Meta-analysis of senescent cell secretomes to identify common and specific features of the different senescent phenotypes: a tool for developing new senotherapeutics

Abstract

DNA damage resulting from genotoxic injury can initiate cellular senescence, a state characterized by alterations in cellular metabolism, lysosomal activity, and the secretion of factors collectively known as the senescence-associated secretory phenotype (SASP). Senescence can have beneficial effects on our bodies, such as anti-cancer properties, wound healing, and tissue development, which are attributed to the SASP produced by senescent cells in their intermediate stages. However, senescence can also promote cancer and aging, primarily due to the pro-inflammatory activity of SASP.Studying senescence is complex due to various factors involved. Genotoxic stimuli cause random damage to cellular macromolecules, leading to variations in the senescent phenotype from cell to cell, despite a shared program. Furthermore, senescence is a dynamic process that cannot be analyzed as a static endpoint, adding further complexity.Investigating SASP is particularly intriguing as it reveals how a senescence process triggered in a few cells can spread to many others, resulting in either positive or negative consequences for health. In our study, we conducted a meta-analysis of the protein content of SASP obtained from different research groups, including our own. We categorized the collected omic data based on: i) cell type, ii) harmful agent, and iii) senescence stage (early and late senescence).By employing Gene Ontology and Network analysis on the omic data, we identified common and specific features of different senescent phenotypes. This research has the potential to pave the way for the development of new senotherapeutic drugs aimed at combating the negative consequences associated with the senescence process. Video Abstract.

Keywords: Meta-analysis; SASP; Secretome; Senescence.

Fig. 1

Overview of the data collection and information. A The flowchart illustrates the inclusion and exclusion process. B The characteristics of protein lists are presented, including species, cell type, elapsed time since stress, and stressor type

Fig. 2

Similarity analysis at different time points since senescence induction. A The data are classified, and the number of protein lists obtained from senescent human cells is indicated. B A box plot displays the overlap rates at each time point. Statistical significance is denoted by asterisks (*p < 0.05, **p < 0.01, ***p < 0.001). C A heatmap depicts the overlap rates, with grey color indicating overlap rates calculated from the same protein lists. The color label positioned at the bottom and right side of the heatmap represents the characteristics of the data

Fig. 3

Similarity analysis among different stressor types. A The data are classified, and the number of protein lists obtained from senescent human cells is indicated. B A box plot displays the overlap rates for each stressor type. Statistical significance is denoted by asterisks (*p < 0.05, **p < 0.01, ***p < 0.001). C A heatmap depicts the overlap rates, with grey color indicating overlap rates calculated from the same protein lists. The color label positioned at the bottom and right side of the heatmap represents the characteristics of the data

Fig. 4

Gene ontology analysis of SASP classified by time series. A A box plot is presented to show the occurrence rates at each time point. The horizontal dashed line represents the threshold for selecting common ontology occurrence rates higher than 75%. B A Venn diagram is provided to identify the common and specific ontologies among the different time points. The names of proteins with central roles in each ontology and frequently included in datasets at each time point are indicated. C The main outcomes of the GO analysis are summarized. Common and specific ontologies are categorized based on their functions and displayed in the box

Fig. 5

REACTOME analysis of SASP classified by time series. A A box plot is displayed to illustrate the occurrence rates at each time point. The horizontal dashed line represents the threshold for selecting common pathway occurrence rates higher than 75%. B A Venn diagram is utilized to identify the common and specific pathways among the different time points. The names of proteins with central roles in each ontology and frequently included in datasets at each time point are indicated. C The main outcomes of the REACTOME analysis are summarized. Common and specific pathways are categorized based on their functions and presented in the box

Fig. 6

Gene ontology analysis of SASP classified by stressor type. A A box plot is presented to display the occurrence rates for each stressor type. The horizontal dashed line represents the threshold for selecting common ontology occurrence rates higher than 75%. B A Venn diagram is utilized to identify the common and specific ontologies among the different stressor types. The names of proteins with central roles in each ontology and frequently included in datasets at each time point are indicated. C The main outcomes of the GO analysis are summarized. Common and specific ontologies are categorized based on their functions and presented in the box

Fig. 7

REACTOME analysis of SASP classified by stressor type. A A box plot is presented to illustrate the occurrence rates for each stressor type. The horizontal dashed line represents the threshold for selecting common pathway occurrence rates higher than 75%. B A Venn diagram is used to identify the common and specific pathways among the different stressor types. The names of proteins with central roles in each ontology and frequently included in datasets at each time point are indicated. C The main outcomes of the REACTOME analysis are summarized. Common and specific pathways are categorized based on their functions and presented in the box