Exosomes from Human Adipose Tissue-Derived Mesenchymal Stem Cells Promote Epidermal Barrier Repair by Inducing de Novo Synthesis of Ceramides in Atopic Dermatitis

Abstract

Atopic dermatitis (AD) is a multifactorial, heterogeneous disease associated with epidermal barrier disruption and intense systemic inflammation. Previously, we showed that exosomes derived from human adipose tissue-derived mesenchymal stem cells (ASC-exosomes) attenuate AD-like symptoms by reducing multiple inflammatory cytokine levels. Here, we investigated ASC-exosomes' effects on skin barrier restoration by analyzing protein and lipid contents. We found that subcutaneous injection of ASC-exosomes in an oxazolone-induced dermatitis model remarkably reduced trans-epidermal water loss, while enhancing stratum corneum (SC) hydration and markedly decreasing the levels of inflammatory cytokines such as IL-4, IL-5, IL-13, TNF-α, IFN-γ, IL-17, and TSLP, all in a dose-dependent manner. Interestingly, ASC-exosomes induced the production of ceramides and dihydroceramides. Electron microscopic analysis revealed enhanced epidermal lamellar bodies and formation of lamellar layer at the interface of the SC and stratum granulosum with ASC-exosomes treatment. Deep RNA sequencing analysis of skin lesions demonstrated that ASC-exosomes restores the expression of genes involved in skin barrier, lipid metabolism, cell cycle, and inflammatory response in the diseased area. Collectively, our results suggest that ASC-exosomes effectively restore epidermal barrier functions in AD by facilitating the de novo synthesis of ceramides, resulting in a promising cell-free therapeutic option for treating AD.

Keywords: ASC-exosomes; anti-inflammation; atopic dermatitis; ceramides; lamella body; restoration; skin barrier.

Figure 1

Characterization of adipose tissue-derived MSCs (ASC)-exosomes. (A) Representative histogram of particle concentration and size distribution of ASC-exosomes measured by nanoparticle tracking analysis (NTA). (B) Representative Cryo-TEM image of ASC-exosomes. Scale bar: 100 nm. (C) ASC-exosomes were analyzed using western blots for the presence of exosomal markers such as Alix, TSG101, CD9, and CD81. (D) ASC-exosomes were analyzed by western blotting for the negative markers of exosomes such as GM130 and Calnexin. Abbreviation: ASC, human adipose tissue-derived mesenchymal stem cell. (E) Surface signature of ASC-exosomes quantified by MACSPlex Exosome Kit (human) in conjunction with flow cytometry. Data indicate APC median signal intensities of ASC-exosomes incubated with the 39 capture beads and stained with a mixture of CD9-, CD63-, and CD81-APC antibodies. Background was corrected by subtraction of median fluorescence APC intensity. n = 10. MW, molecular weight.

Figure 2

Multi-layer omics analysis of ASC-exosomes. (A–C) Proteomic analysis of ASC-exosomes by LC-MS/MS in three independent batches of ASC-exosomes. (A) Venn diagram showing the number of exosomal proteins that were either unique or overlapping in each indicated batch. (B,C) Gene Ontology (GO) analysis of 471 commonly found exosomal proteins. Enrichment of GO molecular function and cellular component performed using DAVID Bioinformatics resources 6.8. Vertical red lines represent the cutoff for significance of -log10(p-value) with Benjamini-Hochberg correction. (D–F) Lipidomics of ASC-exosomes. (D) Lipid composition of ASC-exosomes and their originating cells. A-Venn diagram showing the number of species quantified in this study. (E) Lipid classes detected in ASCs and ASC-exosomes. (F) Enrichment of lipid classes in ASCs or ASC-exosomes calculated as mol% of lipids in these samples. PC: phosphatidylcholine, PE: phosphatidylethanolamine, PI: phosphatidylinositol, SM: sphingomyelin, NEFA: non esterified fatty acid, PS: phosphatidylserine, CER: ceramide, TG: triacylglycerol, DG: Diacylglycerol, Chol ester: cholesterol ester, Chol: Cholesterol, PG: phosphatidylglycerol, dhSM: dihydrosphingomyelin, CL: cardiolipin, dhCER: dihydroceramide, Hex-Cer: hexosylceramides, PA: phosphatidic acid, MG: monoradylglycer.

Figure 3

ASC-exosomes improve atopic dermatitis (AD) induced by repeated exposure to Ox. (A) Schematic diagram of the study protocol. The first two days 2% Ox was applied, before leaving 5 days for recovery. Dexamethasone was topically applied thrice a week as a positive control. (B) Representative dorsal skin photographs of each treatment group showing comparison of AD-like skin lesions (C) Representative images of hematoxylin and eosin (H&E)-stained epidermal histological sections from dorsal skin of mice from different treatment groups showing thickness of epidermis. The sections were visualized with a 20x magnification. Box and whisker plot of (D) epidermal thickness, (E) TEWL, and (F) stratum corneum (SC) hydration. The median and the 5th and 96th percentile are shown. Error bars depict the SEM. ***p < 0.001 vs. vehicle. (G) Effects of ASC-exosomes and dexamethasone on body weight changes. Note that the treatment of dexamethasone induced significant body weight loss. n = 12 per group. ***p < 0.001 determined by one-way ANOVA.

Figure 4

ASC-Exosomes reduce the level of multiple inflammatory cytokines in AD-like skin lesion of SKH-1 mice. The concentrations of (A) IL-4, (B) IL-5, (C) IL-13, (D) TNF-α, (E) IFN-γ, (F) IL-17, (G) IgE, and (H) TSLP in serum or lesional skin tissues were detected by ELISA. (I) Toluidine blue staining of the dorsal skin sections showing infiltration of mast cells into the AD-like skin lesions. Error bars depict the SEM. n  =  8. *: p  <  0.05, **: p  <  0.01, ***: p < 0.001 vs. vehicle. Dex: dexamethasone.

Figure 5

ASC-exosomes increase the level of ceramides in skin lesions. (A–J) Mass spectrometry analysis of ceramide and dihydroceramide species. Lipids were extracted from the homogenate of dorsal skins corresponding to 1 mg protein. n = 8 per group. Error bars depict the SEM. *: p  <  0.05, **: p  <  0.01, ***: p < 0.001 vs. vehicle. Dex: dexamethasone.

Figure 6

ASC-exosomes restore oxazole-induced defects in the lamellar secretory system. Electron micrographs of epidermis. Inserts show high magnification of lamellar bodies and/or processed lamellar lipids as indicated. Note that lamellar bodies are more abundant at the SG-SC interface in the ASC-exosomes treated group compared with the vehicle and dexamethasone groups. KHG: keratohyaline granules, C: corneocyte, arrows: lamellar bodies, arrow heads: lipid secretion points, *: lipid phase separation (indicating incomplete processing). Scale bar: 100 nm.

Figure 7

ASC-exosomes modulate distinct gene expression program during AD pathogenesis. (A) The principal component analysis (PCA) showing the first two principal components of RNA-Seq data regarding their correlation. (B) Heatmap showing 1802 differential expressed gene (DEGs). (C) Volcano plot shows the DEGs in vehicle, ASC-exosomes-, or dexamethasone-treated group. Red dots indicate significantly up-regulated genes and blue dots indicate down-regulated genes.

Figure 8

ASC-exosomes normalize altered gene expression in a murine model of allergic dermatitis. Bubble plot shows the significantly (p < 0.05) altered GO biological processes in vehicle, ASC-exosomes-, and dexamethasone-treated groups compared with the indicated control group. The size of the bubble is proportional to the number of genes. Red bubbles indicate up-regulation and blue bubbles denote down-regulation. GO analysis was performed using DAVID bioinformatics resources 6.8 (https://david.ncifcrf.gov).

Figure 9

Heat map of DEGs between vehicle and ASC-exosomes that mapped to the indicated GO in Figure 8. (A) Skin barrier. (B) Lipid metabolism. (C) Cell cycle. (D) Immune response.

Figure 10

ASC-exosomes do not affect the sphingolipid hydrolysis pathway. (A–G) The lipid profiles were determined by LC-MS/MS measurements. Note that dexamethasone treatment significantly reduces the levels of sphingomyeline except for those with C18 and C20 acyl chains. Error bars depict the SEM. n  =  8. *: p < 0.05, **: p  <  0.01, ***: p < 0.001 vs. vehicle. Dex, dexamethasone.

Figure 11

ASC-exosomes regulate the salvage pathway to increase ceramides in skin lesion. (A–D) The levels of sphingosine and S1P, and the activities of Sphk1 and S1PL were determined by LC-MS/MS measurements. Error bars depict the SEM. n  =  8. *: p < 0.05, **: p  <  0.01, ***: p < 0.001 vs. vehicle. Dex, dexamethasone.