A subset of human dermal fibroblasts overexpressing Cockayne syndrome group B protein resist UVB radiation-mediated premature senescence

Ultraviolet B (UVB) radiation is a major contributor to skin photo-ageing. Although mainly absorbed by the epidermis, UVB photons managing to penetrate the upper dermis affect human dermal fibroblasts (HDFs), leading, among others, to the accumulation of senescent cells. In vitro studies have shown that repeated exposures to subcytotoxic UVB radiation doses provoke HDFs’ premature senescence shortly after the end of the treatment period. Here, we found that repetitive exposures to non-cytotoxic UVB radiation doses after several days lead to mixed cultures, containing both senescent cells and fibroblasts resisting senescence. “Resistant” fibroblasts were more resilient to a novel intense UVB radiation stimulus. RNA-seq analysis revealed that ERCC6, encoding Cockayne syndrome group B (CSB) protein, is upregulated in resistant HDFs compared to young and senescent cells, also confirmed at the protein level. CSB was found to be a key molecule conferring protection towards UVB cytotoxicity, as siRNA-mediated CSB loss-of-expression rendered HDFs significantly more susceptible to a high UVB radiation dose. In accordance, cells from a CSB-deficient patient were found to be dramatically more photosensitive. UVB-resistant HDFs remained normal (able to undergo replicative senescence) and non-tumorigenic. Even though they formed a distinct population in-between young and senescent cells, resistant HDFs retained numerous tissue-impairing characteristics of the senescence-associated secretory phenotype, including increased matrix metalloprotease activity and promotion of epidermoid tumor xenografts in immunodeficient mice. Collectively, here we describe a novel subpopulation of HDFs showing increased resistance to UVB-mediated premature senescence, as well as undesirable traits that may negatively affect skin homeostasis. To elucidate the molecular profile of UVB-resistant human dermal fibroblasts, emerging after exposure of the cells to 10 non-cytotoxic doses of UVB radiation. Total RNA was extracted from three samples per condition (young cells, UVB-reistant cells, ionizing radiation-induced senescent cells). For library preparation, the 3′ mRNA-Seq Library Prep Kit Protocol for Ion Torrent. Quality and quantity of libraries were assessed on a bioanalyzer using the DNA High Sensitivity Kit reagents and protocol. The quantified libraries were pooled at a final concentration of 7 pM, templated and enriched on the Ion Proton One Touch system using the Ion PI™ Hi-Q™ OT2 200 Kit and sequenced using the Ion PI™ Hi-Q™ Sequencing 200 Kit on Ion Proton PI™ V2 chips on an Ion Proton™ System, according to the manufacturer’s instructions FASTQ files were trimmed with Trim Galore (v.0.6.51) to remove low quality read ends using a Phred score of 20. Subsequently, a two steps alignment procedure was applied. Pre-processed reads were aligned against the hg38 reference genome (Ensembl) with HISAT2 (v.2.2.1) and then the reads left unmapped were subjected to a second alignment round using BOWTIE2 (v.2.3.5.1) with the -local and -very-sensitive local switches turned on. Downstream analysis of the resulting BAM files was performed with metaseqR2 R package (v.1.9.2). The filtered gene counts table was subjected to differential expression analysis for the contrasts resistant vs. young cells, IS vs. young cells and IS vs. resistant cells using sequentially DESeq, edgeR, limma, NBPSeq, and NOISeq methods supported by metaseqR2. Their p-values were then combined by the PANDORA algorithm to account, among others, for the false positives reported Benjamini-Hochberg-corrected PANDORA p-values of less than 0.05 and absolute fold change of at least 1.2 were used as differential expression thresholds.