From Hair to Colon: Hair Follicle-Derived MSCs Alleviate Pyroptosis in DSS-Induced Ulcerative Colitis by Releasing Exosomes in a Paracrine Manner

Abstract

Ulcerative colitis (UC) has attracted intense attention due to its high recurrence rate and the difficulty of treatment. Pyroptosis has been suggested to be crucial in the development of UC. Although mesenchymal stem cells (MSCs) are broadly used for UC therapy, they have rarely been studied in the context of UC pyroptosis. Hair follicle-derived MSCs (HFMSCs) are especially understudied with regard to UC and pyroptosis. In this study, we aimed to discover the effects and potential mechanisms of HFMSCs in UC. We administered HFMSCs to dextran sulfate sodium- (DSS-) treated mice and found that the HFMSCs significantly inhibited pyroptosis to alleviate DSS-induced UC. A transwell system and GW4869, an exosome inhibitor, were used to prove the paracrine mechanism of HFMSCs. HFMSC supernatant reduced pyroptosis-related protein expression and promoted cell viability, but these effects were attenuated by GW4869, suggesting a role for HFMSC-released exosomes (Exos) in pyroptosis. Next, Exos were extracted and administered in vitro and in vivo to explore their roles in pyroptosis and UC. In addition, the biodistribution of Exos in mice was tracked using an imaging system and immunofluorescence. The results suggested that Exos not only improved DSS-induced pyroptosis and UC but also were internalized into the injured colon. Furthermore, the therapeutic efficacy of Exos was dose dependent. Among the Exo treatments, administration of 400 μg of Exos per mouse twice a week exhibited the highest efficacy. The differentially expressed miRNAs (DEmiRNAs) between MSCs and MSC-released Exos suggested that Exos might inhibit pyroptosis through tumour necrosis factor-related apoptosis-inducing ligand (TRAIL) signalling and interferon- (IFN-) gamma pathways. Our study reveals that HFMSCs can alleviate pyroptosis in UC by releasing DEmiRNA-containing Exos in a paracrine manner. This finding may lead to new treatments for UC.

Figure 1

Investigation of the effects of HFMSCs on DSS-induced UC. (a) The morphology of HFMSCs was observed under a white-light microscope. (b, c) HFMSCs stained with Alizarin red and Oil red O differentiated into osteoblasts and adipocytes. (d) CK15 was observed by immunofluorescence. (e) Surface markers of HFMSCs, including positive markers (CD90 and CD29) and negative markers (CD31 and CD43), were detected by flow cytometry. HFMSCs were administered to DSS-treated mice. (f, g) The colon lengths of the mice in all the groups were measured and analysed. (h, i) The body weight loss and DAI values were compared among the three groups. (j) The colon histology of the mice in the three groups was detected by HE staining. (k) Mucosal damage and inflammatory infiltration in the colons were determined. All the data are displayed as the means ± SDs. ^∗P < 0.05, ^∗∗P < 0.01, and ^∗∗∗P < 0.001. ns: not significant. The images were obtained with 10x and 20x Olympus dotSlide objectives.

Figure 2

HFMSCs reduced pyroptosis in vivo. To detect the impact of HFMSCs on pyroptosis in DSS-treated mice, (a) the expression of the NLRP3, GSDMD, and PCNA proteins was detected by immunohistochemistry. (b–d) The efficacy of HFMSCs in the model mice was assessed semiquantitatively. (e) Western blotting was carried out to examine the protein expression of NLRP3, GSDMD, cleaved caspase-1, and IL-1β. In addition, (f, g) ELISA was applied to test the protein levels of IL-1β and IL-18 in the serum. All the data are displayed as the means ± SDs. ^∗P < 0.05, ^∗∗P < 0.01, and ^∗∗∗P < 0.001. ns: not significant. The images were obtained with an Olympus dotSlide objective. Scale bar: 100 μm.

Figure 3

HFMSCs exerted a protective effect against pyroptosis in a paracrine manner. (a) After PKH67 staining, HFMSCs were cocultured with MODE-K cells in a transwell system. (b) The fluorescence of PKH67 was then observed by fluorescence microscopy to detect the uptake of HFMSCs by MODE-K cells in a paracrine manner. Scale bar, 50 μm. MODE-K cells were divided into six groups for different treatments. (c–e) The cell viability in each group was determined through EdU and CCK-8 assays. Scale bar, 200 μm. (f) The results of western blotting revealed the levels of pyroptosis-related proteins in all the groups. (g, h) ELISA was used to analyse the IL-1β and IL-18 protein levels in the supernatant of each group. All the data are displayed as the means ± SDs. ^∗P < 0.05, ^∗∗P < 0.01, and ^∗∗∗P < 0.001. ns: not significant.

Figure 4

Exos exerted a therapeutic effect in vivo. Because Exos inhibited pyroptosis in vitro, we hypothesized that Exos could also be effective in vivo. C57BL/6J mice administered with DSS were treated with different doses of Exos (Exo¹: 100 μg, Exo²: 200 μg, and Exo³: 400 μg; twice a week for each mouse). (a) The colons of the mice were extracted, and (b) the colon lengths were compared. (c, d) The body weights and DAI values of the mice in each group were recorded and compared. (e) HE staining was performed to assess colon histology in each group. (f) After comparing the images of HE staining among the five groups, the histology scores with intestinal mucosa ulceration and inflammation were analysed. All the data are displayed as the means ± SDs. ^∗P < 0.05, ^∗∗P < 0.01, and ^∗∗∗P < 0.001. ns: not significant. All images were obtained with an Olympus dotSlide objective. Scale bars: 200 μm, 100 μm.

Figure 5

Exos relieved pyroptosis in vivo. (a–d) Immunohistochemistry and semiquantitative analysis revealed the protein levels of NLRP3, GSDMD, and PCNA in each group. (e–g) Western blotting and ELISA were applied to test the levels of pyroptosis-related proteins in the five groups. All the data are presented as the means ± SDs. ^∗P < 0.05, ^∗∗P < 0.01, and ^∗∗∗P < 0.001. ns: not significant. The images were obtained with an Olympus dotSlide objective. Scale bar: 100 μm.

Figure 6

Distribution of Exos in vivo. To ensure the internalization of Exos in vivo, (a) DiR-labelled Exos were administered to healthy mice and DSS-treated mice. (b) Twenty-four hours after DSS administration, the fluorescence intensity was compared between the two groups. (c) The fluorescence accumulation in the heart, lungs, liver, kidneys, spleen, and intestine was detected. Moreover, (d–o) immunofluorescence revealed colocalization of Exos and the NLRP3, GSDMD, and PCNA proteins. The fluorescence images were obtained with an X spectral imaging instrument and in vivo imaging software (NightOWL II LB983). Immunofluorescence images were captured by 40x fluorescence microscopy (Zeiss-DMI8). Scale bar: 200 μm.

Figure 7

Differential expression and bioinformatics analysis of miRNAs in Exos derived from MSCs. (a) The DEmiRNAs between the EXO group and the control group are shown in a volcano plot, and (b) the differential expression of the top 10 upregulated DEmiRNAs and the top 10 downregulated DEmiRNAs is shown in a heat map. Functional enrichment analysis was performed to examine the (c) CC terms, (d) MF terms, (e) BP terms, and (f) biological pathways for the DEmiRNAs. (g) DEmiRNA-mRNA regulatory network with the top 5 upregulated DEmiRNAs.