합니다(Humbert et al., 2000); 따라서 E2F3 활성의 감소는 노화된 쥐에서 TEC 세포 주기 진행이 감소합니다. 흉선 노화 과정에서 세포 주기 관련 유전자 발현 감소는 다른 연구에서도 확인되었으며, 쥐에서 흉선 노화에 따라 myc 표적 유전자와 리보솜 유전자의 감소가 관찰되었습니다(Cowan et al., 2019). FoxN1MycTg 마우스 모델을 사용한 연구에서, TEC에서 myc가 과발현되는 이 모델에서 저자들은 myc가 TEC에서 리보솜 유전자 발현을 주로 촉진하며, 이는 cyclin D1에 의해 조절되는 유전자와 구분된다는 것을 추가로 입증했습니다(Cowan et al., 2019). 이 대량 RNA-seq 및 scRNA-seq 결과는 흉선 노화 과정에서 TEC의 전사체 및 세포 하위 집합 변화를 개괄적으로 보여줍니다. 노화 과정에서 발현이 감소하는 대표적 cTEC 및 mTEC 전사체는 표 1에 요약되어 있습니다.

TABLE 1.

The expression‎ levels of the representative downregulated genes in mTECs and cTECs during aging

Cell typesGenesAverage expression‎ level (TPM)NewbornAdultAged

mTECs	Cell cycle‐related genes or E2F3 targets	Ccna2	75.87	63.14	41.00
		Ccnb1	93.8	86.26	60.38
		Ccnb2	73.15	56.97	41.63
		Cdk1	96.96	78.58	57.27
		Cdkn2d	38.29	34.65	19.96
	Ribosomal genes	Rpl23a	1224.37	1002.92	760.76
		Rpl10a	361.87	330.44	242.68
		Rps24	601.92	545.33	344.34
		Rps29	1714.98	1339.24	919.09
		Rpl9	1313.21	1105.58	726.28
cTECs	Cell cycle‐related genes or E2F3 targets	Ccna2	53.09	41.22	5.92
		Ccnb1	52.71	53.66	2.13
		Ccnb2	38.14	80.92	8.57
		Cdk1	64.35	21.59	2.75
		Cdkn2d	39.73	45.84	33.06
	Ribosomal genes	Rpl23a	1267.49	827.83	626.10
		Rpl10a	531.99	326.59	210.05
		Rps24	728.18	398.52	180.39
		Rps29	1696.40	881.87	426.52
		Rpl9	2077.67	986.22	662.24

Open in a new tab

Note: According to the RNA‐seq data of Cowan et al. (2019).

In addition to TECs, aging also affects other stromal cells in the thymus. Thymic aging coincides with adipocyte accumulation around the thymus, and the increase in adipose tissue may inhibit thymic function through adipocytokine secretion (Dixit, 2010). Fibroblast percentage also increases in the aging thymus (Figure 1) in species including mice, humans, and fish, suggesting that this may be a conserved feature (Bertho et al., 1997; Gray et al., 2006; Torroba & Zapata, 2003). Recently, a thymic stromal cell transcriptome analysis revealed that proinflammatory gene expression‎ increased with aging in mouse thymic dendritic cells, which in turn may accelerate thymic aging (Ki et al., 2014). Another interesting study demonstrated that thymic B cell function is also impaired with aging in mice; the authors showed that Aire and Aire‐dependent tissue‐restricted antigen expression‎ decline in aging thymic B cells (Cepeda et al., 2018). Thus, aging impairs many cell subsets in the thymic microenvironment.

TECs 외에도 노화는 흉선 내 다른 간질 세포에도 영향을 미칩니다.

흉선 노화는

흉선 주변의 지방세포 축적과 동시에 발생하며,

지방 조직의 증가는 지방세포 사이토킨 분비를 통해 흉선 기능을 억제할 수 있습니다(Dixit, 2010).

노화된 흉선에서 섬유모세포의 비율도 증가합니다(그림 1).

이는 쥐, 인간, 물고기 등 다양한 종에서 관찰되며, 이 현상이 보존된 특징일 수 있음을 시사합니다

(Bertho et al., 1997; Gray et al., 2006; Torroba & Zapata, 2003).

최근 흉선 간질 세포 전사체 분석 결과, 쥐 흉선 ден드리틱 세포에서 염증성 유전자 발현이 노화에 따라 증가했으며, 이는 흉선 노화를 가속화할 수 있음을 보여주었습니다(Ki et al., 2014). 또 다른 흥미로운 연구에서는 쥐의 흉선 B 세포 기능이 노화에 따라 손상된다는 것이 입증되었습니다. 연구자들은 노화된 흉선 B 세포에서 Aire 및 Aire에 의존적인 조직 특이적 항원 발현이 감소함을 보여주었습니다(Cepeda et al., 2018). 따라서 노화는 흉선 미세환경 내 많은 세포 하위 집합의 기능을 손상시킵니다.

2.2. Thymocyte development in the aged thymus

In addition to thymic stromal cells, thymocyte development is also drastically disturbed during thymic aging. Some studies have shown that hematopoietic stem cells (HSCs) of aged mice display an increased bias toward myeloid differentiation concomitant with a diminished lymphoid lineage differentiation ability (Beerman et al., 2010). HSC abnormalities in aged mice may affect the seeding of early T‐lineage progenitors (ETPs) within the thymus. Indeed, ETP frequency declines with aging, and their potential ability to reconstitute the thymus is also reduced (Min et al., 2004, 2005). ETPs from young mice are able to differentiate into all stages of thymocytes when seeded into thymic lobes; in contrast, this differentiation ability is impaired in ETPs from aged mice (Min et al., 2004). However, the effect of aging on HSCs and ETPs is controversial. Zhu et al. established an elegant mouse model in which they transplanted a fetal thymus into the kidney capsule of aged mice, thus providing a young thymic microenvironment for aged lymphohematopoietic progenitor cells (LPCs) (Zhu et al., 2007). Using this model, they demonstrated that the LPCs derived from aged mice and young mice have similar abilities to differentiate into ETPs and subsequent thymocyte subpopulations when transplanted into the young thymic microenvironment, indicating that LPCs do not have a defect synchronized with age‐related thymic involution (Zhu et al., 2007). Another study showed that the ETP defects in aged thymi are mainly due to changes in thymic epithelial architecture, including the poorly defined cortico‐medullary junction and reduced medulla cellularity, rather than ETP‐intrinsic defects (Gui et al., 2007). Thus, although aging may have some effects on HSCs and ETPs, the impairment of ETPs and subsequent thymocyte subpopulations in aged mice can be mainly attributed to thymic microenvironment disruption.

ETPs subsequently differentiate into double negative (DN) (CD4−CD8−) subpopulations that include DN1 (CD44+CD25−), DN2 (CD44+CD25+), DN3 (CD44−CD25+), and DN4 (CD44−CD25−) (Liang, Zhang, Dong, et al., 2021; Luan et al., 2019). The DN subsets subsequently become double positive (DP) (CD4+CD8+) cells that further differentiate into CD4 or CD8 single positive (SP) T cells through the process of positive and negative selection (Germain, 2002). Although both DN and DP population cell numbers are significantly reduced with aging, DN subset frequency increases 2–3 times in aged (24–27 months old) mice compared with young (2–3 months old) mice, whereas the percentage of DP (CD4+CD8+) subpopulations significantly diminishes with age (Thoman, 1995). Among DN subsets, there is a considerable reduction in DN2 and DN3 subset cell numbers with thymic aging in mice (Aspinall, 1997). Additionally, thymic aging is concomitant with the abnormal accumulation of CD3+ DN cells within the thymus (Aw et al., 2009, 2010). Aging also interferes with later stages of thymocyte development. DP and SP thymocytes in aged mice display deregulated CD3 expression‎, which may lead to attenuated TCR‐dependent stimulation (Aw et al., 2009). Indeed, thymocytes from older mice exhibit an impaired mitogen response ability, which is manifested by a failure to upregulate the activation marker CD69 and proliferate (Aw et al., 2010; Djikic et al., 2014). Consistent with impaired thymocyte differentiation in the aged thymus, T‐cell receptor excision circles (TRECs) within the thymus also significantly decline with aging in mice and humans (Ortman et al., 2002; Palmer et al., 2018). Thus, aging impairs multiple thymocyte developmental stages.

2.2. 노화 흉선에서의 흉선 세포 발달

티모스 간질 세포 외에도 티모스 노화 과정에서 티모사이트 발달이 극적으로 방해받습니다. 일부 연구에서는 노화된 쥐의 혈액 줄기 세포(HSCs)가 골수 분화 방향으로의 편향이 증가함과 동시에 림프계 분화 능력이 감소한다는 것이 밝혀졌습니다(Beerman et al., 2010). 노화된 쥐의 HSC 이상은 티모스 내 초기 T계열 전구세포(ETPs)의 정착에 영향을 미칠 수 있습니다. 실제로 ETP의 빈도는 노화에 따라 감소하며, 티무스를 재구성하는 잠재적 능력도 감소합니다(Min et al., 2004, 2005). 젊은 쥐의 ETP는 티무스 소엽에 이식될 때 모든 단계의 티모사이트로 분화할 수 있지만, 노화된 쥐의 ETP는 이 분화 능력이 손상됩니다(Min et al., 2004). 그러나 노화가 HSCs와 ETP에 미치는 영향은 논란의 여지가 있습니다. Zhu et al.은 노화된 쥐의 신장 캡슐에 태아 흉선을 이식하여 노화된 림프혈액 생성 전구세포(LPCs)에 젊은 흉선 미세환경을 제공하는 우아한 쥐 모델을 확립했습니다(Zhu et al., 2007). 이 모델을 통해 그들은 노화된 쥐와 젊은 쥐에서 유래한 LPC가 젊은 흉선 미세환경으로 이식되었을 때 ETP 및 후속 흉선 세포 하위 집단으로 분화하는 능력이 유사함을 보여주었습니다. 이는 LPC가 연령 관련 흉선 퇴화와의 동기화된 결함을 갖지 않음을 시사합니다(Zhu et al., 2007). 또 다른 연구에서는 노화 흉선의 ETP 결함이 주로 흉선 상피 구조의 변화, 즉 명확하지 않은 피질-수질 경계와 수질 세포 밀도의 감소와 관련이 있으며, ETP 내재적 결함 때문은 아니라는 점을 보여주었습니다(Gui et al., 2007). 따라서 노화는 HSCs와 ETP에 일부 영향을 미칠 수 있지만, 노화 마우스에서 ETP의 기능 장애와 후속 흉선 세포 하위 집단의 손상은 주로 흉선 미세환경의 파괴에 기인한다고 볼 수 있습니다.

ETPs는 이후 이중 음성(DN) (CD4−CD8−) 하위 집단으로 분화되며, 이는 DN1 (CD44+CD25−), DN2 (CD44+CD25+), DN3 (CD44−CD25+), 및 DN4 (CD44−CD25−)로 구성됩니다(Liang, Zhang, Dong, et al., 2021; Luan et al., 2019). DN 하위집단은 이후 이중 양성(DP) (CD4+CD8+) 세포로 분화되며, 양성 및 음성 선택 과정을 통해 CD4 또는 CD8 단일 양성(SP) T 세포로 추가 분화됩니다(Germain, 2002). 노화에 따라 DN 및 DP 세포 수는 모두 유의미하게 감소하지만, 노화된 쥐(24–27개월령)에서 DN 하위 집합의 빈도는 젊은 쥐(2–3개월령)에 비해 2–3배 증가하는 반면, DP(CD4+CD8+) 하위 집합의 비율은 연령에 따라 유의미하게 감소합니다(Thoman, 1995). DN 하위집단 중에서는 쥐의 흉선 노화에 따라 DN2 및 DN3 하위집단 세포 수가 크게 감소합니다(Aspinall, 1997). 또한 흉선 노화는 흉선 내 CD3+ DN 세포의 비정상적 축적과 동반됩니다(Aw et al., 2009, 2010). 노화는 흉선 세포 발달의 후반 단계에도 영향을 미칩니다. 노화 마우스의 DP 및 SP 티모사이트는 CD3 발현이 조절되지 않아 TCR 의존적 자극이 약화될 수 있습니다(Aw et al., 2009). 실제로 노화 마우스의 티모사이트는 활성화 표지자 CD69의 발현 증가와 증식 능력이 저하되어 미토겐 반응 능력이 손상됩니다(Aw et al., 2010; Djikic et al., 2014). 노화 흉선에서의 티모사이트 분화 장애와 일치하게, 흉선 내 T세포 수용체 절제 고리(TRECs)도 쥐와 인간에서 노화에 따라 유의미하게 감소합니다(Ortman et al., 2002; Palmer et al., 2018). 따라서 노화는 티모사이트 발달의 여러 단계를 손상시킵니다.

2.3. Thymic involution effects on thymic output

Mature CD4 SP and CD8 SP thymocytes are exported to the periphery where they play a role in immunological surveillance (Liang, Zhang, Zhang, et al., 2021; Zhang et al., 2021). Age‐related thymic involution causes an obvious reduction in the thymic output of naïve T cells and subsequently decreases peripheral T‐cell diversity (Chaudhry et al., 2016; Cowan et al., 2020). Diminished thymic production of naïve T cells leads to homeostatic expansion of existing T cells, resulting in memory T‐cell augmentation (Surh & Sprent, 2000). Although it is well accepted that the thymic output of peripheral naïve T cells progressively declines with aging in mice (den Braber et al., 2012), in humans, the relationship of thymic involution to peripheral naïve T‐cell maintenance is a matter of debate. Many studies using TRECs as a measurement of thymic output demonstrate that peripheral naïve T‐cell thymic output declines with aging in humans (Fagnoni et al., 2000; Ferrando‐Martinez et al., 2011; Mitchell et al., 2010; Naylor et al., 2005). However, Braber et al. showed that adult human peripheral naïve T‐cell pool maintenance occurs almost exclusively through cell proliferation, rather than thymic output (den Braber et al., 2012). Thus, the contribution of thymic output to naïve T‐cell pool maintenance in adults needs further investigation.

Aging also interferes with naïve T‐cell properties and functions (Srinivasan et al., 2021). Naïve T cells from aged mice express elevated levels of senescence markers and display reduced proliferation ability upon antigen stimulation (Akbar & Henson, 2011; Chaudhry et al., 2016). Chemokine receptor expression‎ is also altered in CD4+ T cells of aged mice, exhibiting a deregulation of CCR1, 7, and 8 and CXCR2, 4, and 5, which may impair their migration ability (Mo et al., 2003). The reduced number of naïve T cells together with the disrupted function of naïve T cells during aging leads to impaired immunological surveillance ability in aged organisms.

3. AGE‐RELATED THYMIC INVOLUTION MECHANISMS

3.1. Thymic stromal cell alterations lead to thymic involution

Although the T‐lineage differentiation potential of HSCs and ETPs is partially compromised in aged mice compared with young mice (Min et al., 2004; Zediak et al., 2007), increasing evidence suggests that thymic involution is mainly caused by age‐related thymic stromal cell degeneration, particularly TEC degeneration (Chen et al., 2009; A. V. Griffith et al., 2015; Gui et al., 2012; Zhu et al., 2007). For example, a global transcriptome analysis of thymic stromal cells and lymphocytes revealed that mouse thymic stromal cells, in contrast to lymphocytes, are deficient in catalase (A. V. Griffith et al., 2015). This results in elevated H2O2 levels and stromal cell oxidative damage, which subsequently leads to thymic atrophy. The authors further showed that thymic atrophy could be ameliorated by genetic and biochemical restoration of antioxidant activity (A. V. Griffith et al., 2015). Similar to this study, another publication revealed that the thymi of human Down syndrome patients exhibited premature senescence, and TECs from Down syndrome patients showed increased oxidative stress (Marcovecchio et al., 2021). Using a genome‐wide computational approach, another group showed that age‐associated thymic degeneration is primarily a stromal cell function change (Griffith et al., 2012). Many studies support the pivotal role of thymic stroma in thymic aging. Mackall et al. showed that lethally irradiated older mice exhibit impaired thymopoiesis compared with lethally irradiated young mice after both were injected with young bone marrow (Mackall et al., 1998). A similar experiment showed that intrathymic injection of young ETPs failed to restore normal thymopoiesis in older mice but did so in young mice (Zhu et al., 2007). In contrast, the same study showed that fetal thymi transplants into the kidney capsules of young or old mice had similar thymopoiesis (Zhu et al., 2007). Overall, these findings suggest that the thymic stroma is a key factor in regulating age‐related thymic involution. Likely, the durable identity of the thymus is established by its stromal components because developing thymocytes are only transiently present in the thymus (Petrie & Zuniga‐Pflucker, 2007).

3.2. Molecular regulation of thymic involution

Foxn1 is essential for embryonic thymic organogenesis and TEC maintenance in adults (Zuklys et al., 2016), and emerging evidence suggests that Foxn1 also plays a critical role in preventing age‐related thymic involution (Abramson & Anderson, 2017). Foxn1 expression‎ progressively declines with aging (Figure 2), and Foxn1 overexpression‎ ameliorates age‐related thymic deterioration, indicating that Foxn1 is a pivotal regulator of thymic aging (Bredenkamp et al., 2014; Chen et al., 2009; O'Neill et al., 2016; Rode et al., 2015; Zook et al., 2011). Sun et al. generated a loxP‐floxed‐Foxn1 mouse model carrying a ubiquitous CreERT transgene with a low level of spontaneous activation leading to a gradual loss of Foxn1 expression‎ with age (Sun et al., 2010). By examining this mouse model's phenotype at different ages, Sun et al. demonstrated that gradual Foxn1 loss with age substantially accelerates age‐related thymic involution (Sun et al., 2010). In contrast, Foxn1 overexpression‎ restores most of the changes caused by thymic involution in old mice, including thymic mass enlargement, increased ETP frequency, elevated EpCAM+MHCII+ TEC cell number, and CD4+ and CD8+ naïve compartment expansion in the spleen (Bredenkamp et al., 2014; Zook et al., 2011). Moreover, a recent study showed that engrafting Foxn1‐reprogrammed embryonic fibroblasts could rejuvenate aged thymic architecture and function in both male and female mice (Oh et al., 2020). Collectively, these studies demonstrate a crucial role for Foxn1 in regulating age‐associated thymic degeneration.

FIGURE 2.

Open in a new tab

Age‐related thymic involution mechanisms.

Both positive and negative regulators of thymic involution have been identified. Positive regulators include Foxn1, E2F3, myc, Wnt4, FGF21, KGF, IL‐7, IL‐22, miR‐181a‐5p, Lamin‐B1, Leptin, GH, IGF‐1, Ghrelin, and GHSR, which exhibit reduced activity with age. Negative regulators include Axin, LIF, OSM, IL‐6, SCF, IL‐1β, miR‐125a‐5p, miR‐205‐5p, and follistatin, which exhibit increased activity with age. In addition, CR can attenuate age‐related thymic involution, while obesity and sex hormones exacerbate age‐related thymic involution

The Wnt signaling pathway attenuates with aging in the thymus (Ferrando‐Martinez et al., 2015; Kvell et al., 2010; Yang, Youm, Sun, et al., 2009) (Figure 2). Ferrando‐Martínez et al. found that the nonadipocytic component of the human thymus expresses higher levels of Wnt pathway inhibitors in the elderly than in the young, thus attenuating the Wnt pathway (Ferrando‐Martinez et al., 2015). Using thymic stromal cell transcriptome analysis, Griffith et al. revealed that Wnt signaling deregulation is the most significant hallmark of thymic degeneration (Griffith et al., 2012). A previous study illustrated that the expression‎ of Axin, a Wnt inhibitor, on mTECs and fibroblasts increases with aging in humans and mice, and Axin knockdown by RNA interference ameliorates age‐related thymic degeneration (Yang, Youm, Sun, et al., 2009). Another study showed that Wnt pathway reduction during aging may involve the epithelial–mesenchymal transition (EMT) process in mice, which we will discuss further below (Kvell et al., 2010).

Growth factors and cytokines also play critical roles in age‐related thymic involution. Prolongevity ketogenic hormone fibroblast growth factor 21 (FGF21) expression‎ gradually declines in the thymus with age, and loss of FGF21 function in middle‐aged mice accelerates age‐dependent thymic deterioration (Figure 2), suggesting that FGF21 expression‎ could protect against age‐related thymic involution (Youm et al., 2016). Leukemia inhibitory factor (LIF), oncostatin M (OSM), IL‐6, and stem cell factor (SCF) expression‎ levels all increase with age in mice (Figure 2), and this elevated expression‎ is associated with thymic involution (Sempowski et al., 2000). Studies have shown that these cytokines may originate from adipocytes or TECs (Dooley & Liston, 2012; Ventevogel & Sempowski, 2013). Another proinflammatory cytokine, IL‐1β, is mainly expressed by macrophages in the thymus (Figure 2), and its increased expression‎ levels also lead to thymic involution (Dixit, 2012; Finn et al., 2018; Guarda et al., 2011). Consistent with this, IL‐1β receptor is primarily expressed in TECs, and ablation of Nlrap3 and Asc, which are required for IL‐1β activation, protect against age‐related thymic demise and immunosenescence in mice (Youm et al., 2012). In contrast, some cytokines and growth factors play positive roles in preventing age‐associated thymic degeneration. Studies have shown that IL‐7, IL‐22, and keratinocyte growth factor administration reverse age‐induced thymic involution in humans and mice (Ventevogel & Sempowski, 2013).

MicroRNAs (miRNAs) have been implicated in the aging process in many organisms, and the role of miRNAs in age‐associated thymic deterioration was recently investigated. Guo et al. compared various miRNA expression‎ levels in TECs from 2‐month‐old and 20‐month‐old mice and identified many differentially expressed miRNAs (Guo et al., 2013). Whole thymus miRNA expression‎ has also been examined during aging. Compared with thymi from 1‐month‐old mice, 50 and 81 miRNAs were differentially expressed in thymi from 10‐month‐old and 19‐month‐old mice, respectively (Ye et al., 2014). Among these differentially expressed miRNAs, miR‐181a‐5p and miR‐125a‐5p, which were downregulated and upregulated during aging, respectively, have been studied further. Guo et al. revealed that miR‐181a‐5p expression‎ decreased in TECs of 10‐ to 19‐month‐old mice compared to 1‐month‐old mice, and miR‐181a‐5p promoted mTEC proliferation by targeting transforming growth factor beta receptor I (Tgfbr1), which exhibits increased expression‎ with aging. This suggests that miR‐181a‐5p could prevent age‐related thymic demise by interfering with TGFβ signaling that could negatively regulate the development of mTECs and promote thymic involution (D. G. Guo et al., 2016; Hauri‐Hohl et al., 2008, 2014; Xu et al., 2018). In contrast to miR‐181a‐5p, Xu et al. revealed that miR‐125a‐5p expression‎ increased in TECs of aged mice compared with TECs of young mice. They found that miR‐125a‐5p suppressed Foxn1 expression‎, which may underlie its role in promoting age‐related thymic involution (Xu et al., 2018; Xu, Sizova, et al., 2017). A more recent study showed that miR‐205‐5p expression‎ in TECs markedly increased with aging in mice, and miR‐205‐5p promoted age‐associated thymic involution by inhibiting TEC proliferation (Gong et al., 2020). Another study compared differentially expressed miRNAs in the thymi of 10‐month‐old newborn babies and 70‐year‐old humans and showed that 106 miRNAs were significantly changed in elderly thymi (Ferrando‐Martinez et al., 2015). Furthermore, some of the altered miRNAs in this study, such as miR‐25, miR‐134, and miR‐7f, could modulate the Wnt pathway (Ferrando‐Martinez et al., 2015). In addition to age‐related thymic involution, microRNAs also regulate stress‐induced thymic involution. Papadopoulou et al. showed that miR‐29a could prevent pathogen‐associated thymic involution via targeting the IFN‐α receptor in TECs, and Hoover et al. revealed that miR‐205 expression‎ in TECs could maintain thymopoiesis following inflammatory perturbations in mice (Hoover et al., 2016; Papadopoulou et al., 2012). Thus targeting miRNAs may be a potential strategy to rejuvenate age‐induced diminished thymic function (Xu, Zhang, et al., 2017).

Recently, some new thymic involution regulators have been identified. Lamin‐B1 is a cellular architectural protein that has recently been shown to play a critical role in preventing thymic aging in mice (Yue et al., 2019). Yue et al. demonstrated that the increased proinflammatory cytokines produced by thymic myeloid immune cells during aging diminishes Lamin‐B1 expression‎ in TECs and promotes cell senescence, which subsequently induces age‐related thymic involution (Yue et al., 2019) (Figure 2). Other recent studies have shown that imbalances in follistatin, activin A, and BMP4 signaling drive thymic involution in mice (Lepletier et al., 2019), while liver X receptors, a class of nuclear receptors that sense intracellular oxysterols and cholesterol biosynthetic pathway intermediates, may protect against premature thymic involution in mice (Chan et al., 2020). Additionally, results from our lab showed that TEC‐specific deletion of tuberous sclerosis complex 1 (Tsc1), a negative regulator of mTOR activity (Liang, Zhang, Zhang, et al., 2021), also accelerates thymic involution in mice (unpublished data). Interestingly, sirtuin 6 (Sirt6) is a chromatin deacylase that has been implicated as a key factor in aging (Chang et al., 2020); however, our recent publication showed that Sirt6 deficiency in TECs has no obvious effects on thymic aging in mice (Zhang et al., 2021).

3.3. Sex hormones in thymic involution

Steroid hormone levels change dramatically with aging, and steroid hormones play a critical role in promoting age‐related thymic involution (Gui et al., 2012). The role of sex hormones in thymic involution was first reported in 1904 in a study that found that castrated cattle had enlarged thymi (Henderson, 1904). Additionally, the fact that the thymus degenerates most rapidly after puberty, when steroid hormone production reaches its peak, further supports the role of steroid hormones in thymic involution (Abramson & Anderson, 2017). Thymic involution is also more rapid in males than in females (Gui et al., 2012; Hun et al., 2020), implying that androgens may have a more dramatic impact on thymic involution. Although both TECs and thymocytes express androgen receptors (Olsen et al., 2001), androgen‐mediated thymic involution is caused by direct impact on TECs rather than thymocytes because TEC‐specific (but not thymocyte‐specific) androgen receptor deletion leads to androgen‐mediated thymic involution resistant in mice (Olsen et al., 2001). More recently, a comprehensive transcriptome analysis showed that sexual dimorphism significantly affects cTECs (Dumont‐Lagace et al., 2015). cTECs from male mice display low proliferation rates, and androgen‐dependent signaling represses the expression‎ of genes involved in cTEC development and function, such as Foxn1, Dll4, Psmb11, and Ctsl (Dumont‐Lagace et al., 2015). Consistently, another study demonstrated that sex steroid blockade could increase Dll4 expression‎ and its downstream targets on cTECs in mice, which further promotes thymopoiesis by modulating Notch signaling (Velardi et al., 2014). Notably, although castration is an effective way to regenerate the aged thymus, the thymic regrowth induced by castration is transient (Griffith et al., 2012).

Pregnancy also causes thymic involution, mainly mediated by progesterone (Clarke & Kendall, 1989). Studies have shown that progesterone receptor expression‎ in thymic stromal cells is required for thymic involution during pregnancy in mice (Tibbetts et al., 1999). Interestingly, thymic involution during pregnancy may be essential for normal fertility (Tibbetts et al., 1999). A recent study uncovered that RANK expression‎ in TECs promoted sex hormone‐mediated thymic involution and natural regulatory T‐cell development during pregnancy, which is critical for successful pregnancy and prevention of gestational diabetes (Paolino et al., 2021).

3.4. Metabolic regulation of thymic involution

Caloric restriction (CR) has long been known to play a critical role in increasing life span. Recently, CR was also shown to be effective at preventing age‐related thymic involution. Yang et al. showed that CR could inhibit thymic adipogenesis and reduce age‐related thymic involution in mice (Yang, Youm, Vandanmagsar, et al., 2009). Another study conducted on nonhuman primates obtained similar results; the authors showed that long‐term CR effectively improves naïve T‐cell production and preserves T‐cell receptor repertoire diversity (Messaoudi et al., 2006). Thymus transcriptome analysis in short‐term CR mice showed that CR altered catalytic activity and metabolic processes (Omeroglu Ulu et al., 2018). Short‐term CR also altered the expression‎ of leptin, ghrelin, Igf1, and adiponectin, some of which were reported to be associated with age‐related thymic involution (Omeroglu Ulu et al., 2018).

Obesity increases the risk of infections and cancer, which may be partly ascribed to obesity's negative impact on thymic involution. High‐fat diet fed mice display disrupted thymic structure, including a reduced medullary region and an absence of the cortico‐medullary junction (Gulvady et al., 2013). Diet‐induced obesity also leads to thymocyte apoptosis, reduces thymic output, and compromises TCR repertoire diversity in mice (Yang, Youm, Vandanmagsar, et al., 2009). Progressive adiposity in middle‐aged humans also decreases thymic output (Yang, Youm, Vandanmagsar, et al., 2009). Resveratrol, a phytoalexin produced from plants, has been shown to have the potential to inhibit obesity‐induced thymic involution (Gulvady et al., 2013; Wei et al., 2020). Leptin is a potent adipokine that is responsible for sensing a positive energy balance state and reducing food intake (Friedman & Halaas, 1998). Leptin (ob/ob mouse)‐ and leptin receptor (db/db mouse)‐deficient mice display severe obesity that subsequently causes significant thymic involution (Dixit, 2012; Howard et al., 1999). Leptin administration rescues this accelerated thymic involution in ob/ob mice (Howard et al., 1999). Another study showed that leptin receptor is mainly expressed in the medullary region of the thymus (Gruver et al., 2009). Consistent with results in mice, human patients with loss‐of‐function leptin and leptin receptor mutations also display T‐cell functional defects that could be partially reversed by recombinant leptin administration (Farooqi et al., 2002, 2007). Furthermore, naïve CD8 T‐cell maintenance in nonagenarians has been shown to be associated with high leptin levels (Chen et al., 2010).

Growth hormone (GH) and its proximal mediator, IGF‐1, play critical roles in preventing age‐associated thymic involution. Indeed, GH removal by hypophysectomy leads to thymic atrophy in mice and humans (Napolitano et al., 2008; Savino et al., 2002). Circulating GH levels decline with aging, and GH administration partially ameliorates age‐related thymic involution in mice (Taub et al., 2010). Randomized clinical studies in middle‐aged HIV patients showed that GH treatment increases thymic mass and elevates TRECs in peripheral T cells (Napolitano et al., 2008). Similarly, exogenous administration of IGF‐1 enhances thymopoiesis mainly through TEC expansion in mice (Chu et al., 2008). Furthermore, subcutaneous transplantation of GH3 pituitary adenoma cells, which secrete growth hormone, reverses thymic aging in rats (Kelley et al., 1986). A recent clinical trial showed that recombinant human GH administration combined with dehydroepiandrosterone and metformin could promote thymic regeneration and increase protective immunological changes (Fahy et al., 2019). However, GH application in clinical practice needs to be carefully considered due to its significant side effects (Taub et al., 2010). Ghrelin is a stomach hormone that can induce strong GH‐releasing activity through binding to its receptor‐specific 7‐transmembrane GH secretagogue receptor (GHSR) (Sato et al., 2012). Ghrelin and GHSR expression‎ within the thymus decline with progressive age, and administration of both ghrelin and ghrelin‐receptor agonists alleviate age‐associated thymic deterioration in mice and humans (Dixit et al., 2007, 2009; Smith et al., 2007). Consistent with this, genetic studies revealed that ghrelin and GHSR deficiency accelerates age‐associated thymic demise in mice (Youm et al., 2009). Thus, many hormones contribute positively or negatively to thymic involution.

3.5. Adipocyte origin during thymic involution: The EMT process

Adipogenesis is a notable feature of thymic involution (Dixit, 2010). Recently, some progress has been made in understanding how adipocytes are formed during thymic aging. Using genetically modified reporter mice, Youm et al. first reported that TECs can transition to mesenchymal cells through a mechanism called EMT (Youm et al., 2009). These mesenchymal cells are highly plastic and have the potential to differentiate into adipocytes (Mani et al., 2008). Indeed, these mesenchymal cells express pro‐adipogenic genes, which provide a possible adipocyte origin in the thymus (Youm et al., 2009). Some regulators play critical roles in the EMT process. Peroxisome proliferator‐activated receptor gamma (PPARγ), a member of the nuclear receptor superfamily of ligand‐activated transcription factors, is involved in adipocyte development (Tontonoz & Spiegelman, 2008). Thus, it is reasonable to speculate that PPARγ may play a key role in thymic involution. Indeed, an adipocyte‐lineage‐specific constitutively active PPARγ transgene and administration of rosiglitazone, a PPARγ signaling activator, both promote age‐related thymic involution in mice (Youm et al., 2010). In fact, the Ghrl–GHSR interaction and CR both protect against thymic involution by inhibiting EMT and adipogenesis in mice (Yang, Youm, Vandanmagsar, et al., 2009; Youm et al., 2009), and CR‐mediated thymic involution inhibition also involves PPARγ downregulation in mouse thymic stromal cells (Yang, Youm, & Dixit, 2009). Furthermore, decreased Wnt4 and increased LAP2α during thymic aging may promote direct TEC trans‐differentiation into pre‐adipocytes or cause EMT and subsequent pre‐adipocyte differentiation (Kvell et al., 2010). A recent study showed that CD147 deletion from T cells in mice could prevent thymic involution by inhibiting TEC EMT, implying that the interaction between thymocytes and TECs contributes to age‐related thymic involution (Chen et al., 2020). Although some progress has been made, adipocyte origin during thymic aging needs further investigation.

4. CONCLUSIONS

Age‐related thymic involution contributes significantly to immunosenescence. Although some progress has been made in understanding the molecular regulation of thymic involution, the detailed molecular regulation network is still unclear. Comprehensive information about age‐related thymic involution is needed to promote thymic rejuvenation in the elderly. With advances in transcriptome analysis, significant progress has been made in understanding overall thymic stromal cell changes during aging, and the use of scRNA‐seq has revealed comprehensive TEC subset changes. Thymic aging is associated with the downregulation of cell cycle‐related genes and ribosome biogenesis‐related genes in TECs. Recent genetic studies have also identified some new thymic aging regulators, including FGF21, lamin‐B1, liver X receptors, and some miRNAs. With the current understanding of age‐related thymic involution, we can speculate that thymic stromal cells (especially TECs) offer potential targets for thymic rejuvenation in the elderly.

4. 결론
연령 관련 흉선 위축은 면역 노화에 크게 기여합니다. 흉선 위축의 분자적 조절 메커니즘에 대한 이해는 일부 진전을 보였지만, 상세한 분자적 조절 네트워크는 여전히 명확하지 않습니다. 노인에서의 흉선 재생을 촉진하기 위해 연령 관련 흉선 위축에 대한 포괄적인 정보가 필요합니다. 전사체 분석의 발전으로 노화 과정에서 티모스 간질 세포의 전반적인 변화에 대한 이해가 크게 진전되었으며, scRNA-seq의 활용은 티모스 상피 세포(TEC) 하위 집단의 포괄적인 변화를 밝혀냈습니다. 티모스 노화는 TEC에서 세포 주기 관련 유전자와 리보솜 생합성 관련 유전자의 발현 감소와 연관되어 있습니다. 최근 유전적 연구에서는 FGF21, lamin-B1, 간 X 수용체, 일부 miRNA 등 새로운 흉선 노화 조절 인자들이 식별되었습니다. 현재의 흉선 노화 관련 이해를 바탕으로, 흉선 간질 세포(특히 TEC)는 노인에서의 흉선 재생을 위한 잠재적 표적이 될 수 있다고 추측할 수 있습니다.

AUTHOR CONTRIBUTIONS

Zhanfeng Liang wrote the manuscript; Xue Dong and Qian Zhang designed figures; Zhaoqi Zhang analyzed the RNA‐seq data; Yong Zhao reviewed the manuscript and supervised the work.

CONFLICT OF INTEREST

All authors declare no conflicts of interest.

ACKNOWLEDGMENT

This work was supported by grants from the National Natural Science Foundation for General and Key Programs (31930041, Y.Z., 31800754, Z.L.), the National Key Research and Development Program of China (2017YFA0105002, 2017YFA0104401, and 2017YFA0104402, Y.Z.), and the Knowledge Innovation Program of Chinese Academy of Sciences (XDA16030301, Y.Z.).

Liang, Z. , Dong, X. , Zhang, Z. , Zhang, Q. , & Zhao, Y. (2022). Age‐related thymic involution: Mechanisms and functional impact. Aging Cell, 21, e13671. 10.1111/acel.13671

DATA AVAILABILITY STATEMENT

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

REFERENCES

1. Abramson, J. , & Anderson, G. (2017). Thymic epithelial cells. Annual Review of Immunology, 35, 85–118. 10.1146/annurev-immunol-051116-052320 [DOI] [PubMed] [Google Scholar]
2. Akbar, A. N. , & Henson, S. M. (2011). Are senescence and exhaustion intertwined or unrelated processes that compromise immunity? Nature Reviews Immunology, 11(4), 289–295. 10.1038/nri2959 [DOI] [PubMed] [Google Scholar]
3. Aspinall, R. (1997). Age‐associated thymic atrophy in the mouse due to a deficiency affecting rearrangement of the TCR during intrathymic T cell development. Journal of Immunology, 158(7), 3037–3045. [PubMed] [Google Scholar]
4. Aspinall, R. , & Andrew, D. (2000). Thymic atrophy in the mouse is a soluble problem of the thymic environment. Vaccine, 18(16), 1629–1637. 10.1016/S0264-410x(99)00498-3 [DOI] [PubMed] [Google Scholar]
5. Aspinall, R. , Pido‐Lopez, J. , Imami, N. , Henson, S. M. , Ngom, P. T. , Morre, M. , Niphuis, H. , Remarque, E. , Rosenwirth, B. , & Heeney, J. L. (2007). Old rhesus macaques treated with interleukin‐7 show increased TREC levels and respond well to influenza vaccination. Rejuvenation Research, 10(1), 5–17. 10.1089/rej.2006.9098 [DOI] [PubMed] [Google Scholar]
6. Aw, D. , & Palmer, D. B. (2012). It's not all equal: A multiphasic theory of thymic involution. Biogerontology, 13(1), 77–81. 10.1007/s10522-011-9349-0 [DOI] [PubMed] [Google Scholar]
7. Aw, D. , Silva, A. B. , Maddick, M. , von Zglinicki, T. , & Palmer, D. B. (2008). Architectural changes in the thymus of aging mice. Aging Cell, 7(2), 158–167. 10.1111/j.1474-9726.2007.00365.x [DOI] [PubMed] [Google Scholar]
8. Aw, D. , Silva, A. B. , & Palmer, D. B. (2009). Is thymocyte development functional in the aged? Aging (Albany NY), 1(2), 146–153. 10.18632/aging.100027 [DOI] [PMC free article] [PubMed] [Google Scholar]
9. Aw, D. , Silva, A. B. , & Palmer, D. B. (2010). The effect of age on the phenotype and function of developing thymocytes. Journal of Comparative Pathology, 141, S45–S59. 10.1016/j.jcpa.2009.10.004 [DOI] [PubMed] [Google Scholar]