Introduction

Cirrhosis is a familiar end-stage of various chronic liver diseases, which is caused by diverse factors, such as viral infection, alcohol, drugs, and chemical toxicities (Goodman 2007; Huang et al. 2010; Kisseleva et al. 2010; Silveira et al. 2010). According to the epidemiological survey, there are 112 million cirrhosis patients worldwide, and the mortality rate can rise to 88% within 5 years (D’Amico et al. 2018, 2020). Cirrhosis is usually manifested as the accumulation of extracellular matrix (ECM) in liver tissues caused by the activation of hepatic stellate cells (HSCs), causing severe fibrosis of liver tissues, and then loss of original liver function (Shiels et al. 2017; Hu et al. 2020b). The most effective treatment for cirrhosis is liver transplantation, but there are problems such as scarcity of donor transplantation, immune rejection, complex surgery, and high cost (Lurie et al. 2015; Roehlen et al. 2020). Although hepatocyte transplantation has emerged as an alternative method to restore liver function and advance liver regeneration, this method is limited by the fact that hepatocytes are prone to deactivation in vitro (Piscaglia et al. 2010). Clearly, more effective and safer therapies are needed for the treatment of cirrhosis.

In recent years, massive experiments indicated that transplantation of mesenchymal stem cells (MSCs) is a safe and effective transitional therapy for end-stage liver disease before liver transplantation, which can advance liver repair and regeneration and notably ameliorate the liver function and quality of life of patients (Hu et al. 2019). Adipose mesenchymal stem cells (ADMSCs) are widely concerned in tissue reconstruction and cell transplantation applications due to their characteristics of easy collection, rich stem cells, strong proliferation, and anti-infection (Seki et al. 2013; Qu et al. 2017). However, the effect of stem cell transplantation alone in the therapy of cirrhosis is not ideal, so new methods to promote the higher performance of stem cells are in urgent need of development (Keating 2012). An increasing number of studies have used MSCs as the carrier of exogenous genes to transfect genes with clear biological functions into MSCs, so as to generate a role together to improve the effect of MSCs in reducing liver fibrosis (Seo et al. 2014; Zhu et al. 2021). Hepatocyte growth factor (HGF) has been proven to manifest a pivotal function in liver regeneration and protection, and the up-regulation of HGF expression can reduce liver fibrosis (Seo et al. 2014). Kim et al. pointed out that MSCs overexpressing HGF can improve liver function and alleviate liver fibrosis compared with MSCs alone (Kim et al. 2014).

Mounting evidence has strongly supported that MSCs improve the internal microenvironment of the liver and accelerate the regeneration of damaged liver tissue by releasing microbubbles (exosomes) (Terai and Tsuchiya 2017; Driscoll and Patel 2019; Zhao et al. 2019). MSC-derived exosomes (MSC-Exo) can reproduce the biological activity of MSCs and may be used as a substitute for whole-cell therapy (Lou et al. 2017). Compared with MSCs, MSC-Exo not only has a smaller nanometer diameter and does not block the pulmonary microvascular system during migration but also has higher safety and lower immunogenicity, and cannot directly form tumors (Börger et al. 2017). Therefore, the application of MSC-Exo has more advantages than MSCs. It has been reported that HGF-modified hADMSC exosomes can repair full-thickness skin defects in diabetic mice (Cao et al. 2022). However, it is unclear whether ADMSCHGF-Exo is also stronger than ADMSC-Exo in the treatment of cirrhosis.

It has been reported that the Rho-GTPase signal is a key pathway leading to liver injury, liver fibrosis, hepatocellular carcinoma, and cirrhosis (Hennenberg et al. 2009; Ren et al. 2021; Wang et al. 2022). Suppression of the Rho-GTPase signal can effectively alleviate liver injury (Choi et al. 2006). Wells et al. clarified that Rho family GTPases were activated during HGF-mediated prostate cancer-cell scattering (Wells et al. 2005). However, whether ADMSCHGF-Exo affects the progression of cirrhosis by regulating Rho-GTPase signaling remains unclear. Therefore, this study was designed to analyze the anti-fibrosis role of exosomes secreted by HGF-modified-mouse ADMSCs on activated HSCs and to preliminarily explore the possible mechanism, thereby providing new ideas and theoretical basis for the clinical treatment of cirrhosis.

Materials and methods

Animals

Five 3-week-old female BALB/c mice and thirty-six 6-8-week-old male BALB/c mice, specific pathogen-free (SPF), were acquired from Shanghai Sippe-Bk Lab Animal Co., Ltd. (China) with the certificate number SCXK (Hu) 2018-0008. All mice were reared at an ambient temperature of 22 ± 2 °C , a humidity of 55 ± 5%, and 12-h of circulating light. All procedures involving mice were authorized via the Ethics Committee of Zhejiang Eyong Pharmaceutical Research and Development Center (SYXK (Zhe) 2021-0033).

Isolation and identification of ADMSCs

Each mouse underwent euthanize by inhaling excess CO2. The subcutaneous adipose tissue of the medial inguinal region of the mouse was quickly separated and then digested in type I collagen (C1696, Bioswamp, China) at 37 °C for 50 min. After centrifugation, the pelleted cells were re-suspended in DMEM/F12 medium (ZQ-600, Zqxzbio, China) augmented with 10% exosome-depleted fetal bovine serum (FBS, 11,011 − 8615, Tianhang, China) and 1% penicillin-streptomycin liquid (P1400, Solarbio, China). After centrifugation again, the bottom cells were dripped into phosphate buffer saline (PBS, SH30256.01, Hyclone, USA) followed by centrifugation. The centrifuged cells (2 × 105/mL) were grown in DMEM/F12 medium with 10% exosome-depleted FBS under a cell incubator (BB150, Thermo Fisher Scientific, USA). The medium was changed initially after 24 h and then every 3 days. The morphology of ADMSCs with different generations was observed under an optical microscope (AE2000, Motic, China).

Phenotypic analysis of the third passage ADMSCs was carried out through the flow cytometer. The above cells were suspended in PBS and cell density was diluted to 1 × 106/100 µL. Thereafter, anti-CD45 antibody (560,695), anti-CD44 antibody (553,134), anti-CD34 antibody (551,387), and anti-CD105 antibody (562,759) were added (4 °C, 0.5 h, dark room). After washing, the above cells were assessed by applying a flow cytometer (C6, BD, USA).

Construction and transduction of lentiviral vector

Genechem (China) supplied a blank lentiviral vector encoding only green fluorescent protein (GFP) and a HGF-overexpressing lentiviral vector encoding GFP and HGF. The packaging and concentration of lentivirus was completed by Genscript Co., Ltd. (China). ADMSCs (1 × 105 cells/well) were appended at 24-pore plates 24 h before transfection. The next day, the cell culture medium was replaced with fresh medium containing 6 µg/mL polybrene (H8761, Solarbio, China) and viral suspension, which was then reacted at 37℃ for 24 h. Thereafter, the medium containing the virus was replaced with a fresh medium and incubated for 48 h. For efficiency of GFP expression, a fluorescence microscopy (Ts2-FC, Nikon, Japan) was applied. The stable transfected cells were screened using a fresh complete medium containing purinomycin.

Exosome preparation and identification

For the acquisition of exosomes, exosome isolation kits (41201ES25, Yeasen, China) was selected. The culture medium (including exosome-depleted FBS) of ADMSCs, ADMSCNC, and ADMSCHGF was harvested and centrifuged. The obtained supernatant (10 mL) was injected into a new centrifuge tube with exosome separation reagents (2.5 mL). After being reacted at 4 °C for 2 h, the mixture underwent centrifugation (10,000 g, 60 min). The collected precipitate was re-suspended in 100 µL PBS and centrifuged again at 12,000 g for 2 min. For purification of exosomes, the supernatant was passed through a filter membrane, about 13 µg exosomes were obtained. The morphology of exosomes was identified under a transmission electron microscope (TEM, H-600, Hitachi, Tokyo, Japan). The size distribution of ADMSC-Exo was quantified by nanoparticle tracking analysis (NTA). Furthermore, the representative markers (CD63, CD81, and CD9) of exosomes were identified through western blot.

Cell culture and experimental design

WHELAB (Shanghai, China) supplied mouse HSCs (C2211). HSCs were grown in DMEM medium (SH30243.01, Hyclone, USA) with 10% FBS, which were then put into the cell incubator (37 °C, 5% CO2).

To analyze the anti-fibrotic effect of exosomes secreted by ADMSCs overexpressing HGF on activated HSCs, HSCs were allocated to the control group (HSCs were cultured normally), TGF-β group (HSCs were subjected to 10 ng/mL TGF-β for 48 h) (Wang et al. 2018), TGF-β + ADMSC-Exo group, TGF-β + ADMSCNC-Exo group, and TGF-β + ADMSCHGF-Exo group. After 2 days of induction with TGF-β, the latter three groups of HSCs were treated with ADMSC-Exo, ADMSCNC-Exo, and ADMSCHGF-Exo at 400 µg/µL and 50 µL, respectively, for 48 h.

Cell vitality assay

MTT kit (E606334) supplied by BBI Life Science (UK) was selected to analyze cell viability of HSCs. HSCs were suspended and adjusted to 1 × 104 cells/mL, which were then loaded in the 96-pore plates for one day. Then, the cells were treated according to grouping. After 48 h, 10 µL MTT solution was dripped into each well. After being reacted for 4 h in the cell incubator, absorbance of these cells at 450 nm was tested by a CMaxPlus microplate reader (Molecular Devices, USA) to calculate the cell viability.

Apoptosis assay

Annexin V-FITC/PI Apoptosis Detection Kit (CW2574S, CWBIO, China) was taken to analyze the apoptosis of HSCs. The HSCs were cultured in 6-well plates at the density of 1.5 × 106 per well. After being treated as grouping, the cells were washed and the cell concentration was adjusted to 1 × 106/mL. After centrifugation, the precipitated HSCs with 100 µL binding buffer were stained with 5 µL Annexin V-FITC at 37 °C away from light for 10 min. Then, 5 µL PI was added (37 °C, 5 min). After adding a binding buffer, apoptosis of HSCs was monitored under the flow cytometer.

Cell cycle assay

A cell cycle kit (CCS012, Multi Sciences, China) was applied according to its manual. Firstly, the treated HSCs were washed, digested, and centrifuged (1000 rpm, 3 min). Subsequently, 1 mL pre-cooled PBS was taken to re-suspend the pelleted HSCs. The re-suspended HSCs underwent centrifugation. HSCs of each well were fixed with 1 mL pre-cooled 70% ethanol at 4 °C for 2 h prior to washing. After that, HSCs were subjected to 500 µL staining solution in the darkroom at 37 °C for 0.5 h. In the end, cellular distribution was measured by the flow cytometer.

Immunofluorescence

HSCs (3 × 104 per well) were inoculated into the 6-well plate with cover glass for one day. Next, HSCs received different treatments. The fixation of HSCs was conducted with 4% paraformaldehyde (P0099, Beyotime, China). Thereafter, 0.5% Triton X-100 (10,789,704,001, Roche, USA) was exploited to permeate the cell membrane. Then, 3% bovine serum albumin (BSA, 4240GR100, BioFRoxx, Germany) was utilized to seal HSCs for 30 min. A further all-night reaction was conducted with anti-α-SMA antibodies (1:200, Ab5831, abcam, UK) at 4 °C. After being reacted with goat anti-mouse IgG H&L (Alexa Fluor® 594, 1:500, ab150116, Abcam, UK), they were then subjected to DAPI (ab104139, Abcam, UK). Lastly, the cover glass was observed with a fluorescence microscope (Ts2-FC, Nikon, Japan) after adding mounting media.

Quantitative real-time polymerase chain reaction (QRT-PCR)

The total RNA from ADMSCs, ADMSC-Exo and HSCs were isolated by Trizol reagent (B511311, Sangon, China). After being converted into cDNA with the use of cDNA First Strand Synthesis Kit (CW2569, CWBIO, China), cDNA was amplified with the assistance of SYBR Green qPCR kit (CW2601, CWBIO, China) under a PCR system (LightCycler® 96, Roche, USA). β-actin was set as the normalizer and 2−ΔΔCt was performed for data evaluation. Primers were presented in Table 1.

Table 1 Quantitative real-time polymerase chain reaction primers
Full size table

Western blot

Following extraction of protein from ADMSCs and HSCs using RIPA lysis buffer (P0013D, Beyotime, China), the protein was quantified by BCA kit (pc0020, Solarbio, China). After denaturation, the denatured protein underwent electrophoresis, which was then electroblotted onto Polyvinylidene Fluoride (PVDF) membrane. Thereafter, 5% BSA was applied to shake the membrane at 37℃ for 2 h. Afterward, the membrane was probed with appropriate primary antibodies (4℃, all night). After another 60 min of treatment with Anti-Rabbit IgG H&L (HRP) antibody (1:5000, ab7090, Abcam, UK) or Rabbit Anti-Mouse IgG H&L (HRP) antibody (1:5000, ab6728, Abcam, UK), the immunoreactivity was generated using ECL reagent (GK10008, GlpBio, USA). The images were obtained by a chemiluminescence apparatus (ChemiScope 6000, Clinex, China). The primary antibodies of CD63 (1:1000, ab217345), CD81 (1:8000, ab109201), CD9 (1:1000, ab307085), α-SMA (1:1000), Collagen I (Col-I, 1:1000, ab260043), RhoA (1:5000, ab187027), Cdc42 (1:10000, ab187643), Rac1 (1:1000, ab155938), P27 (1:1000, ab193379), and β-actin (1:5000, ab8227) were bought from Abcam (UK).

Establishment of liver injury model in mice

Sixty mice were randomized to the control, CCl4, CCl4+ADMSC-Exo, CCl4+ADMSCNC-Exo, and CCl4+ADMSCHGF-Exo groups. For induction of cirrhosis, CCl4 dissolved in olive oil was applied in this study. Specifically, except for the control group, the mice in other four groups were injected subcutaneously with 40% CCl4-olive oil solution (5 mL/kg body weight) on the first day. Thereafter, the latter four groups of mice were injected subcutaneously with 40% CCl4-olive oil solution (3 mL/kg body weight) twice a week for 6 weeks. In addition, the latter four groups were fed with high-fat and low-protein feed and drunk with 0.5% alcohol. Mice in the control group were given a subcutaneous injection of the same volume of olive oil and fed with normal feed and normal drinking water. After successful modeling, the exosomes of ADMSCs, ADMSCNC, and ADMSCHGF (400 µg/µL, 200 µL) were injected into the tail vein of mice once a week for 4 weeks. At the same time, the mice in the control and CCl4 groups received PBS with the same amount. After 2 weeks and 4 weeks of administration (8 weeks and 10 weeks of modeling), the blood samples from the orbital vein of mice were harvested and centrifuged to obtain the serum.

Enzyme-linked immunosorbent assay (ELISA) assay

In this part, the mouse alanine aminotransferase (ALT) kit (ml063179), mouse aspartate aminotransferase (AST) kit (ml058577), and mouse serum albumin (ALB) kit (ml057991) were introduced from MEIMAIN (China). Total bilirubin (TBIL) kit (C019-1-1) was introduced from Jiancheng (China). The concentrations of ALT, AST, ALB, and TBIL of the serum in mice were determined as instructions.

Statistics

Data analysis obtained from triplicate independent experiments was conducted by SPSS software (16.0, IBM, USA). The measurement data were presented as mean ± standard deviation. One-way ANOVA with Turkey test was utilized for multiple comparisons. For those with unequal variances, the Kruskal-Wallis H test was applied. A statistically significant difference can be unveiled if p < 0.05.

Results

Isolation and identification of ADMSCs

As displayed in Fig. 1A, under an optical microscope, ADMSCs were observed to grow rapidly, most of which were fibrous or fusiform. After 24 h of culture, some ADMSCs began to adhere to the wall; after 72 h of culture, a large number of ADMSCs showed adherent growth. Next, the flow cytometer analysis was adopted to detect the surface markers of ADMSCs. The data presented in Fig. 1B found that CD44 and CD105 were positively expressed, while CD34 and CD45 were negatively expressed in ADMSCs, indicating the successful isolation of ADMSCs.

Fig. 1
figure 1

Morphological and phenotypic features of ADMSCs. A Spindle-like morphology of ADMSCs was observed under microscopy with a magnification of 100× and 200×. B Flow cytometer revealed the expressions of CD44, CD105, CD34, and CD45 in ADMSCs. Note: ADMSCs, adipose-derived mesenchymal stem cells.

Full size image

HGF overexpression transfection didn’t affect the properties of mesenchymal stem cells and Exo

There was no notable difference in the level of mouse HGF between ADMSCs and ADMSCNC but the level of mouse HGF in the ADMSCHGF group was higher than that in the ADMSCNC group, suggesting the transfection was performed successfully (Fig. 2A, p < 0.01). Then, the flow cytometer was applied to detect the surface markers of ADMSCs after transfection. The proportion of CD45 and CD34 positive cells in the ADMSCHGF group and ADMSCNC group was less than 1%, while the proportion of CD44 and CD105 positive cells was more than 90% (Fig. 2B).

Next, we isolated exosomes from ADMSCs, ADMSCNC, and ADMSCHGF. The exosomes were observed as oval vesicular structures by TEM (Fig. 2C). NTA showed that the particle size of exosomes was concentrated in the range of 50–100 nm (Fig. 2D). In addition, as illustrated in Fig. 2E and F, when compared with the ADMSC-Exo group, the expressions of HGF mRNA and protein in the ADMSCNC-Exo group did not change significantly. However, relative to the the ADMSCNC-Exo group, the expressions of HGF mRNA and protein the in ADMSCHGF-Exo group were significantly upregulated (p < 0.01). The further western blot assay exhibited that the expressions of exosome-related markers CD9, CD63, and CD81 in the ADMSC-Exo, ADMSCNC-Exo, and ADMSCHGF-Exo groups were not significantly changed (Fig. 2E).

(A) HGF overexpression in ADMSCs was verified by qRT-PCR after transfection. (B) Flow cytometer analysis for the expressions of CD45, CD44, CD105, and CD34 in ADMSCs. (C) ADMSC-Exo morphology was observed using TEM. Scale bar, 100 nm. (D) The size distribution of ADMSC-Exo was quantified by NTA. (E) The expression of HGF mRNA in ADMSC-Exo was detected by qRT-PCR. (F) The expression of HGF protein in ADMSC-Exo was detected by western blot. (G) The expression levels of CD63, CD81 as well as CD9 proteins in ADMSC-Exo were detected using western blot. @p < 0.05, @@p < 0.01 vs. ADMSCs.#p < 0.05, ##p < 0.01 vs. ADMSCNC-Exo. The results were presented as the mean ± standard deviation, n = 3. Note: Exo, exosome; HGF, hepatocyte growth factor; TEM, transmission electron microscopy; NTA, nanoparticle tracking analysis.

Fig. 2
figure 2

Overexpression of HGF did not affect the expressions of markers for ADMSCs and Exo

Full size image

HGF overexpression in ADMSC-Exo suppressed the activation of TGF-β-mediated HSCs

The MTT assay indicated that TGF-β induction prominently elevated cell viability of HSCs and the effect was partially offset with the addition of ADMSC-Exo/ADMSCNC-Exo/ADMSCHGF-Exo (Fig. 3A, p < 0.05). The flow cytometer assay proposed that apoptosis of HSCs was largely repressed after TGF-β treatment and the repressed capability was weakened by ADMSC-Exo/ADMSCNC-Exo/ADMSCHGF-Exo (Fig. 3B, p < 0.05). Importantly, the group that received ADMSCHGF-Exo facilitated apoptosis of TGF-β-induced HSCs compared to the cells that received ADMSCNC-Exo (Fig. 3B, p < 0.05). Additionally, the cell cycle assay indicated that induction with TGF-β statistically raised the cells in the S stage but dismissed the percentage of cells in the G0/G1 stage and the G2/M stage in HSCs (Fig. 3C, p < 0.01). After ADMSC-Exo/ADMSCNC-Exo/ADMSCHGF-Exo mediation, fewer cells were noted to be arrested at the S phase, while more cells were arrested at the G0/G1 phase and G2/M phase (Fig. 3C, p < 0.05).

Fig. 3
figure 3

ADMSC-derived exosomal HGF suppressed the activation of TGF-β-mediated HSCs.  A MTT assay of cell viability in HSCs after incubation with ADMSC-Exo. Flow cytometry assay for cell apoptosis B and cell cycle C in HSCs. & p  < 0.05, && p  < 0.01 vs. Control; +p  < 0.05, ++p  < 0.01 vs. TGF-β; *p  < 0.05 vs. TGF-β + ADMSC NC -Exo. The results were presented as the mean ± standard deviation, n  = 3. Note: HSCs, hepatic stellate cells; TGF-β, transforming growth factor-β

Full size image

ADMSCHGF-Exo suppressed the α-SMA expression in TGF-β-mediated HSCs

By conducting an immunofluorescence assay, we found that relative to the control group, the fluorescence intensity of α-SMA protein in the TGF-β, TGF-β + ADMSC-Exo, ATGF-β + DMSCNC-Exo, and TGF-β + ADMSCHGF-Exo groups was higher (Fig. 4). Interestingly, ADMSC-Exo/ADMSCNC-Exo/ADMSCHGF-Exo intervention weakened the fluorescence intensity of α-SMA protein in HSCs induced by TGF-β (Fig. 4).

Fig. 4
figure 4

ADMSC-derived exosomal HGF suppressed α-SMA expression in TGF-β-mediated HSCs

Full size image

The expression of α-SMA in HSCs was detected by immunofluorescence with a magnification of 200× (scale bar = 50 μm), n = 3. Note: α-SMA, α-smooth muscle actin.

ADMSCHGF-Exo restrained collagen fiber production in TGF-β-treated HSCs by rho pathway

As displayed in Fig. 5A and B, TGF-β treatment was found to enhance the gene and protein levels of α-SMA, Col-I, Rho A, Cdc42, and Rac1 but lessen the gene and protein level of P27 in HSCs (p < 0.01). Compared to the TGF-β group, the gene and protein levels of α-SMA, Col-I, Rho A, Cdc42, and Rac1 were lower but the gene and protein expression of P27 was higher in the TGF-β + ADMSC-Exo, TGF-β + ADMSCNC-Exo, and TGF-β + ADMSCHGF-Exo groups (Fig. 5A and B, p < 0.05). More importantly, the regulatory effect of ADMSCHGF-Exo treatment on the above-mentioned proteins was stronger than that of ADMSCNC-Exo (Fig. 5A and B, p < 0.05).

Fig. 5
figure 5

ADMSC-derived exosomal HGF restrained collagen fiber production in TGF-β-treated HSCs by Rho pathway  QRT-PCR A and western blot B analysis for the mRNA and protein levels of α-SMA, Col-I, Rho A, Cdc42, Rac1 as well as P27 in HSCs. &p < 0.05, &&p < 0.01 vs. Control; +p < 0.05, ++p < 0.01 vs. TGF-β; *p < 0.05, **p < 0.01 vs. TGF-β + ADMSCNC-Exo. The results were presented as the mean ± standard deviation, n = 3.

Full size image

ADMSCHGF-Exo improved serum liver indices in CCl4-induced mice

After 2 weeks of treatment, relative to the control group, the contents of ALT, AST, and TBIL in the CCl4 group were all increased but the content of ALB was decreased (Fig. 6, p < 0.05). After 4 weeks of treatment, except CCl4 + ADMSCHGF-Exo group, the contents of ALT, AST, and TBIL in the other groups were higher but the content of ALB was lower relative to the control group (Fig. 6, p < 0.05). Also, ADMSC-Exo/ADMSCNC-Exo/ADMSCHGF-Exo induction effectively decreased the contents of ALT, AST, and TBIL but enhanced the content of ALB in CCl4-induced mice (Fig. 6, p < 0.05). We also found that ADMSCHGF-Exo treatment had a stronger regulatory effect on serum liver index than ADMSCNC-Exo treatment (Fig. 6, p < 0.05).

Fig. 6
figure 6

ADMSC-derived exosomal HGF improved serum liver indices in CCl4-induced mice

Full size image

After 2 and 4 weeks of ADMSC treatment, the contents of serum ALT, AST, ALB as well as TBIL in CCl4-induced mice were detected by ELISA. &p < 0.05, &&p < 0.01 vs. Control; +p < 0.05, ++p < 0.01 vs. CCl4; *p < 0.05, **p < 0.01 vs. CCl4 + ADMSCNC-Exo. The results were presented as the mean ± standard deviation, n = 6. Note: ALT, alanine aminotransferase; AST, aspartate aminotransferase; ALB, albumin; TBIL, total bilirubin.

Discussion

The previous study noted that human MSCs have low immunogenicity, can promote the repair of liver injury and liver regeneration, apparently mitigate liver fibrosis, and ameliorate liver function and quality of life (Cao et al. 2020). The specific markers CD45, CD34, CD44, and CD105 on the surface of mouse ADMSCs were identified by flow cytometer. Consistent with what Deng et al. reported (Deng et al. 2014), ADMSCs were negative for the level of hematopoietic markers CD45 and CD34, but positive for CD105 and CD44, which verified that we successfully isolated mouse ADMSCs. MSCs are regarded as the best human cell type that can expand the production of exosomes (Hu et al. 2019). In terms of surface markers, MSC-Exo expresses not only common marker proteins such as CD81, CD63, and CD9 but also specific marker proteins derived from MSCs such as CD29, CD44, and CD73, etc., these markers can be used as reference indicators for the isolation and identification of exosomes (Yu et al. 2014). In this study, TEM, NTA, and western blot techniques were used to identify that the exosomes obtained from ADMSCs, ADMSCNC, and ADMSCHGF, and found these exosomes are extracellular vesicles with biological activity.

MSCs are reported to advance liver regeneration by modulation of immune cells (Hu et al. 2020). More and more research results indicated that MSC-Exo exhibits a more critical role in the repair of liver tissue damage than MSCs. For example, Rong et al. found that hBMMSC-Exo effectively alleviated liver fibrosis, including inhibition of activated HSCs, reduction of collagen deposition, enhancement of liver function, inhibition of inflammation, increase of liver cell regeneration, and repression of Wnt/β-catenin signaling by constructing CCl4-treated rat liver fibrosis model and injecting hBMMSC-Exo (Rong et al. 2019). Some scholars also confirmed that MSCs and MSC-Exo lessen serum transaminase level and liver necrosis, and have the potential to reduce liver injury and improve organ regeneration after liver ischemia-reperfusion injury (Anger et al. 2019). According to the report, MSCs can reduce liver fibrosis by inhibiting the proliferation of activated HSCs and promoting their apoptosis (Ma et al. 2022). Cell proliferation depends on the normal operation of the cell cycle, and MSCs can impede the activation and proliferation of HSCs by changing the cell cycle, thus alleviating liver fibrosis (Wang et al. 2017). Qin et al. found that BMSCs might modulate HSCs and cyclin D1 by the RhoA-P27 signaling, which led to the cell cycle G1/S phase transition, restrained the proliferation, and induced apoptosis of HSCs (Qin et al. 2012). This research was the first to check the cell proliferation and cell cycle of exosomes secreted by HGF-modified mouse ADMSCs on activated HSCs. After HSCs were treated with ADMSC-Exo/ADMSCNC-Exo/ADMSCHGF-Exo, fewer cells were noted to be arrested at the S phase, cell viability was obviously repressed, and cell cycle inhibitor P27 expression was evidently enhanced. Importantly, HGF-modified ADMSC-Exo was superior to ADMSC-Exo in regulating the proliferation and cell cycle of activated HSCs.

In order to explore the mechanism of HGF-modified ADMSC-Exo against hepatic fibrosis, the expressions of fibrosis proteins α-SMA and Col-I in HSCs were detected after treatment. In in vitro experiments, HGF-transfected hUCMSCs prominently weakened the expressions of Col-I, Col-III, and Smad2 in HSCs-T6 (Yin et al. 2021). In TGF-β-induced HSCs, it was observed that the activation ability of HSCs was strongly weakened after BMSCs intervention, and the expressions of α-SMA and Col-I were largely decreased (Qiao et al. 2018). Another study found that Rho GTPase signaling (RhoA, Rac1, and Cdc42) displays an important role in controlling the high expression of α-SMA in TGF-β-mediated MSCs (Ge et al. 2018). These researchers further demonstrated that RhoA, Rac1, and Cdc42 modulate α-SMA expression in MSCs through a RhoA/ROCK/pMLC2/F-actin/MRTFa pathway (Ge et al. 2018). Consistent with the report of Li et al. (Li et al. 2012), TGF-β was found to enhance the gene and protein levels of Rho A, Cdc42, and Rac1 in HSCs. Importantly, ADMSC-Exo, ADMSCNC-Exo, and ADMSCHGF-Exo effectively repressed the gene and protein levels of α-SMA, Col-I, Rho A, Cdc42, and Rac1 in TGF-β-treated HSCs, and ADMSC-Exo with HGF transfection had the better effect, revealing that ADMSCHGF-Exo repressed the activation of HSCs by restraining Rho pathway.

In order to further confirm the protective effect of HGF-modified ADMSC-Exo on liver injury, this study also conducted an in vivo experiment. CCl4-mediated liver damage in mice is widely applied as an animal model for inducing liver fibrosis because it is highly similar to the morphological and functional changes observed in human cirrhosis (Cao et al. 2017). Therefore, we used this model to study the hepato-protective effect of ADMSCHGF-Exo on mice with liver injury. In this study, serum ALT, AST, ALB, and TBIL levels were measured to evaluate CCl4-induced liver injury in rats. The serum liver enzymes (ALT, AST, ALB, and TBIL) are important indicators of liver synthesis function (He et al. 2021). Zhang et al. found that the concentrations of serum ALT, AST, and TBIL decreased apparently after hUC-MSCs intervention in a rat model of acute liver failure (ALF) (Zhang et al. 2018). This study indicated that exosomes secreted by ADMSCs overexpressing HGF effectively decreased the contents of ALT, AST, and TBIL but enhanced the content of ALB in CCl4-induced mice. The improvement in liver function proposed that ADMSCHGF-Exo accelerated liver recovery from the progression of cirrhosis.

To sum up, ADMSCHGF-Exo alleviated the activation of HSCs by weakening the expression of RhoA, Cdc42, and Rac1, the key molecules in the Rho pathway, thus reducing the production of collagen. The in vivo experiments verified that ADMSCHGF-Exo accelerated the recovery of liver from the progression of cirrhosis. In this study, we conducted a preliminary discussion from the cell model level and measured the serological indicators of liver function at the animal model level. In the future, we will further study the relevant mechanisms from the animal model level to provide a more theoretical basis for clinical transformation to finally solve the difficulties in the treatment of cirrhosis.