Canine Mesenchymal-Stem-Cell-Derived Extracellular Vesicles Attenuate Atopic Dermatitis

Abstract

Atopic dermatitis (AD) is a chronic inflammatory skin disease that is associated with systemic inflammation and immune modulation. Previously, we have shown that extracellular vesicles resulting from human adipose-tissue-derived mesenchymal stem cells (ASC-EVs) attenuated AD-like symptoms by reducing the levels of multiple inflammatory cytokines. Here, we aimed to investigate the improvement of canine AD upon using canine ASC-exosomes in a Biostir-induced AD mouse model. Additionally, we conducted in vivo toxicity studies to determine whether they targeted organs and their potential toxicity. Firstly, we isolated canine ASCs (cASCs) from the adipose tissue of a canine and characterized the cASCs-EVs. Interestingly, we found that cASC-EVs improved AD-like dermatitis and markedly decreased the levels of serum IgE, ear thickness, inflammatory cytokines, and chemokines such as IL-4 and IFN-γ in a dose-dependent manner. Moreover, there was no systemic toxicity in single- or repeat-dose toxicity studies using ICR mice. In addition, we analyzed miRNA arrays from cASC-EVs using next-generation sequencing (NGS) to investigate the role of miRNAs in improving inflammatory responses. Collectively, our results suggest that cASC-EVs effectively attenuate AD by transporting anti-inflammatory miRNAs to atopic lesions alongside no toxicological findings, resulting in a promising cell-free therapeutic option for treating canine AD.

Keywords: adipose tissue; canine atopic dermatitis; extracellular vesicles; mesenchymal stem cells.

Figure 4

cASC-EVs alleviate dermatitis in Biostir-induced atopic dermatitis-like skin lesions. (A) Schedule of the in vivo animal experiments. (B) Photographs of the dorsal lesion area of mice on days 0 and 28. (C) Clinical evaluation using the score index of AD, each week for four weeks. (D) Ear thickness in mice with Biostir-induced AD-like skin lesions. (E) IgE levels in the total plasma of Biostir-induced AD mouse model. (F,G) IL-4 and IFN-gamma mRNA levels, measured by RT-qPCR (n = 3). Data are mean ± SD. ** p < 0.01 vs. group 2, * p < 0.05 vs. group 2. G1 = Negative control group, G2 = Biostir-induced AD group, G3 = Biostir-induced AD + cASC-EV 1.00 × 10⁹ treated group, G4 = Biostir-induced AD + cASC-EV 3.33 × 10⁹ treated group, G5 = Biostir-induced AD + cASC-EV 1.00 × 10¹⁰ treated group, G6 = Biostir-induced AD + prednisolone 10 mg/kg treated group.

Figure 1

Characterization of cASCs. (A–C) Expression of cell surface markers of cASCs isolated from adipose tissue preserved for 24 h, determined by flow cytometry. (D) Expression of multipotency markers (Sox2, Oct4, and Nanog) by RT-PCR in cASCs during continuous passages. GAPDH was used as an internal control. The cASCs were positive for all markers. CMT-U27 (canine mammary cancer cell line) is used for comparison.

Figure 2

Differentiation of canine adipose-derived mesenchymal stem cells (cASCs) into osteocyte, chondrocyte, and adipocyte lineages. (A) Images of alizarin red staining for osteogenic lineage. Images of Alcian blue staining for chondrogenic lineage. Images of Oil Red O staining for adipogenic lineage. (B) Reverse-transcript quantitative PCR (RT-qPCR) data on gene expression relating to osteogenic, chondrogenic, and adipogenic factors (BNP2 and RUNX2 for osteogenic, Sox9 and aggrecan for chondrogenic, alongside FAS and SREBP1 for adipogenesis, respectively; n = 3 for each lineage). Data are mean ± SD. * p < 0.05 vs. Control group, ** p < 0.001 vs. Control group.

Figure 3

Isolation and characterization of canine ASC-EVs. (A) The schematic summary of the EV separation methods. (B) Histogram of particle concentration and size distribution of canine ASC-EVs measured by nanoparticle tracking analysis (NTA) (n = 3). The red part of the histogram represents the range of deviation. (C–E) Particle concentration, protein concentration, and purity of canine ASC-EVs (n = 3). (F) Histograms showing canine ASC-EVs stained with anti-human CD81 antibody, which are shown in comparison to an IgG1 isotype-stained negative control. (G) Calnexin and concentration of canine ASC-EVs in each batch (n = 3).

Figure 5

Inhibitory effects of EVs on AD-like lesions in Biostir-AD induced atopic dermatitis-like phenotypes in a murine model. (A) Representative H&E (hematoxylin & eosin) and TB (toluidine blue) staining images of dorsal skin in normal mice and mice with Biostir-induced AD topically treated with EVs in a dose-dependent manner (1.00 × 10⁹, 3.33 × 10⁹, and 1.00 × 10¹⁰ particles/mL) and prednisolone (10 mg/kg; positive control). Representative immunohistochemical staining images revealed infiltration of TSLP and CD86 in dermatitis. Positively stained cells are shown in brown. (B) Graph indicating the number of infiltrated mast cells in the tissue, determined by toluidine blue staining. (C) Graph indicating the intensity of immune-positive cells of TSLP expression in dermatitis. (D) Graph indicating the intensity of immune-positive cells of CD86 expression in dermatitis. Data are mean ± SD. ** p < 0.01 vs. group 2. G1 = Negative control group, G2 = Biostir-induced ADgroup, G3 = Biostir-induced AD + cASC-EV 1.00 × 10⁹ treated group, G4 = Biostir-induced AD + cASC-EV 3.33 × 10⁹ treated group, G5 = Biostir-induced AD + cASC-EV 1.00 × 10¹⁰ treated group, G6 = Biostir-induced AD + prednisolone 10 mg/kg treated group.

Figure 6

miRNA expression profiles in different batches of cASC-derived EVs. (A) The heatmap describes the total miRNA profiles in EVs. (B) The ratio of different miRNAs in the top 20 miRNAs in EVs. (C–E) Gene Ontology (GO) analysis of the targets of cASC-EV miRNA. Enrichment of GO biological process, molecular function and pathway performed using DAVID Bioinformatics resources 6.8.