• Log in with Facebook Log in with Twitter Log In with Google      Sign In    
  • Create Account
  LongeCity
              Advocacy & Research for Unlimited Lifespans

Photo

Cellular senescence and chronological age in various human tissues: A systematic review and meta‐analysis

aging cell cycle proteins cellular senescence human immunosenescence

  • Please log in to reply
No replies to this topic

#1 Engadin

  • Guest
  • 198 posts
  • 580
  • Location:Madrid
  • NO

Posted 08 December 2019 - 03:16 PM


.

 

 

 

 

 

F U L L   T E X T   S O U R C E :   Aging Cell

 

 

 

 

 

Abstract

 
Senescent cells in tissues and organs are considered to be pivotal to not only the aging process but also the onset of chronic disease. Accumulating evidence from animal experiments indicates that the magnitude of senescence can vary within and between aged tissue samples from the same animal. However, whether this variation in senescence translates across to human tissue samples is unknown. To address this fundamental question, we have conducted a systematic review and meta‐analysis of all available literature investigating the magnitude of senescence and its association with chronological age in human tissue samples. While senescence is higher in aged tissue samples, the magnitude of senescence varies considerably depending upon tissue type, tissue section, and marker used to detect senescence. These findings echo animal experiments demonstrating that senescence levels may vary between organs within the same animal.
 
 
1 INTRODUCTION
 
Senescence, the process by which cells stop dividing and enter a state of permanent growth arrest, has been identified as a key hallmark of aging (López‐Otín, Blasco, Partridge, Serrano, & Kroemer, 2013). A higher number of senescent cells have not only been identified in tissue of older individuals (Wang et al., 2009), but also in individuals of higher biological age (Waaijer et al., 2015, 2012), in cancer patients undergoing chemotherapy (Demaria et al., 2017) and at sites of pathology of age‐related diseases (Childs, Durik, Baker, & van Deursen, 2015). Animal experiments have shown that senescent cell numbers can vary across tissue samples within one animal (Baker et al., 2016, 2011; Bussian et al., 2018; Krishnamurthy et al., 2004; Wang et al., 2009). This variation in cells expressing senescence markers may reflect different rates of tissue renewal and/or different triggers of senescence, within tissue types.
 
Seminal experiments that eliminate senescent cells in progeroid and aged mice demonstrate a markedly improved health span after senescence cells are cleared (Baker et al., 2016; Hickson et al., 2019; Jeon et al., 2017; Ogrodnik et al., 2019; Zhu et al., 2015). However, the clearance of senescent cells within and across different tissue samples from the same animal can be variable (Baker et al., 2016; Ogrodnik et al., 2019). Whether this variation in senescent cell clearance is a result of the treatment, age or disease is largely unknown. Currently, whether senescent cells accumulate at the same rate within and across aging tissue samples within humans is unknown. As such, to assess and collate the current evidence and understanding of senescence cells within aged human tissue samples, a systematic review and meta‐analysis of the current literature was conducted.
 
 
2 RESULTS
 
2.1 Selection of included studies
 
Figure 1 illustrates the literature search process. The search retrieved 9,740 articles. After exclusion of duplicates, 5,594 articles were screened for titles and abstracts of which 797 were screened for full text.
 
 
acel13083-fig-0001-m.jpg
 
 
 
Figure 1. Overview of the search strategy
 
 
 
2.2 Characteristics of included studies
 
Overall, 51 articles were included in the systematic review. Table S1 provides a comprehensive overview of included articles. The total number of included participants was 2,072. Overall, most studies reported an age range rather than a mean/median age of the study population. As such, the age of participants included in this review ranged from neonatal to 120 years. Sex distributions of the study population were given in 33 of 51 articles (65%), in which 52% (n = 874/1,694) of the participants are female. Cell cycle regulators (n = 34/51 articles), particularly p16INK4a (n = 27/51 articles), were the most used markers for detection of senescence in human tissue samples. The majority of articles (n = 36/51) used only one senescence marker to assess the relationship between senescence and chronological age. Overall, 14 different tissue types have been used to assess the relationship between senescence and age with skin (n = 12/51 articles), kidney (n = 11/51 articles), and blood (n = 6/51 articles) utilized most often. Eight articles that have utilized skin, brain, and kidney tissue have provided an in‐depth analysis of these samples based on tissue structure (e.g. dermis vs. epidermis) or cell type (e.g. astrocytes vs. oligodendrocytes).
 
 
2.3 Systematic review: qualitative description
 
Of the 51 articles assessed, 31 reported a significantly positive association between senescence and chronological age for at least one senescence marker while two describe a negative association between senescence and age. Eleven articles also reported a higher level of senescence with chronological age, but these findings were not statistically significant. In addition, while some articles (n = 7/51) described a positive association between senescence and chronological age, quantitative data to support this association have not been provided by the authors. Further information regarding the overall qualitative outcomes for each study can be found in Table S2.
 
 
2.4 Meta‐Analysis: Association between the magnitude of senescence and chronological age
 
Overall, 38 articles provided quantified data that could be used in the meta‐analysis, altogether describing 83 associations between senescence and chronological age. For the detailed meta‐analysis, including all articles and associations, refer to Figures S1 and S2. Figure 2 illustrates the overall correlation between senescence and age, subgrouped by tissue type (2A) and senescence marker type (2B). Overall, a positive correlation between senescence and age was observed (r = .28, p = .000, 95% prediction interval = −0.12 through 0.60, τ2 = 0.042). Pancreas (r = .90, p < .001, one article, two markers, one tissue type), brain (r = .70, p < .001, two articles, one marker, four tissue types), and lung tissue (r = .66, p < .001, one article, two markers, one tissue type) showed the highest correlation between senescence and age. Of the different tissue types assessed, senescence in adipose, gut, prostate, and thymus tissues was not significantly correlated with age. Senescence markers identified as DNA damage markers had the highest correlation with age (r = .69, p < .001), followed by cell cycle regulators (r = .33 p < .001), whereas proliferation markers had the lowest correlation with age (r = .08, p = .43).
 
 
acel13083-fig-0002-m.jpg
 
Figure 2. Overall forest plots of the meta‐analysis for senescence and chronological age subgrouped by tissue (a) and marker type (b).
 
 
Figure 3a describes the association of senescence with age in multiple tissue sections within a specific tissue. Overall, a positive association (r = .31, p < .001), between senescence and age, was observed within the kidney tissue. However, while a positive association was observed between senescence and age within skin sections (dermis, epidermis, facial, and abdominal), significant heterogeneity was also observed (r = .30, p = .03, τ2 = 0.067). The overall association between senescence and age within this analysis also demonstrated significant heterogeneity (r = .25, 95% prediction interval = −0.21 through 0.62, τ2 = 0.049). Figure 3b describes the association of senescence with age where multiple senescent markers have been used in one tissue sample, the gold standard for detecting senescent cells within tissue samples (Gorgoulis et al., 2019). Overall, a positive association between senescence and age was observed irrespective of the marker used to determine senescence. However, significant heterogeneity was observed (r = .24, 95% prediction interval = −0.17 through 0.64, τ2 = 0.040), and within one tissue sample, the expression of senescent markers varied (Song, Yang, Xie, Zang, & Yin, 2008).
 
 
acel13083-fig-0003-m.jpg
 
 
 
Figure 3. Forest plots of the subgroup meta‐analysis of senescence and chronological age for multiple associations extracted from the same article assessing senescence marker expression across tissue sections (a) and senescence marker expression within a tissue sample (b).
 
 
2.5 Meta‐analysis: change in slope of senescence per 10 years of life
 
To determine the magnitude of the change in senescence level with age, a meta‐analysis using mean change in β was performed. For the detailed meta‐analysis, including all articles and associations refer to Figures S3 and S4. Figure 4 illustrates the overall findings of the meta‐analysis subgrouped by tissue and marker type. Overall, there was a significantly higher level of senescence per 10 years of age across tissue samples; the combined overall effect size indicated a 0.16 ± 0.02 standardized unit (S.U) (p < .001, 95% prediction interval = 0.02–0.29, τ2 = 0.0045) and higher magnitude of senescence per 10 years. The highest magnitude of senescence per 10 years of age was observed in brain (β = 6.23 ± 0.55 S.U) and adipose (β = 1.88 ± 0.24 S.U) tissue. Similarly, senescence markers identified as DNA damage markers were higher per 10 years of age (β = 1.99 ± 0.70 S.U), whereas the senescence‐associated secretory phenotype (SASP) had the lowest magnitude of senescence per 10 years of age (β = 0.04 ± 0.08 S.U). Significant heterogeneity (I2 = 81.1%–98.6%) was observed for all tissue types and marker subgroups included in the meta‐analysis.
 
 
acel13083-fig-0004-m.jpg
 
Figure 4. Overall forest plots of meta‐regression analyses for senescence load per 10 years of chronological age, expressed as the change in the overall slope (standardized units), subgrouped by tissue (a), and marker type (b).
 
 
2.6 Publication bias
 
Publication bias was assessed for all outcomes via visually inspecting the asymmetry of the funnel plot (Figure 5). The funnel plot including all data points demonstrated asymmetry. Smaller studies that do not show a positive association between senescent makers and age are underrepresented in this review. This, combined with Egger's regression intercept analysis, indicated publication bias toward the positive finding (p = .001).
 
 
 
 
 
.../...
 
 
 
 
 
 

 







Also tagged with one or more of these keywords: aging, cell cycle proteins, cellular senescence, human immunosenescence

1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users