Transfer learning identifies in vivo senescence heterogeneity and contributions to vascularization and matrix production across species and diverse pathologies

Abstract

Cellular senescence is a state of permanent growth arrest that plays an important role in wound healing, tissue fibrosis, and tumor suppression. Despite senescent cells’ (SnC) pathological role and therapeutic interest, their phenotype in vivo remains poorly defined. Here, we developed an in vivoderived senescence signature using a foreign body response (FBR) fibrosis model in a SnC reporter mouse. We identified pericytes and “cartilage-like” fibroblasts as senescent and defined cell typespecific senescence associated secretory phenotypes (SASP). Transfer learning and senescence scoring identified these two SnC populations along with endothelial and epithelial SnCs in new and publicly available murine and human data single cell RNAseq (scRNAseq) datasets from diverse pathologies. Signaling analysis uncovered crosstalk between SnCs and myeloid cells via an IL34- CSF1R-TGFßR signaling axis, contributing to tissue balance of vascularization and matrix production. Overall, our study provides a senescence signature and a computational approach that may be broadly applied to identify transcriptional profiles and SASP factors produced by SnCs that regulate tissue structure and pathology.

Senescent and disease-associated microglia are modifiable features of aged brain white matter

“…4A ). We developed a 50-gene custom panel of senescence-associated genes based on a two-step process of (1) selection of genes from published senescence profiles 20 , 33 , 34 , 35 and (2) confirmation of brain cell expression in aged brain single-cell RNA sequencing and spatial transcriptome datasets 30 , 36 , 37 , which we integrated in tandem with the 950-plex CosMx mouse neuroscience panel ( Extended Data Table 3 ). Microglia were identified based on previously established marker selection including Csf1r , Ctss , Cx3cr1 , Hexb , Selplg , Itgam , P2ry12, Tmem119 , and Trem2 38 , 39 .…”

Substrate stiffness modulates the emergence and magnitude of senescence phenotypes in dermal fibroblasts

“…While it appears that at the protein level the phenotypes do not manifest, the observations that the gene programs are initiated bodes well for RNA-based signatures and identification of senescence in softer matrices. As such, approaches using resources like SenMayo and SenSIG, which are RNA-based signature may be useful 42,43 .…”

Immunoengineering Biomaterials for Musculoskeletal Tissue Repair across Lifespan

“…Cherry et al expanded on this work, demonstrating that heterogeneous senescent populations with discrete phenotypes emerge following biomaterial implantation. [41] Using a novel transgenic p16-Cre;Ai14 mouse strain, they developed a bulk RNA-seq profile for in vivo FBRassociated senescence. Then, they applied a computational transfer learning technique to identify senescence phenotypes in single-cell RNA-seq datasets.…”

“…Adapted with permission. [41] Copyright 2023, Springer Nature BV. B) SnC subtypes identified from single cell dataset in A.…”