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

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.


Introduction
Ulcerative colitis (UC), an immune-associated inflammatory disease, has attracted attention globally due to its increasing incidence and recurrence rates [1] and poses great threats to health care and the economy worldwide [2]. Investigating the pathogenesis of UC and identifying more effective treatments are popular research topics. Although the pathogene-sis of UC remains underexplored, this condition is believed to be caused by an abnormal mucosal immune response and excessive inflammation in response to bacterial antigens [3]. The currently available medical treatments for UC, which include 5-aminosalicylic acid (5-ASA), hormones, immunosuppressants, and biological agents, have resulted in some improvements [4,5]; however, due to their limited efficacy and complications, some patients still experience relapse and must consider surgical treatment [6]. Thus, more advanced treatments for UC are urgently needed.
Due to the anti-inflammatory and immunomodulatory abilities of mesenchymal stem cells (MSCs) [7,8], transplantation of these cells has shown encouraging efficacy in UC [9][10][11]. Compared with MSCs from other sources, hair follicle-derived MSCs (HFMSCs) are more abundant, do not have age limitations, and are easier to obtain in a minimally invasive manner from patients [12]. HFMSCs also have much lower immunogenicity than other MSCs and do not have associated ethical issues [13]. In addition, a previous study has indicated that HFMSCs have greater proliferation ability than bone marrow MSCs [14]. Given these characteristics and the multidirectional differentiation potential of HFMSCs [13], these cells may have improved therapeutic prospects for UC.
MSCs have been demonstrated to exhibit homing, differentiation, and paracrine signalling abilities to exert antiinflammatory and immunomodulatory effects [15]. However, recent studies have suggested that MSCs exert their effects through paracrine pathways rather than homing and differentiation pathways [15,16]. MSC-released exosomes (Exos), which are particularly important paracrine components of MSCs, are significantly involved in intercellular signal communication [17]. Various proteins, mRNAs, miRNAs, and other molecules in MSC-released Exos are believed to influence the biological processes of target cells [18]. In particular, miRNAs in MSC-released Exos are strongly recommended for the treatment of UC [15,19,20]. Fully exploring the mechanisms of Exos may promote understanding of the functions of MSCs. Furthermore, given their substantial genetic material content, nonimmunogenicity, small size, and high transport efficiency [18], Exos may be promising treatment agents for UC.
Recent studies have shown that the nucleotide oligomerization domain-(NOD-) like receptor pyrin domaincontaining protein 3 (NLRP3) inflammasome has vital impact on the immune and inflammatory responses of the intestinal mucosa in UC [21,22]. The NLRP3 inflammasome is also the initial factor in pyroptosis that exacerbates UC [23]. Activated NLRP3 recruits the apoptosisassociated speck-like protein containing a caspase recruitment domain (ASC) protein, and the caspase-1 protein assembles into the inflammasome to cleave the caspase-1 protein and produce large amounts of interleukin-1β (IL-1β) and interleukin-18 . Then, these proinflammatory cytokines are released through gasdermin D (GSDMD) to initiate pyroptosis in UC [24,25]. A recent study has demonstrated that effectively suppressing NLRP3-induced pyroptosis can improve experimental colitis [26]. In addition, Cai et al. [27] reported that the NLRP3 inflammasome and pyroptosis pathway in UC can be blocked by MSCderived Exos containing the miRNA 378a-5p. These findings indicate that NLRP3 and the pyroptosis pathway may be the targets of MSCs for UC treatment.
In this study, we applied HFMSCs in UC to evaluate the therapeutic effects and explore the potential mechanisms of HFMSCs. The results may lay a theoretical foundation for the application of HFMSCs in UC treatment. 2.2. Isolation of HFMSCs. HFMSCs were isolated as previously described [28,29]. The skin and hair follicles of healthy mice were collected for extraction of HFMSCs. After several rounds of disinfection, the tissues were cut into 3 * 3 mm 2 blocks and incubated with type I collagenase (0.1%, Sigma-Aldrich, USA) at 37°C for 1-2 h. Under a stereomicroscope, hair follicles were then extracted and placed into type I collagenase for 3-4 h and trypsin for 1 h. Foetal bovine serum (FBS, ScienCell, USA) was used for the neutralization of trypsin. HFMSCs were obtained from the neutralizing solution after being centrifuged at 1000 rpm for 10 min. The HFMSCs were plated in Dulbecco's modified Eagle's medium/F12 (DMEM/F12, Gibco, USA) containing 15% FBS and 1% penicillin-streptomycin (Gibco). After 10-14 days, the cells reached 70-80% confluence. The cells were then expanded to passages 2-5 for subsequent experiments.

Establishment and Evaluation of Dextran Sulfate
Sodium-(DSS-) Induced UC. To establish a UC animal model as previously described [33], mice were given 2.5% DSS (MW = 36, 000 − 50,000 Da; MP Biomedicals, Canada) dissolved in drinking water and allowed to drink freely for 7 days and then to recover for 3 days. All mice were randomly divided into 4 groups: the control group (n = 12), the DSS+PBS group (n = 12), the DSS+HFMSC group (n = 6), and the DSS+Exo group (n = 18). Each mouse in the HFMSC treatment group was treated with 3 × 10 6 HFMSCs via the tail vein on the 3 rd day. The mice in the DSS+Exo group were administered Exos via the tail vein on the 3 rd and 5 th days. The DSS+Exo group was divided into three subgroups for administration of different doses of Exos; 100 μg, 200 μg, and 400 μg of Exos per mouse were injected twice a week into the mice in the DSS+Exo 1 , DSS +Exo 2 , and DSS+Exo 3 groups, respectively.
To assess the severity of UC [34], body weight loss, diarrhoea, and bloody stool were recorded daily. The disease activity index (DAI) was evaluated according to previously described methods [34]. On the 11 th day, the mice were sacrificed. The colon length of each mouse (from the rectum 2.6. HE Staining. Mouse colons were fixed with 4% paraformaldehyde, dehydrated with alcohol, cleared with xylene, embedded in paraffin, and cut into 5 μm sections. HE staining was carried out as described previously [15].

Immunohistochemistry and Immunofluorescence
Staining. Colon tissues were sectioned into 4 μm sections. Immunohistochemistry was performed, and the results were analysed as described above [36]. The immunohistochemistry images were obtained with an Olympus (BX41) microscope and semiquantitatively analysed with Fiji software.
Exos labelled with PKH67 were administered to the mice in the DSS+Exo 3 group, and frozen colon sections from the DSS+Exo 3 group were subjected to immunofluorescence staining. Immunofluorescence staining was carried out as described previously [36]. Immunofluorescence images were then collected with a fluorescence microscope (Zeiss-DMI8).
All primary antibodies are presented in Table S1.

Western
Blotting. Colons and cell samples were lysed for extraction of proteins, and the BCA method was applied to measure the protein concentration. Western blotting was implemented as previously described [15]. Images of the protein bands were then collected using an ImageQuant LAS 334 4000 mini machine (GE). The primary antibodies are shown in Table S1.
2.9. Enzyme-Linked Immunosorbent Assay (ELISA). Mouse venous blood and the supernatant of MODE-K cells were collected and centrifuged. ELISA was used to examine the levels of IL-1β and IL-18 in the samples. All experimental procedures were performed according to the protocols provided with the ELISA kits (Boster, China).

Cell Viability Analysis.
Cell viability was assessed with a Cell Counting Kit-8 (CCK-8) (APExBIO-K1018) and 5ethynyl-2 ′ -deoxyuridine (EdU) imaging kits (UE, China). A total of 2 × 10 3 MODE-K cells per well were cultured in 96-well plates and treated as described above. The CCK-8 and EdU assays were performed according to the manufacturer's recommended procedures.
2.11. Tracking of Labelled Exos In Vivo. Exos were labelled with 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindotricarbocyanine iodide (DiR; Thermo Fisher Scientific/Invitrogen, USA), and DiR-labelled Exos (Exos DiR , 400 μg per mouse) were administered to healthy mice and DSS-treated mice. Images of the mice and tissues were obtained 24 h after the administration of Exo DiR using an X spectral imaging instrument and in vivo imaging software (NightOWL II LB983).

Differentially Expressed miRNA (DEmiRNA) Analysis and Functional Enrichment Analysis.
From the Gene Expression Omnibus (GEO) database, we obtained the dataset GSE71241 (https://www.ncbi.nlm.nih.gov/geo/query/acc .cgi?acc=GSE71241) of miRNAs in human MSC-released Exos. Exos derived from MSCs were used for the experimental group (EXO, n = 9). MSC samples were used for the control group (Control, n = 9). DEmiRNAs were identified from the EXO and Control groups using the limma package with the criteria of an adjusted P value < 0.05 and a jlog 2FCj value > 1. Functional enrichment of the DEmiRNAs was performed via Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) analyses using FUN-RICH software to recognize the cellular components (CCs), molecular functions (MFs), biological processes (BPs), and related biological pathways involved. The upregulated DEmiRNAs were sorted by their |log2FC| values, and the top 5 miRNAs were used to predict the target genes using the miRTarBase database.

Statistical Analysis.
All experiments were carried out three times. The data are shown as the means ± SDs and were analysed by one-way analysis of variance (ANOVA) with GraphPad Prism 8.0. Values of P < 0:05 were considered to indicate significance.

HFMSCs Attenuated DSS-Induced UC.
HFMSCs were isolated from mouse hair follicles as described above [28,29]. Under a white light microscope, HFMSCs were observed as adherent cells exhibiting spindle-like shapes (Figure 1  Oxidative Medicine and Cellular Longevity 98.5%, CD90: 98.4%) and low expression of negative surface markers of MSCs (CD31: 1.71%, CD43: 1.74%) (Figure 1(e)). Thus, HFMSCs were confirmed to be MSCs derived from hair follicles. On the 3 rd day, the mice began to develop diarrhoea, weight loss, and bloody stool, and these symptoms persisted until the end of the DSS intervention. The morphology and length of the mouse colons in the three groups were assessed and compared on the 11 th day (Figure 1(f)). The colon length in the HFMSC treatment group was greater than that in the model group but less than that in the control group (Figure 1(g)). Based on the symptoms recorded daily for 10 days, the model group treated with DSS+PBS presented a significantly decreased body weight and elevated DAI values. However, these changes were markedly improved in the HFMSC treatment group (Figures 1(h) and 1(i)). Furthermore, as demonstrated by the histological analysis of the colon, the DSS+HFMSC group showed obviously decreased mucosal damage and clearly lower inflammatory infiltration than the DSS+PBS group (Figures 1(j) and 1(k)). Together, these results confirmed that HFMSCs substantially alleviated DSS-induced UC.

HFMSCs Reduced Pyroptosis to Relieve DSS-Induced UC.
NLRP3 inflammasome-induced pyroptosis has been proven to play a key role in UC [21,22]. To identify the effect of HFMSCs on pyroptosis, the protein expression of NLRP3, GSDMD, and proliferating cell nuclear antigen (PCNA) was detected by immunohistochemistry (Figure 2(a)). According to the semiquantitative analysis, the DSS+PBS group showed the highest number of positively stained cells with the NLRP3 and GSDMD proteins, while the staining for these proteins was distinctly reduced in the HFMSC treatment group (Figures 2(b) and 2(c)). In addition, the DSS+HFMSC group presented the highest numbers of PCNA-stained cells among the three groups ( Figure 2(d)).
To further prove the ability of HFMSCs to inhibit pyroptosis, western blotting and ELISA were used to detect the levels of pyroptosis-related proteins. Lower protein levels of NLRP3, GSDMD, cleaved caspase-1, and IL-1β were observed in the HFMSC treatment group than in the model group, as shown in Figure 2(e). Statistical analysis of these protein levels is exhibited in Figure S1. ELISA revealed that HFMSCs markedly reduced the protein levels of IL-1β and IL-18 (Figures 2(f) and 2(g)). Thus, we concluded that HFMSCs could distinctly suppress pyroptosis in UC.

HFMSCs Inhibited Pyroptosis In Vitro in a Paracrine
Manner. Recent studies have confirmed that the paracrine mechanism of MSCs is crucial and effective in many diseases [15,37,38]. To discover the role of the paracrine pathway in the effects of HFMSCs, a transwell system was implemented. In this system, MODE-K cells were plated in the lower chamber and cocultured with PKH67-labelled HFMSCs in the upper chamber (Figure 3(a)). After 24 h, marked green fluorescence in MODE-K cells was detected using fluorescence microscopy, which indicated the paracrine uptake of HFMSC components by MODE-K cells (Figure 3  Oxidative Medicine and Cellular Longevity 3(e), the supernatant of HFMSCs markedly increased the viability of MODE-K cells; the next-greatest viability was observed in the LPS+ATP+HFMSC+GW4869 group. In addition, no significant differences were found among the LPS+ATP group, the DMSO treatment group, and the GW4869 treatment group. Although the supernatant of HFMSCs effectively protected cells from pyroptosis, this protection was attenuated by GW4869, which implies that Exos are crucial for the paracrine pathway of HFMSCs. Furthermore, western blotting and ELISA were used to test the levels of pyroptosis-related proteins in all the groups. Compared with those in the model group, the levels of the proteins NLRP3, GSDMD, cleaved caspase-1, IL-1β, and IL-18 were prominently reduced in the LPS+ATP +HFMSC group and decreased to a lesser extent in the LPS+ATP+HFMSC+GW4869 group (Figures 3(f)-3(h) and Figure S2). There were still no significant differences among the LPS+ATP group, the DMSO treatment group, and the GW4869 treatment group. The abovementioned findings demonstrated that HFMSCs attenuated pyroptosis in a paracrine manner by releasing Exos.

HFMSC-Released Exos Attenuated Pyroptosis In Vitro.
We extracted Exos by differential centrifugation. The typical cup-and sphere-shaped morphologies were observed by TEM ( Figure S3A). NTA results demonstrated that the Exo particle size was approximately 90 nm ( Figure S3B). The surface proteins CD9 and TSG101 were enriched in Exos compared with HFMSCs, whereas the protein calnexin was rarely expressed in Exos, as indicated by western blotting ( Figure S3C). Considering the uptake of HFMSCs by MODE-K cells in the transwell system, we detected the internalization of PKH67-labelled Exos by MODE-K cells using immunofluorescence. Green fluorescence appeared in MODE-K cells after incubation with Exos ( Figure S3D), which implied that the HFMSC components were internalized by MODE-K cells via released Exos. To further verify the effect of the supernatant of HFMSCs on 10 Oxidative Medicine and Cellular Longevity pyroptosis through Exos, Exos were applied to inhibit pyroptosis in vitro. As evidenced by EdU and CCK-8 assays, Exos markedly promoted the viability of MODE-K cells compared with that in the LPS+ATP group ( Figures S3E-S3G). Western blotting demonstrated that Exos significantly inhibited pyroptosis by decreasing the protein levels of NLRP3, GSDMD, cleaved caspase-1, and IL-1β ( Figure S3H). Moreover, the expression of the proteins IL-1β and IL-18 in the supernatant of MODE-K cells treated with Exos was lower than that in the LPS +ATP group, as shown by ELISA ( Figures S3I and S3J). In general, these results confirmed that Exos could effectively limit pyroptosis in vitro and were key components of HFMSCs.

Exos Exerted a Therapeutic Effect In Vivo.
Based on the therapeutic effect of Exos in vitro, we hypothesized that Exos could also be effective in vivo. C57BL/6J mice treated with DSS for 7 days were administered different doses of Exos (Exo 1 : 100 μg, Exo 2 : 200 μg, and Exo 3 : 400 μg for each mouse) on the 3 rd and 5 th days. The colons of the mice in the five groups were measured and analysed on the 11 th day. In Figures 4(a) and 4(b), the colon lengths in the Exo 2 and Exo 3 treatment groups were clearly greater than those in the model group and the DSS+Exo 1 group, whereas there was no significant difference in colon length between the Exo 2 and Exo 3 treatment groups. A significant difference was also not observed between the model group and the DSS+Exo 1 group. After DSS administration for 3 days, the body weight of the mice, particularly in the model group, distinctly decreased over time. However, as shown in Figure 4(c), the Exo treatment groups exhibited less weight loss with increasing doses of Exos, although the data for the Exo 3 and Exo 2 treatment groups did not differ. As shown in Figure 4(d), the DAI scores in the Exo 3 treatment group were significantly lower than those in the model group, followed by those in the DSS+Exo 2 group. However, the DAI scores in the Exo 1 treatment group did not significantly differ from those in the model group. Intestinal mucosal ulceration and inflammation gradually improved with increasing doses of Exos (Figures 4(e) and 4(f)). Based on the abovementioned results, we concluded that Exos relieved DSS-induced UC in a dose-dependent manner. Furthermore, we found that twice-weekly injection of 400 μg of Exos was the most appropriate dose for alleviation of DSSinduced UC.

Exos Protected DSS-Treated Mice from Pyroptosis.
We performed immunohistochemical staining of the colons from each group. As shown in Figures 5(a)-5(c), decreasing numbers of NLRP3-and GSDMD-positive cells were detected with increasing doses of Exo, and the lowest numbers were found in the DSS+Exo 2 group and the DSS+Exo 3 group. Additionally, the analysis of NLRP3-positive cells showed no clear distinction between the model group and the Exo 1 treatment group. In contrast, increased numbers of PCNA-positive cells were detected in the Exo treatment groups ( Figure 5(d)). Interestingly, the numbers of PCNApositive cells did not significantly differ among the Exo treatment groups. Western blotting revealed that the NLRP3, GSDMD, cleaved caspase-1, and IL-1β protein levels decreased along with increasing Exo doses ( Figure 5(e)).

12
Oxidative Medicine and Cellular Longevity Statistical analysis of these protein levels was shown in Figure S4. The ELISA also showed that the protein expression of IL-1β and IL-18 was negatively correlated with the dose of Exos (Figures 5(f) and 5(g)). Accordingly, the above results indicated that Exos prominently blocked pyroptosis in a dose-dependent manner and promoted regeneration to some extent.

Exos Entered the Damaged Colon and Influenced
Pyroptosis. Based on the effects of Exos in reducing pyroptosis, we attempted to determine the biological distribution of Exos. Previous studies have demonstrated that Exos infused into mice with colitis can reach the damaged colon and improve colitis [39,40]. Accordingly, considering the uptake of Exos in vitro, DiR-labelled Exos were injected into healthy mice and DSS-treated mice, and the internalization of the Exos was examined in vivo ( Figure 6(a)). After 24 h, the DiR dye was highly concentrated in the livers and spleens of mice in both groups. In addition, the colons of the mice in the DSS+Exo DiR group presented clear DiR fluorescence (Figures 6(b) and 6(c)), which suggested that the Exos were internalized into the damaged colon, in line with the findings of previous studies [39,40]. Moreover, considering the ability of Exos to reduce pyroptosis, we tracked the distributions of PKH67-labelled Exos in the DSS+Exo 3 group by immunofluorescence. As shown in Figures 6(d)-6(o), Exos appeared at the same site as the NLRP3, GSDMD, and PCNA proteins, which further implied that Exos might target the injured colon to improve pyroptosis and promote regeneration in vivo.

Discussion
In this study, we proved that HFMSCs could effectively relieve DSS-induced UC and pyroptosis by releasing Exos. Furthermore, our bioinformatics results indicated that the TRAIL and IFN-gamma signalling pathways may be involved in the effects of Exo treatment on pyroptosis. Our findings reveal the efficacy and mechanism of HFMSCs against UC, potentially providing a promising treatment for UC.
Hair follicles, as natural reservoirs of MSCs, are recognized as having advantages over other sources of MSCs [13]. In previous studies, HFMSCs have been suggested to be effective in the treatment of hair loss [30], pancreatitis [44], and liver cirrhosis [45] due to their anti-inflammatory properties and their ability to differentiate into parenchymal cells. However, there have been few studies on HFMSCs in the context of UC. In this study, HFMSCs were revealed to diminish colon shortening, body weight loss, bleeding, and colon injury, in line with the effects of MSCs derived from other sources on UC [9][10][11]. It has been shown that dental pulp MSCs transfected with hepatocyte growth factor can transdifferentiate into intestinal stem cells to reduce inflammation and restore mucosal integrity in UC [9]. However,  14 Oxidative Medicine and Cellular Longevity   Sala et al. [10] reported that the anti-inflammatory ability of MSCs in UC is more dependent on paracrine release of tumour necrosis factor-induced protein 6 than on homing to the damaged colon. Recently, the paracrine function of MSCs was proven to be closely related to the release of Exos [46,47]. After GW4869 treatment, the effects of MSCs have been found to prominently decrease, consistent with our findings [47]. Studies on the application of Exos have further proven the key role of Exos in MSCs. Exos derived from human umbilical cord MSCs are believed to release miRNA 378a-5p to macrophages, potentially targeting NLRP3 to attenuate colitis [27]. Adipose MSC-released Exos have been suggested to regulate Foxp3+ Treg cells in the spleen and lymph nodes to reduce the inflammatory cytokine release induced by DSS [48]. As shown in our findings, NLRP3induced pyroptosis was significantly inhibited by Exos, improving DSS-induced UC. Collectively, the evidence suggests that MSC-released Exos can modulate immune responses and control inflammatory responses to attenuate UC [49].
NLRP3 has been suggested to be the key protein leading to severe inflammation and the pyroptosis pathway in UC [21,22]. A previous study has demonstrated that NLRP3 −/ − mice treated with DSS are not susceptible to UC and have lower levels of IL1β and IL18 than wild-type mice [50]. Pterostilbene derivatives have been revealed to improve   16 Oxidative Medicine and Cellular Longevity experimental colitis by suppressing NLRP3-induced pyroptosis [26]. Another study has suggested that MSC-derived exosomal miR-378a-5p can target the mRNA of NLRP3 to inhibit pyroptosis and attenuate UC [27]. Gu et al. [51] reported that MSC-derived exosomal miR-181a limits the expression of proinflammatory factors (TNF-α, IL-6, IL-1β, IL-17, and IL-18) and improves epithelial integrity in UC. NLRP3 and NLRP3-induced pyroptosis may become therapeutic targets for UC. In addition, PCNA was found to be highly expressed in the HFMSC and Exo treatment groups, which implied that some potential functions of HFMSCs and Exos need to be explored in the future.
Emerging studies have demonstrated that miRNAs are highly enriched in Exos and perform multiple biological functions [20,38]. We did not detect the levels of miRNAs in Exos but used bioinformatics tools to analyse the DEmiR-NAs and their enriched functions. The TRAIL signalling pathway and IFN-gamma pathway were revealed to be enriched for the DEmiRNAs. Interestingly, these pathways have been demonstrated to positively regulate the NLRP3 inflammasome through the protein MLKL [42]. The protein TRAIL has also been revealed to interact with DR5 and NFκB to stimulate the NLRP3 pathway [52,53]. Another study has proven that IFN-gamma can inhibit mitochondrial integrity in Paneth cells by inducing mTORC1-dependent pyroptosis and stimulating intestinal inflammation [41]. Together, the evidence suggests that DEmiRNAs in MSCreleased Exos may reduce pyroptosis by targeting the TRAIL signalling and IFN-gamma pathways, which should be further verified in our future research.
There is no clear standard for the dosage of MSCreleased Exos for UC treatment. It has been reported that 200 μg of bone marrow MSC-released Exos can protect mice that drink 5% DSS for 7 days from developing severe colitis [54]. Another study has demonstrated that three injections of 400 μg/mouse human umbilical cord MSC-released Exos can downregulate IL-1β protein levels in mice with UC induced by treatment with 3% DSS for 11 days [39]. Treatment of mice with two injections of 60 μg of olfactory/ ecto-MSC-released Exos can alleviate 2.5% DSS-induced acute UC [19]. We hypothesize that the different effective frequencies and doses of MSC-released Exos may be attributable to the different sources of MSCs and the different DSS interventions. In our study, three Exo treatment groups (Exo 1 : 100 μg, Exo 2 : 200 μg, and Exo 3 : 400 μg; twice a week in each mouse) were used to determine the appropriate dose of Exos. Based on the excellent amelioration of UC and pyroptosis, twice-weekly injection of 400 μg of Exos was concluded to be the appropriate dosage of Exos for acute UC. This finding may provide a new dosage reference for the application of Exos in acute UC.

Conclusions
Our findings confirm that HFMSCs exert therapeutic effects against DSS-induced UC and pyroptosis by releasing Exos in a paracrine mechanism. The evidence obtained with HFMSCs, as novel advantageous MSCs, may provide new insights for research on MSCs in UC. Based on the bioinfor-matics results, pyroptosis is supposed to be a potential target of Exos for the treatment of UC.