Extracellular Vesicles Derived from Senescent Fibroblasts Attenuate the Dermal Effect on Keratinocyte Differentiation

Abstract

The skin is a multilayered and primary defensive organ. Intimate intercellular communication in the skin is necessary to ensure effective surveillance. Extracellular vesicles (EVs) are being explored for their involvement in intercellular skin communication. The aim of this study was to evaluate how human dermal fibroblasts (HDFs) accelerate EV production during senescence and the effects of senescence-associated EVs on epidermal homeostasis. Replicative senescent HDFs were assessed with senescence-associated β-galactosidase staining and the expression of senescence-related markers. Isolated EVs were characterized by dynamic light scattering and EV marker expression. EVs secreted from untreated young or senescent HDFs, or from those treated with a nSMase inhibitor, antioxidant, and lysosomal activity regulators, were determined by sandwich ELISA for CD81. Human epidermal keratinocytes were treated with young- and senescent HDF-derived EVs. Compared to young HDFs, senescent HDFs produced relatively high levels of EVs due to the increased nSMase activity, oxidative stress, and altered lysosomal activity. The nSMase inhibitor, antioxidant, and agents that recovered lysosomal activity reduced EV secretion in senescent HDFs. Relative to young HDF-derived EVs, senescent HDF-derived EVs were less supportive in keratinocyte differentiation and barrier function but increased proinflammatory cytokine IL-6 levels. Our study suggests that dermis-derived EVs may regulate epidermal homeostasis by reflecting cellular status, which provides insight as to how the dermis communicates with the epidermis and influences skin senescence.

Keywords: epidermal homeostasis; exosome; extracellular vesicle; human dermal fibroblast; lysosomal activity; senescence.

Figure 1

EV secretion was increased in replicative senescent dermal fibroblasts. (A) Senescence-associated (SA)-β-gal assay in young HDFs (PDL < 10) and senescent HDFs (PDL > 50). Scale bar = 50 μm. Data are means ± SD of three independent experiments on two independent senescent cell lines (*** p < 0.001). (B) Western blot analysis of p21 and p16 in young versus senescent HDFs. GAPDH was the loading control. (C) Levels of EVs derived from equal numbers of young and senescent HDFs. Protein concentrations in isolated EVs were determined by BCA assay. Data are means ± SD of four independent experiments using two independent senescent cell lines (*** p < 0.001). (D) Dynamic light scattering analysis of EVs derived from young and senescent HDFs. (E) Western blot analyses of EV markers. Five micrograms of EV protein were subjected to immunoblot analysis with anti-CD81, anti-CD9, anti-Alix, and anti-HSP90. GAPDH was the loading control.

Figure 2

EV biogenesis increased in senescent dermal fibroblasts. (A) Schematic image of sandwich ELISA using anti-CD81 antibody to detect EVs. (B) Conditioned media used to culture equal numbers of young and senescent HDFs were harvested at the indicated time points. EV levels were quantitated by sandwich ELISA for CD81. Data are means ± SD of three independent experiments using a senescent cell line (*** p < 0.001). (C) HDFs were labeled with a fluorescent lipid molecule N-Rh-PE (red) for the indicated time periods and co-stained with anti-CD63 antibody (green). Nuclei were stained with 4′6-diamidino-2-phenylindole (DAPI; blue). Fluorescence images were taken under a confocal microscope. Representative images are shown. Scale bar = 20 μm.

Figure 3

Enhanced nSMase activity in senescent HDFs increased EV production. (A) nSMase activity was determined for equal numbers of young and senescent HDFs with the nSMase activity assay kit. (B) Senescent HDFs were treated with vehicle (-) or GW4869 (5 μM) for 24 h. Relative EV quantity was analyzed by sandwich ELISA for CD81 in each conditioned medium. C–D Equal numbers of senescent HDFs were treated with vehicle (-) or N-acetyl cysteine (NAC; 5 mM) for 24 h and nSMase activity was measured (C). Senescent HDFs were treated with NAC and the relative EV quantity was measured by sandwich ELISA for CD81 in each conditioned medium (D). Data are means ± SD of three independent experiments using a senescent cell line (* p < 0.05; ** p < 0.01; *** p < 0.001).

Figure 4

Dysfunctional lysosomal activity in senescent HDFs increased EV secretion. (A) Lysosomal pH in young and senescent HDFs was measured with a LysoSensor Yellow/Blue DND-160. (B) Representative confocal images for LAMP1 (green) and TFEB (red) in young and senescent HDFs. Nuclei were stained with DAPI (blue). Scale bar = 10 μm. (C) Senescent HDFs were treated with vehicle (-) or KU60019 (0.5 μM) for 25 d and lysosomal pH was measured. (D) Confocal images for LAMP1 (green) and TFEB (red) in senescent HDFs either untreated or treated with KU60019. Nuclei were stained with DAPI (blue). Scale bar = 10 μm. (E) Senescent HDFs were treated with vehicle (-) or KU60019 (0.5 μM) for 25 d and relative EV quantity was measured by sandwich ELISA for CD81 in each conditioned medium. (F) Young HDFs were treated with vehicle (-) or bafilomycin A (100 nM) for 24 h and relative EV quantity was measured by sandwich ELISA for CD81 in each conditioned medium. Data (in A,C,E,F) are means ± SD of three independent experiments using a senescent cell line (** p < 0.01; *** p < 0.001).

Figure 4

Dysfunctional lysosomal activity in senescent HDFs increased EV secretion. (A) Lysosomal pH in young and senescent HDFs was measured with a LysoSensor Yellow/Blue DND-160. (B) Representative confocal images for LAMP1 (green) and TFEB (red) in young and senescent HDFs. Nuclei were stained with DAPI (blue). Scale bar = 10 μm. (C) Senescent HDFs were treated with vehicle (-) or KU60019 (0.5 μM) for 25 d and lysosomal pH was measured. (D) Confocal images for LAMP1 (green) and TFEB (red) in senescent HDFs either untreated or treated with KU60019. Nuclei were stained with DAPI (blue). Scale bar = 10 μm. (E) Senescent HDFs were treated with vehicle (-) or KU60019 (0.5 μM) for 25 d and relative EV quantity was measured by sandwich ELISA for CD81 in each conditioned medium. (F) Young HDFs were treated with vehicle (-) or bafilomycin A (100 nM) for 24 h and relative EV quantity was measured by sandwich ELISA for CD81 in each conditioned medium. Data (in A,C,E,F) are means ± SD of three independent experiments using a senescent cell line (** p < 0.01; *** p < 0.001).

Figure 5

Senescent HDF-derived EVs attenuate the dermal effect on keratinocyte differentiation but evoke proinflammatory cytokine IL-6. Human epidermal keratinocytes (HEKs) in culture were treated with EVs derived from 5 × 10⁵ young and senescent HDFs for 4 days. HEKs cultured without EV treatment for 4 d served as the control (-). Cells were analyzed for mRNA (A) and protein (B) expression of keratinocyte differentiation-related (KRT1, LOR) or barrier function-related (BLMH) markers by qRT-PCR and western blot analyses, respectively. The mRNA levels were normalized to those of RPL13A. GAPDH was the loading control for western blot analysis. KRT1, keratin 1; LOR, loricrin; BLMH, bleomycin hydrolase. (C,D) HEKs were treated with EVs derived from 5 × 10⁵ young and senescent HDFs. Conditioned media at 24 h and 4 d post-treatment with EVs were harvested for the determination of secreted IL-8 (C) and IL-6 (D) levels, respectively, with specific ELISA kits. Data (in A,C,D) are means ± SD of three independent experiments using two senescent cell lines (* p < 0.05; ** p < 0.01; *** p < 0.001; ns, not significant).