Abstract

Background

   Exosomes derived from pre-stimulated mesenchymal stem cells (MSCs) have
   improved therapeutic effects in disease-associated microenvironments.
   In this study, we investigated the therapeutic potential of exosomes
   from MSCs stimulated with plasma from patients with liver failure
   (LF-Exos).

Methods

   Untreated exosomes (NC-Exos) and LF-Exos were extracted and
   characterized by nanoparticle tracking analysis (NTA), transmission
   electron microscopy (TEM), western blotting, and miRNA sequencing. We
   then examined the protective effects of LF-Exos on hepatocytes acutely
   injured by D-galactosamine (D-GalN)/lipopolysaccharide (LPS)
   co-treatment and on a mouse model of acute liver failure (ALF).
   Apoptosis was assessed using the CCK-8 assay and flow cytometry. Liver
   tissue damage was examined by hematoxylin and eosin staining and
   immunohistochemistry. The levels of signaling pathway proteins were
   determined by western blotting.

Results

   Stimulation with plasma from patients with liver failure significantly
   altered the morphology of MSCs and reduced their proliferative
   activity. Gene chip analysis identified 31 differentially expressed
   miRNAs, and further analysis showed that these differentially expressed
   miRNAs may affect the PI3K-AKT signaling pathway. Compared to NC-Exos,
   LF-Exos induced AKT phosphorylation in hepatocytes and liver tissues,
   inhibited D-GalN/LPS-induced apoptosis in hepatocytes, and reduced
   pathological liver injury in the mouse model of ALF.

Conclusion

   The biological effects of Exos were improved after stimulation with
   plasma from patients with liver failure. LF-Exos may inhibit the
   activity of the NLRP3 inflammasome and activate the PI3K-AKT signaling
   pathway to exert protective effects on acutely injured hepatocytes and
   a mouse model of ALF.

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s13287-025-04163-2.

   Keywords: Exosomes, Mesenchymal stem cells, Liver failure

Background

   Liver function impairment has a variety of etiologies. When liver
   failure develops, short-term mortality is high [[30]1, [31]2].
   Currently, the most effective clinical treatment for liver failure is
   liver transplantation; however, there are problems such as a shortage
   of donors, high cost, and immune rejection [[32]3]. Therefore,
   additional treatment modalities are urgently needed.

   Mesenchymal stem cells (MSCs) are low immunogenic cells that possess
   self-renewal and multi-lineage differentiation potentials [[33]4].
   Previous studies have shown that transplantation of MSCs is an
   effective treatment for liver failure, and they play a role in the
   treatment of various diseases by regulating immune function, promoting
   cell regeneration, and tissue repair [[34]5]. However, MSC
   transplantation is associated with the risks of capillary blockage and
   teratoma formation and has demanding transport and storage conditions
   [[35]6–[36]8]. Research has shown that MSC-derived exosomes (MSC-Exos)
   play a similar therapeutic function as parent cells [[37]9] and have
   advantages over parent cells, including a smaller size, a lower
   immunogenic response in affected tissue, no risk of tumor formation,
   and being easier to produce and store [[38]10–[39]13]. Therefore,
   cell-free therapy based on MSC-Exos has been widely researched in
   recent years.

   Previous studies have shown that exosomes are heterogeneous; exosomes
   derived from different parent cells or parent cells cultured under
   different conditions show differential expression of nucleic acids,
   lipids, proteins, and other contents [[40]14]. The heterogeneity of
   exosomes affects their functional impact on recipient cells and plays
   various roles in immunomodulation, cell proliferation, and apoptosis
   [[41]15, [42]16]. Applying the microenvironment of the disease to
   condition MSCs can enhance the therapeutic role of MSC-Exos [[43]17,
   [44]18]; therefore, based on the heterogeneity of exosomes, training
   MSCs and deriving “trained” MSC-Exos for targeted therapeutic effects
   were explored in the current study.

   In this study, we stimulated MSCs with plasma from patients with liver
   failure (LF plasma) and extracted plasma-stimulated MSC-derived
   exosomes (LF-Exos), observed the differences between LF-Exos and
   untreated exosomes (NC-Exos), and assessed the therapeutic effects of
   LF-Exos in acutely injured hepatocytes and a murine model of acute
   liver failure (ALF).

Methods

Cell culture

   Umbilical cord mesenchymal stem cells (hUCMSCs) were generously
   provided by National Engineering Research Center of Cell Products,
   AmCellGene Engineering Co., Ltd Tianjin China, and cultured in
   low-glucose Dulbecco’s modified Eagle’s medium (DMEM, Gibco, USA,
   12320032) containing 10% fetal bovine serum (FBS, Vivacell, China,
   [45]C04400) and 1% penicillin–streptomycin (Solarbio, China, P1400).
   After 12 h, the medium was replaced with culture medium containing LF
   plasma; medium without added plasma was used as a control. When the
   cells reached 70% confluence, they were incubated in a serum-free
   medium for 48 h, and the supernatant was collected.

   A mouse AML12 hepatocyte cell line was obtained from Wuhan Pricella
   Biotechnology Co., Ltd. AML12 cells were cultured in DMEM/F12 (Gibco,
   USA, 11330032) containing 10% FBS; 1% penicillin–streptomycin; insulin,
   transferrin, and selenium (ITS) Liquid Media Supplement (Beyotime,
   China, C0341); and 40 ng/mL dexamethasone (MCE, China, HY-14648).

   Murine macrophage RAW264.7 cells were obtained from our central
   laboratory and cultured in DMEM containing 10% FBS and 1%
   penicillin–streptomycin.

Proliferation analysis

   MSCs were cultured overnight in 96-well plates (3,000 cells/well) and
   treated with different concentrations of LF plasma. A CCK-8 (Dojindo,
   Japan, CK04) solution was added to the wells according to the
   manufacturer’s protocol to measure the cell proliferation, and the
   optical density at 450 nm (OD450) was read using a multifunctional
   microplate reader (Thermo, Synergy H1).

Exosome isolation and identification

   The cell culture supernatant was centrifuged at 4 °C and 300×g for
   10 min to remove cells, 3,000×g for 15 min to remove dead cells, and
   10,000×g for 30 min to remove subcellular components. The supernatant
   was transferred to an ultracentrifuge tube and centrifuged at 120,000×g
   for 90 min to obtain the sediment containing exosomes, which were
   resuspended in pre-cooled PBS. The protein concentration of the
   exosomes was determined using a bicinchoninic acid (BCA) assay
   (Huaxingbio, China, HX18651), and exosomes were identified using
   transmission electron microscopy (TEM), nanoparticle tracking analysis
   (NTA), and western blotting.

Exosome miRNA sequencing and sequence analysis

   Total RNA was extracted from the exosomes using TRIzol reagent (Life
   Technologies, Carlsbad, CA, USA) and purified using an RNeasy Mini Kit
   (Qiagen, Valencia, CA, USA, 74104). A GeneChip miRNA array (Thermo
   Fisher, USA, 952052) was used to detect the relative expression levels
   of exosomal miRNAs. Biotinylated cDNA was prepared according to the
   standard Affymetrix protocol, using the Ambion^® WT Expression Kit.
   Following labeling, the fragmented cDNA was hybridized, washed, and
   labelled using an Affymetrix Fluidics Station 450. All arrays were
   scanned using the Affymetrix GeneChip Command Console (AGCC).

Treatment of cells with exosomes

   AML12 cells were cultured in cell culture plates (2.5 × 10^5 cells/mL)
   for 24 h. Next, 4 mg/mL D-Galactosamine (D-GalN; Solarbio, China,
   G8110), 40 µg/mL lipopolysaccharide (LPS; Sigma-Aldrich, L2880), and
   20 µg/mL exosomes were added to the AML12 cells and incubated for 24 h.
   The CCK-8 and 7AAD Apoptosis Assay Kits (Elabscinence, China,
   E-CK-A218) were used to measure the effect of exosomes on AML12 cells,
   according to the manufacturer’s instructions.

   RAW264.7 macrophages were cultured in 6-well plates (5 × 10^5 cells/mL)
   overnight and then co-cultured with exosomes (20 µg/mL) for 6 h.
   Subsequently, 200 ng/mL LPS was added to the wells and incubated for
   24 h, followed by total cell protein collection.

Mouse model of ALF and treatment with exosomes

   To establish the ALF model, 8-week-old male C57BL/6 mice were injected
   with D-GalN (250 mg/kg) and LPS (2.5 mg/kg), followed by immediate tail
   vein injection of NC-Exos, LF-Exos (100 µg), or an equal volume of PBS.
   The mice were anesthetized 6 h after the intervention using 5%
   isoflurane (RWD Life Science, China, H-601) by inhalation, obtained
   retro-orbital blood sampls, and euthanized by cervical dislocation,
   liver samples were collected for further analysis. This study has been
   reported in accordance with the ARRIVE guidelines 2.0, and additional
   supporting documents are provided in the supplementary materials.

   Blood samples were centrifuged at 3,000 rpm for 15 min to collect
   serum, and the serum levels of alanine aminotransferase (ALT) and
   aspartate aminotransferase (AST) were measured according to the
   manufacturer’s protocols for the test kits (NJJCBIO, China, C009/C010).

   A portion of liver tissue was fixed in 4% paraformaldehyde, dehydrated
   with concentration gradients of ethanol, and embedded in paraffin.
   Successive sections were cut and underwent hematoxylin and eosin (HE)
   and TUNEL staining to observe the histopathological changes. The
   remainder of the tissue samples was preserved at -80℃ for western blot
   analysis.

Western blot analysis

   Proteins from cell or tissue samples were extracted using RIPA lysate
   buffer containing 1% protein phosphatase inhibitor, quantified using
   BCA analysis, diluted with sample buffer, and boiled at 105 °C.
   Proteins were separated using sodium dodecyl sulfate-polyacrylamide gel
   electrophoresis (SDS-PAGE) and transferred to polyvinylidene fluoride
   (PVDF) membranes. After blocking with skim milk, membranes were
   incubated overnight with the primary antibody at 4 °C on a shaker. The
   blots were incubated with the horseradish peroxidase (HRP)-conjugated
   secondary antibody (1:5000) for 1 h at room temperature. The primary
   anti-AKT (1:1000, [46]AB179463) antibody was purchased from Abcam.
   Anti-phospho-AKT (1:2000, #4060), anti-phospho-PI3K (1:1000, #4228),
   and anti-ASC (1:1000, #67824) antibodies were purchased from CST. The
   anti-PI3K (1:1000, [47]R22768) antibody was purchased from Zenbio.
   Anti-NLRP3 (1:1000, AG-20B-0014) and anti-mouse caspase-1 (1:1000,
   AG-20B-0042) antibodies were purchased from Adipogen. Anti-mouse
   cleaved IL-1β (1:2000; AF-401-SP) was purchased from R&D Systems.
   Anti-GAPDH (1:2000, Proteintech, 60004-1) and anti-lamin B (1:1000,
   Proteintech, 10895-1-AP) antibodies were used as internal controls.

Statistical analysis

   SPSS software (version 19) was used for statistical analyses, and
   GraphPad Prism (version 8.3) was used to create images. Data are
   presented as the mean ± standard deviation ( Inline graphic ± SD). The
   Student’s t-test was used for comparisons between two samples, and
   one-way analysis of variance (ANOVA) was used for comparisons between
   multiple samples. P-values < 0.05 were considered statistically
   significant.

Results

LF plasma affected the morphology and proliferation of MSCs

   We observed the morphology of MSCs under a microscope. Normal cultured
   MSCs showed a spindle-like shape, grew in spiral arrangements, were
   packed tightly with a uniform morphology and size, and had clear ovoid
   nuclei in the center of the cells. MSCs cultured with LF plasma were
   wider and had enlarged gaps between cells, a disordered cellular
   arrangement, and many coarse particles around the nuclei (Fig. [48]1A).
   The CCK-8 analysis showed that the proliferation curves of normal
   cultured MSCs showed typical “S” shapes. However, increasing
   concentrations of LF plasma inhibited the growth of MSCs (Fig. [49]1B).

Fig. 1.

   [50]Fig. 1
   [51]Open in a new tab

   Morphology and proliferation of MSCs. A. Representative photographs of
   MSCs stimulated with LF plasma after 48 h; B. Proliferation rate of
   MSCs stimulated with LF plasma in different concentrations (scale bar
   = 500 μm)

Identification of MSC-Exos

   Two groups of MSC-Exos were isolated from the cell culture supernatant.
   TEM showed that the particles extracted from the two groups of
   supernatants had teacup or elliptical disc shapes and double-membrane
   structures typical of exosomes (Fig. [52]2A). NTA revealed that the
   NC-Exos and LF-Exos size was 30–200 nm, which was the size of exosomes
   (Fig. [53]2B). Western blotting confirmed that both NC-Exos and LF-Exos
   expressed the specific exosomal surface markers CD63, TSG101, and Alix
   and did not express calnexin, which is not expressed on exosomes
   (Fig. [54]2C). In addition, the protein content of LF-Exos was
   significantly higher than that of NC-Exos (Fig. [55]2D). These results
   confirm that the extracted particles were exosomes.

Fig. 2.

   [56]Fig. 2
   [57]Open in a new tab

   Characterization of NC-Exos and LF-Exos. A. TEM Images showed that both
   groups of exosomes had a teacup or ellipsoidal morphology (scale bar
   = 200 nm); B. NTA detected the range of particle size distribution of
   NC-Exos and LF-Exos (size: nm); C. Western Blot confirmed the
   expression of exosomal surface markers TSG101, Alix, CD63, and
   endoplasmic reticulum protein Calnexin in NC-Exos and LF-Exos
   (Full-length blots are presented in Supplementary Fig. [58]1); D.
   Protein concentration in NC-Exos and LF-Exos; ***P < 0.001

LF plasma induces changes in MSC-Exos miRNA

   We sequenced the miRNA extracted from NC-Exos and LF-Exos and screened
   for miRNAs that were differentially expressed between NC-Exos and
   LF-Exos. We identified 31 differentially expressed miRNAs; of these, 11
   were upregulated and 20 were downregulated (|fold change|>1.2,
   P < 0.05) (Fig. [59]3A-B).

Fig. 3.

   [60]Fig. 3
   [61]Open in a new tab

   Analysis of differentially expressed miRNAs between NC-Exos and
   LF-Exos. A. Volcano plot of differentially expressed miRNAs between
   NC-Exos and LF-Exos (|Fold change|>1.2; P < 0.05); B. Heat map of
   differentially expressed miRNAs between NC-Exos and LF-Exos; C. GO and
   KEGG pathway enrichment analyses of target genes matched upregulated
   differentially expressed miRNAs; D. GO and KEGG pathway enrichment
   analyses of target genes matched to upregulated differentially
   expressed miRNAs

   The miRanda and RNAhybrid algorithms were used to predict target gene
   interaction binding sites for miRNAs to match the target relationship
   between the predicted miRNAs and mRNAs. Using the intersection of the
   matched mRNAs from the two databases, 5,933 predicted mRNAs were
   obtained. The predicted target mRNAs were subjected to gene ontology
   (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway
   analyses. GO enrichment analysis showed that these genes were
   significantly enriched in the regulation of signaling, cell junctions,
   and protein binding. KEGG signaling pathway enrichment analysis showed
   that these genes were significantly enriched in the PI3K-AKT, Wnt, and
   other signaling pathways (Fig. [62]3C-D).

MSC-Exos inhibited apoptosis in D-GalN/LPS-induced hepatocytes

   The CCK-8 assay results showed that MSC-Exos promoted the viability of
   AML12 cells under the inhibitory effect of D-GalN/LPS and that the
   viability of AML12 cells was significantly increased in the LF-Exos
   group (Fig. [63]4A).

Fig. 4.

   [64]Fig. 4
   [65]Open in a new tab

   Protective effect of NC-Exos and LF-Exos on D-GalN/LPS-induced AML12
   cells. A. Percentage of cell proliferation of D-GalN/LPS-induced AML12
   cells co-cultured with NC-Exos or LF-Exos for 24 h; B. Percentage of
   cell apoptosis of D-GalN/LPS-induced AML12 cells co-cultured with NC
   Exos or LF-Exos for 24 h; C. Annexin V/7AAD staining of
   D-GalN/LPS-induced AML12 cells; *P < 0.05; ***P < 0.001;****P < 0.0001

   The apoptosis assay revealed that D-GalN/LPS significantly increased
   the proportion of apoptotic cells, and both NC-Exos and LF-Exos
   reversed D-GalN/LPS-induced apoptosis in AML12 cells. The apoptosis
   rate of AML12 cells in the LF-Exos group was significantly lower than
   that in the NC-Exos group (Fig. [66]4B-C).

MSC-Exos alleviated liver injury in a d-GalN/LPS-induced ALF mouse model

   We observed the survival time of ALF mice. The mice started to die 3 h
   after D-GalN/LPS injection, and the mortality rate reached 100% within
   8 h. Some mice survived in both groups following tail vein injection of
   NC-Exos or LF-Exos, but the survival analysis showed no significant
   difference (Fig. [67]5A).

Fig. 5.

   [68]Fig. 5
   [69]Open in a new tab

   MSC-Exos ameliorated D-GalN/LPS-induced ALF in mice. A. Survival curves
   after mice were injected with D-GalN/LPS (n = 10); B. Serum ALT level
   of mice in each group; C. Serum AST levels of mice in each group; D.
   Hepatic organ coefficient of mice in each group; E. Gross morphology of
   the liver and HE staining analyzed liver sections in each group (scale
   bar = 250 μm); F. TUNEL staining of liver sections in each group;
   *P < 0.05; **P < 0.005; ****P < 0.0001

   Mice in the model group had significantly elevated serum ALT and AST
   levels, indicating that D-GalN/LPS caused significant liver injury.
   After injection with NC-Exos or LF-Exos, the serum ALT and AST levels
   were significantly decreased. However, the serum ALT level of the
   LF-Exos group was significantly lower than that of the NC-Exos group
   (Fig. [70]5B-C). This may be because ALT is mainly distributed in the
   cytoplasm of hepatocytes and is more sensitive to early liver injury,
   while AST is mainly distributed in the mitochondria of hepatocytes, and
   mitochondrial AST is only released when hepatocytes are severely
   damaged and necrotic [[71]19], Therefore, we only observed a
   significant decrease in ALT in the mouse model of ALF.

   The livers of mice in the model group were generally congested and
   enlarged, and the hepatic organ coefficient (liver weight/body weight;
   LW/BW) was significantly increased (Fig. [72]5D). The mouse liver
   pathology was examined. HE staining showed that the liver structure in
   normal mice was clear and obvious, and the hepatocytes were tightly
   arranged around the central vein. The livers of the model group were
   significantly damaged, blurred liver lobule structures, large areas of
   denatured and necrotic hepatocytes, and inflammatory cell infiltration.
   In the NC-Exos and LF-Exos groups, the hepatic lobular structure was
   restored and the necrotic area and inflammatory cell infiltration were
   reduced; the LF-Exos group had more appreciable improvements than the
   NC-Exos group (Fig. [73]5E).

   Mouse hepatocyte apoptosis was detected using TUNEL staining;
   TUNEL-positive hepatocytes were observed in the liver tissue of the
   model group, and the number of TUNEL-positive cells in the NC-Exos and
   LF-Exos groups was reduced (Fig. [74]5F).

Effect of MSC-Exos on the PI3K-AKT signaling pathway and NLRP3 inflammasome
in D-GalN/LPS-injured hepatocytes and ALF mice

   Based on the miRNA sequencing analysis, we speculated that LF-Exos
   might ameliorate ALF progression via the PI3K-AKT signaling pathway. In
   addition, the PI3K-AKT signaling pathway may be associated with the
   NLRP3 inflammasome [[75]20, [76]21], and activation of the NLRP3
   inflammasome is also a factor in liver failure exacerbation [[77]22,
   [78]23].

   We analyzed the expression of proteins in the PI3K-AKT signaling
   pathway and NLRP3 inflammasome and found that LF-Exos significantly
   increased the expression of PI3K and p-AKT in AML12 cells and ALF mice.
   In addition, the expression of NLRP3 in AML12 cells, RAW264.7 cells,
   and ALF mice was significantly reduced (Fig. [79]6).

Fig. 6.

   [80]Fig. 6
   [81]Open in a new tab

   LF-Exos ameliorates ALF by promoting the PI3K-AKT signaling pathway and
   inhibiting NLRP3 inflammasome activation. A. Western blot analysis of
   the protein expression in AML12 cells; B. Western blot analysis of the
   protein expression in RAW264.7 cells; C. Western blot analysis of the
   protein expression in the livers of mice in each group. Full-length
   blots are presented in Supplementary Figures [82]2–[83]4, respectively.
   *P < 0.05, **P < 0.005, ***P < 0.001

Discussion

   MSC-Exos have demonstrated therapeutic effects similar to those of MSCs
   and are an effective therapeutic prospect for ALF [[84]24–[85]26].
   Researchers have found that “training” MSCs may improve the therapeutic
   effects of MSCs and MSC-Exos [[86]27–[87]29], and effective MSC
   stimulation may make MSC-Exos more targeted and efficient in treating
   diseases. Thus, determining effective stimulation has become a popular
   topic in exosome research. In this study, we selected LF plasma for MSC
   stimulation, observed the differences between NC-Exos and LF-Exos, and
   preliminarily verified the therapeutic effects of LF-Exos on ALF.

   Exosome heterogeneity leads to different compositions of MSC-Exos that
   have been stimulated in different ways [[88]30]. miRNA is one of the
   most abundant components of exosomes and the main biologically active
   substance [[89]31]. miRNA can inhibit gene expression and protein
   synthesis by competitively binding to mRNA [[90]32], which plays
   important roles in physiopathological processes. Shao et al. [[91]31]
   stimulated hUCMSCs with IL-6, obtained exosomes (Exos-IL6), and
   sequenced small RNAs extracted from Exos-IL6 and Exos-NC; they
   identified 37 differentially expressed miRNAs that were widely involved
   in the regulation of IL-6-related signaling pathways, including the
   PI3K-AKT signaling pathway. Ma et al. [[92]33] obtained exosomes from
   TNF-α-pretreated MSCs (T-Exos); some of the 180 miRNAs differentially
   expressed between T-Exos and NC-Exos were closely related to immunolog-
   and inflammation-related signaling pathways. Our study found that
   stimulation of MSCs with LF plasma affected their morphology and
   inhibited their proliferation. Sequencing results showed 31
   differentially expressed miRNAs between NC-Exos and LF-Exos, and KEGG
   analysis predicted that these differentially expressed miRNAs were
   related to the PI3K-AKT signaling pathway.

   The mechanisms by which MSC-Exos ameliorate ALF are complex and varied;
   however, most are closely related to the regulation of immune
   homeostasis, inhibition of apoptosis, and promotion of the regenerative
   functions of the remaining cells. The PI3K-AKT signaling pathway is a
   classical pathway that is closely related to a wide range of
   physiological and pathological processes, such as cell growth,
   proliferation, migration, metabolism, and immunomodulation [[93]34,
   [94]35]. Previous studies have revealed that IGF-1R/PI3K/AKT signaling
   plays an important role in the development of acetaminophen
   (APAP)-induced ALF [[95]36]. The aberrant activation of the NLRP3
   inflammasome mediates liver failure by stimulating procaspase-1 and
   pro-IL-1β signaling [[96]37]. Zhang et al. found TNF-α-pretreated
   hUCMSC-Exos affect downstream ALT, AST, and inflammatory factors by
   inhibiting the activation of the NLRP3-associated pathway [[97]38]. A
   relationship between the PI3K-AKT signaling pathway and the NLRP3
   inflammasome has been reported. Zhao et al. [[98]20] revealed that AKT
   acts directly on NLRP3, inhibits NLRP3 oligomerization during LPS
   priming, and reduces cytokine secretion following NLRP3 activation.
   Wang et al. [[99]21] found that NLRP3 inflammasome activation is
   attributable to inhibition of the ROS-PI3K/AKT pathway. Tabaa et al.
   [[100]27] found that suppression of NLRP3 activation regulates the
   PI3K/AKT/mTOR cascade. Our results suggest that treatment with LF-Exos
   can reduce ALT and AST levels and alleviate the liver damage caused by
   ALF. Exosomes can activate the PI3K-AKT signaling pathway via miRNAs
   and may inhibit the activation of the NLRP3 inflammasome, which
   promotes cell proliferation, inhibits apoptosis, and causes
   immunomodulation.

   MSC-Exos have been widely noticed as an emerging therapeutic
   technology, and many studies have confirmed that MSC-Exos have a good
   therapeutic effect on liver failure; however, obtaining more
   therapeutically useful exosomes and clinically applying them remains a
   great challenge. Our team has conducted research on bioartificial liver
   therapy, which involves combining extracorporeal circulation with a
   bioreactor. Hepatocytes are usually inserted into the bioreactor, and
   the patient’s plasma is exchanged with the cells in the bioreactor and
   then returned to the blood, thus promoting the repair of liver damage.
   Our results showed that the modification of MSCs with LF plasma can
   alter the miRNA expression in LF-Exos and may play a role in inhibiting
   apoptosis and regulating immune responses in injured cells and ALF mice
   by activating the PI3K-AKT signaling pathway and inhibiting the
   activation of the NLRP3 inflammasome. LF-Exos may play an enhanced
   therapeutic role in ALF with the placement of MSCs in the bioreactor to
   create an MSC-based bioartificial liver. This will provide a basis for
   future research on MSC-based bioartificial livers.

Conclusion

   In summary, our study revealed that LF plasma-stimulation of MSCs can
   regulate the biological function of MSC-Exos and that LF-Exos can play
   a protective role against acute liver injury or liver failure by
   activating the PI3K-AKT signaling pathway and inhibiting the activation
   of the NLRP3 inflammasome. This study lays the foundation for obtaining
   therapeutically useful MSC-Exos and developing novel bioartificial
   liver systems.

Electronic supplementary material

   Below is the link to the electronic supplementary material.
   [101]Supplementary Material 1^ (731.1KB, xlsx)
   [102]Supplementary Material 2^ (9.3MB, pdf)
   [103]Supplementary Material 3^ (369.4KB, docx)

Acknowledgements