Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 30 of the 30 references it cites were found in a registry or an index; 15 of the 20 sources its claims cite were checked for support.

Ensemble of Time-Evolving SASP Gene Sets Identifies IGFBP7 and CDKN1A as a Potential Marker Pair for Senescent Fibroblast Subpopulations Across Tissues

Open access

Abstract

The senescence-associated secretory phenotype (SASP) is a hallmark of senescent cells and plays a critical role in the development and progression of various age-related diseases, including cancer, cardiovascular disorders, and neurodegenerative diseases. In this study, we characterize SASP heterogeneity using single-cell RNA sequencing (scRNA-seq) data, focusing on the transcriptional signatures associated with elevated expression of individual SASP genes in mature senescent cells, as well as time-dependent variation in SASP expression across the early and mature senescent states in the WI-38 human lung fibroblast cell line. We generated multiple gene sets, each representing the transcriptional landscape linked to high expression of a specific SASP gene, and integrated them into an ensemble that reflects the temporal dynamics of SASP gene expression. Applying SASP scores derived from this ensemble of gene sets (SASP scores/EGS) to publicly available scRNA-seq datasets from human lung, skin, and eye tissues enabled the identification of senescent fibroblasts and revealed IGFBP7 as a consistently upregulated marker in p21+ or p16+ fibroblasts across diverse human tissues. Our framework supports improved detection of both early and mature fibroblast replicative senescent cells, offering valuable insights into aging and age-related disease research.

Overview

CoChat Summary

TL;DR

The paper builds an ensemble of SASP-based gene sets from WI-38 fibroblast scRNA-seq to capture temporal and subpopulation heterogeneity in senescence and then applies it to human lung, skin, and eye fibroblast datasets. It concludes that IGFBP7 is the most consistent transcriptional marker, while pairing p21/CDKN1A with IGFBP7 helps identify senescent fibroblasts more reliably across tissues.

Key findings

  • In WI-38 fibroblasts, SASP genes were not uniformly expressed: only subsets of cells within senescent stages showed high expression, including distinct early- and mature-senescence subpopulations.
  • The authors constructed 62 paired upregulated and downregulated gene sets from 106 genes of interest and used them to compute senescence scores.
  • Using the ensemble, they selected the 100 lowest-score and 100 highest-score cells and reported perfect accuracy in three-fold cross-validation for these high-confidence cells.
  • In human tissues, the method identified 100 predicted senescent cells out of 11,889 fibroblasts in lung tissue as <1% of fibroblasts, and it found IGFBP7 to be consistently upregulated across lung, skin, and eye fibroblasts; in lung tissue, 119 cells were top-ranked by SASP scores/EGS, and all 119 had zero proliferation score, whereas 7 of 119 top SenMayo-score cells had nonzero proliferation scores.

Methods

  • The study used scRNA-seq time-course data from WI-38 human lung fibroblasts across multiple population doublings (PDL 25, 29, 33, 46, 50) and public DISCO-annotated fibroblast scRNA-seq datasets from lung, skin, and eye tissues.
  • Genes of interest came from SASP literature and SenMayo: the paper reports 106 genes of interest, including SASP factors, CDKN1A/CDKN2A/TP53, and other related genes; gene sets were built by comparing cells overexpressing each gene (z-score > 1.5) against control PDL_25p cells.
  • Senescence classification used mean z-scores of upregulated minus downregulated genes, aggregation across the 62 gene-set pairs, a threshold of 2.1 for WI-38 cross-validation, and tissue analyses that selected the n lowest and n highest scoring cells, followed by differential-expression analysis with Seurat’s FindMarkers().

Limitations

  • The paper states that SASP expression differs between cell line and tissue, and that markers like IGFBP7 alone may not be sufficient because other stressed cells can also upregulate it.
  • The authors note misclassifications due to overlap between score distributions and suggest these may reflect ambiguous transitional cells between normal and senescent states.
  • Evident from the text: the approach relies on public scRNA-seq datasets and focuses on fibroblasts from a small set of tissues, so generalization beyond those datasets and cell types is not established in the paper.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3390/ijms27073012

Cites

30 of 30 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Transcriptional heterogeneity of SASP genes and other genes of interest in the WI-38 human lung fibroblast cell line. (A) Flow diagram across temporal PDL groups shows the number of cells with high gene expression (z-score > 1.5) for five selected genes (i.e., CDKN1A (p21), CDKN2A (p16), MIF, IGFBP7, and SERPINE2). The heatmap displays the gene expression in WI-38 cells at PDL_50np of the five genes and some additional genes of interest (i.e., IGFBP3, IGFBP4, IGFBP5, IGFBP6, TIMP1, TIMP2, COL1A1, CTSB, PAPPA). (B) Percentage of cells with high SASP gene expression across PDL groups. (C) Number of cells with high gene expression and co-expression at PDL_50np, along with dotted lines indicating the intersection of sets when considering the five genes. For example, there were 43 cells where CDKN2A, IGFBP7, and SERPINE2 were highly expressed.
Figure 1 Transcriptional heterogeneity of SASP genes and other genes of interest in the WI-38 human lung fibroblast cell line. (A) Flow diagram across temporal PDL groups shows the number of cells with high gene expression (z-score > 1.5) for five selected genes (i.e., CDKN1A (p21), CDKN2A (p16), MIF, IGFBP7, and SERPINE2). The heatmap displays the gene expression in WI-38 cells at PDL_50np of the five genes and some additional genes of interest (i.e., IGFBP3, IGFBP4, IGFBP5, IGFBP6, TIMP1, TIMP2, COL1A1, CTSB, PAPPA). (B) Percentage of cells with high SASP gene expression across PDL groups. (C) Number of cells with high gene expression and co-expression at PDL_50np, along with dotted lines indicating the intersection of sets when considering the five genes. For example, there were 43 cells where CDKN2A, IGFBP7, and SERPINE2 were highly expressed.CC BY · PubMed Central
Figure 2 Construction and application of an ensemble of gene sets to identify senescent cells in human tissues. (A) Flow diagram outlines the process of constructing the ensemble gene sets. Upregulated and downregulated gene sets for each gene of interest, including SASP genes, were obtained by comparing cells in which each gene was overexpressed (z-score > 1.5) against control cells from the PDL_25p group. Lung fibroblast senescence scores for each gene set were calculated as the mean z-score of upregulated genes minus the mean z-score of downregulated genes. Z-scores were computed using the mean and standard deviation of gene expression in PDL_25p cells, which were stored during the earlier stage of ensemble gene set construction. Color scale represents gene expression levels, ranging from dark blue (lowest expression) to dark red (highest expression). (B) To identify more confidently classified normal and senescent fibroblasts in human lung tissue, we selected 100 cells from each group with the lowest and highest senescence scores, respectively. (C) Euler diagram illustrates the co-expression patterns of CDKN1A, IGFBP7, and SERPINE2 in 100 senescent/stressed fibroblasts in lung tissue. (D) Log-normalized expression values of IGFBP7 are shown for two groups: control (100 normal fibroblasts) and senescent (100 senescent/stressed fibroblasts) in lung tissue. The median expression levels between the two groups were significantly different (Wilcoxon rank-sum test, **** p < 1 × 10−9).
Figure 2 Construction and application of an ensemble of gene sets to identify senescent cells in human tissues. (A) Flow diagram outlines the process of constructing the ensemble gene sets. Upregulated and downregulated gene sets for each gene of interest, including SASP genes, were obtained by comparing cells in which each gene was overexpressed (z-score > 1.5) against control cells from the PDL_25p group. Lung fibroblast senescence scores for each gene set were calculated as the mean z-score of upregulated genes minus the mean z-score of downregulated genes. Z-scores were computed using the mean and standard deviation of gene expression in PDL_25p cells, which were stored during the earlier stage of ensemble gene set construction. Color scale represents gene expression levels, ranging from dark blue (lowest expression) to dark red (highest expression). (B) To identify more confidently classified normal and senescent fibroblasts in human lung tissue, we selected 100 cells from each group with the lowest and highest senescence scores, respectively. (C) Euler diagram illustrates the co-expression patterns of CDKN1A, IGFBP7, and SERPINE2 in 100 senescent/stressed fibroblasts in lung tissue. (D) Log-normalized expression values of IGFBP7 are shown for two groups: control (100 normal fibroblasts) and senescent (100 senescent/stressed fibroblasts) in lung tissue. The median expression levels between the two groups were significantly different (Wilcoxon rank-sum test, **** p < 1 × 10−9).CC BY · PubMed Central
Figure 3 Heterogeneity of gene expression in SASP genes and other genes of interest in various human tissue fibroblasts. (A) A flow diagram illustrates the size of groups in WI-38 cells at PDL_50np, as well as fibroblasts in human lung, skin, and eye tissues, where the size of gene groups for five selected genes (i.e., CDKN1A (p21), CDKN2A (p16), MIF, IGFBP7, and SERPINE2) is the number of cells with high gene expression (z-score > 1.5) for a gene or for co-expression of multiple genes. This panel also includes a UMAP visualization of senescent p21+ fibroblasts predicted by SASP scores/EGS in lung tissue. (B) Log-normalized counts of gene expression are displayed for selected SASP genes, such as IGFBP7 and MMP1, in normal fibroblasts in WI-38 (PDL_25p) and mature senescent fibroblasts in WI-38 (PDL_50np), as well as in control (normal) fibroblasts and senescent/stressed fibroblasts from human lung, skin, and eye tissues. Red font color in gene names indicates the marker pair IGFBP7 and CDKN1A, and green font color highlights the genes of interest mentioned in the main text.
Figure 3 Heterogeneity of gene expression in SASP genes and other genes of interest in various human tissue fibroblasts. (A) A flow diagram illustrates the size of groups in WI-38 cells at PDL_50np, as well as fibroblasts in human lung, skin, and eye tissues, where the size of gene groups for five selected genes (i.e., CDKN1A (p21), CDKN2A (p16), MIF, IGFBP7, and SERPINE2) is the number of cells with high gene expression (z-score > 1.5) for a gene or for co-expression of multiple genes. This panel also includes a UMAP visualization of senescent p21+ fibroblasts predicted by SASP scores/EGS in lung tissue. (B) Log-normalized counts of gene expression are displayed for selected SASP genes, such as IGFBP7 and MMP1, in normal fibroblasts in WI-38 (PDL_25p) and mature senescent fibroblasts in WI-38 (PDL_50np), as well as in control (normal) fibroblasts and senescent/stressed fibroblasts from human lung, skin, and eye tissues. Red font color in gene names indicates the marker pair IGFBP7 and CDKN1A, and green font color highlights the genes of interest mentioned in the main text.CC BY · PubMed Central

Verification

Claims the paper makes

14 Valid · 0 Not valid · 6 Not checked · 84 Own results and interpretations

53 more claims

Create a free account to see all 104 claims and to verify the 1 reference still open.

See all claims — free account
  • It may state two separate claims as one. 33
  • Probably not a substantial claim. 16
  • The quote only partly states this claim. 2

Create a free account to see these claims.

Citations

Cites

Show all 29 references

Cited by

See all 1

Metadata

Authors
Kim, H., Kummerfeld, E., Niedernhofer, L. J., Aliferis, C., Robbins, P. D., & Wang, J.
Affiliations
University of Minnesota (US)
Corresponding author
Wang, Jinhua
Venue
International Journal of Molecular Sciences
Publisher
MDPI AG
Published
2026-03-26
Volume
27
Issue
7
Pages
3012
ISSN
1422-0067 (ISSN-L)
OpenAlex
W7140452661
Cited by
1
References
29
Citations by year
2026: 1
FWCI
4.51
Type
journal article
Language
en
Open access
Locations
  • International Journal of Molecular Sciences (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (publishedVersion, cc-by)
Funders
  • National Institutes of Health (5P30CA077598)
  • National Institute on Aging (5U54AG079754-04)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Cell Biology › Telomeres, Telomerase, and Senescence
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Aging › Genetics, Aging, and Longevity in Model Organisms
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Single-cell and spatial transcriptomics
Keywords
senescence-associated secretory phenotype, cellular senescence, senescent fibroblasts, IGFBP7, p21, senescence markers, senescence heterogeneity, replicative senescence, p16, aging, age-related diseases, WI-38 fibroblasts, single-cell transcriptomics
MeSH terms
Senescence-Associated Secretory Phenotype, Single-Cell Gene Expression Analysis, Cell Line, Fibroblasts, Humans, Lung, Organ Specificity, Biomarkers, Cellular Senescence, Insulin-Like Growth Factor Binding Proteins, Gene Expression Profiling, Cyclin-Dependent Kinase Inhibitor p21, Single-Cell Analysis

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Ensemble of Time-Evolving SASP Gene Sets Identifies IGFBP7 and CDKN1A as a Potential Marker Pair for Senescent Fibroblast Subpopulations Across Tissues. https://app.cochat.ai/papers/10.3390/ijms27073012
The paper
Kim, H., Kummerfeld, E., Niedernhofer, L. J., Aliferis, C., Robbins, P. D., & Wang, J. (2026). Ensemble of Time-Evolving SASP Gene Sets Identifies IGFBP7 and CDKN1A as a Potential Marker Pair for Senescent Fibroblast Subpopulations Across Tissues. International Journal of Molecular Sciences. https://doi.org/10.3390/ijms27073012

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.