Introduction

Various hepatoxic factors, such as viruses, drugs, lipid deposition, and autoimmune responses, induce acute or chronic liver injury, [1, 2] and 3.5% of all worldwide deaths result from liver cirrhosis, liver failure, or hepatocellular carcinoma [3]. Orthotopic liver transplantation is performed in end-stage liver disease when the remanent liver tissue is unable to regenerate in sufficient time. However, liver transplantation is currently limited by few liver donors, expensive surgical costs, and severe immune rejection. Cell therapy, including hepatocyte transplantation and stem cell transplantation, has recently become an attractive option to reduce the overall need for liver transplantation and reduce the wait time for patients. However, the clinical application of hepatocyte transplantation is limited by the finite proliferative capacity and missing liver functions of human primary hepatocytes, even though this strategy is simpler, less invasive, and safer than surgery [4]. Stem cells are multipotent cells that are able to proliferate, self-renew, and differentiate into various somatic cells in vitro, and they have attracted much attention for restoring tissue function in vivo with little immune rejection. Stem cells can be classified into three categories: embryonic stem cells (ESCs), induced pluripotent stem cells (iPSCs), and mesenchymal stem cells (MSCs) [5]. ESCs are stem cells derived from early-stage embryos and have high potential to undergo multilineage differentiation. Takahashi and Yamanaka transformed mouse fibroblasts into iPSCs by retroviral transduction of four transcription factors (TFs), Oct4, Sox2, Klf4c, and c-Myc [6]. The multipotency and self-renewal ability of these reprogrammed cells are similar to those of ESCs [7]. MSCs can be isolated from various tissues, such as adipose tissue, bone marrow, the placenta, the amnion, menstrual blood, and other tissues [8,10,11,66]. Quantification of EVs remains extremely challenging, and new methods, such as nanoflow cytometry, direct stochastic optical reconstruction microscopy, ExoCounter with optical disc technology, and imaging flow cytometry, have been developed to effectively quantify EVs [67,68,69,70]. The purification process can be described as three steps: filtration to remove the cellular debris, concentration of the CM, and exosome isolation from the concentrated CM [71]. To achieve higher specificity, investigators should remove impurities step-by-step. Several proteins, including CD9, CD63, CD81, tumor susceptibility gene (TSG)101, and Alix, which are highly enriched in exosomes, are recommended as specific markers for determining the identity of EVs [72,73,74,75]. To determine the purity for quality control, the particle-to-protein, protein-to-lipid, or RNA-to-particle ratios are common markers for monitoring the purity of EVs [72]. There are also some impurities, such as antibiotics and serum, that should be excluded from the cell culture medium when monitoring the removal of potential hazardous factors [76]. On the other hand, another criterion is the absence of several proteins, including histones, lamin A/C, glucose-regulated protein 94, Golgi matrix protein 130, and cytochrome C, since strict cellular localization prohibits the enrichment of these intracellular proteins [72]. Although various biological and biochemical assays are used to determine the potency of EVs or exosomes, it is still difficult to establish a single, standard potency assay since the cargos of EVs in vivo are so complex [77, 78].

MSC-derived soluble factors are effective for treating liver diseases

Transplantation of MSC-CM and exosomes is the main therapy for treating acute or chronic liver injury (Table 1). Transplantation significantly improved liver function by inhibiting apoptosis, necrosis, oxidative stress, inflammation, infiltration of immune cells, immune rejection, HSC activation, etc. (Fig. 1).

Table 1 Transplantation of MSC-CM and exosomes is an effective therapy for attenuating acute or chronic liver injury via different mechanisms
Full size table
Fig. 1
figure 1

MSCs isolated from various tissues such as adipose tissue, bone marrow, endometrium, umbilical cord, and other tissues are able to secret a large number of soluble factors. These soluble factors have various biological functions including immunoregulation, inhibition of HSC activation, anti-inflammation, and anti-oxidative stress. Moreover, they further decrease the cell death rate and improve liver regeneration for preventing liver injury and maintaining liver homeostasis

Full size image

MSC-CM for treating liver diseases

Infusion of MSC-CM immediately before liver irradiation obviously inhibited the apoptosis of sinusoidal endothelial cells (SECs) and the histopathological changes in the irradiated liver by attenuating liver inflammation [79]. MSC-CM attenuated liver injury and provided a survival benefit in rats with 5% reduced-size liver transplantation (RSLT) by reducing the apoptosis rate of hepatocytes and SECs and reducing the secretion of proinflammatory cytokines, infiltration of neutrophils, and activation of Kupffer cells. Furthermore, the expression of vascular endothelial growth factor (VEGF) and matrix metallopeptidase 9 was also improved to promote liver regeneration in the liver grafts of recipient rats [80]. MSC-derived CM partially ameliorates ALF while significantly suppressing fibrogenesis, necroinflammation, liver apoptosis, and activation of HSCs and improving glycogen synthesis and storage and liver regeneration. Furthermore, MSC-CM reduced infiltrating macrophages, converted CD4+ T lymphocytes into anti-inflammatory Tregs and Th2 cells, and facilitated HSC death in liver tissue [81]. Both MSC-CM and interleukin (IL)-1Ra isolated from MSC-CM significantly reduced liver fibrosis by attenuating liver inflammation and inducing alpha-1 type I collagen, transforming growth factor (TGF)-β1, and tissue inhibitor of metalloproteases 1 [82].

MSC-derived exosomes for treating liver diseases

MSC-derived soluble factors isolated from MSC-CM can also protect against liver inflammation and liver injury. Tamura et al. demonstrated that exosomes escape being trapped in lung tissue and are largely distributed in injured liver tissue. MSC-derived exosomes comparably improved liver functions and hepatocyte proliferation but decreased the apoptosis rate and scope of liver necrotic areas by exerting anti-inflammatory effects and activating Tregs in comparison to those of the MSC group [83]. These exosomes were proven to exert protective effects on liver injury models in a dose-dependent manner. Du et al. demonstrated that MSC-derived exosomes significantly upregulated hepatocyte proliferation in a dose-dependent manner and could directly fuse with target hepatocytes. Consequently, MSC-derived exosomes decreased the histopathological scores and serum levels of aminotransferases by inhibiting hepatocyte necrosis and sinusoidal congestion and upregulating hepatocyte proliferation in the ischemia/reperfusion injury model [84]. Exosomes from human umbilical cord-derived MSCs significantly decreased the serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) and repaired injured liver tissue in an ALF mouse model after reducing TXNIP/NLRP3 inflammasome activation and its downstream inflammatory factors, including caspase-1, IL-1β, and IL-6 [85]. MSC-derived exosomes dose-dependently inhibited apoptosis in D-galactoside (D-GalN)/lipopolysaccharide (LPS)-treated AML12 cells in vitro. In addition, transplantation of exosomes significantly inhibited hepatocyte apoptosis and increased the survival rates of D-GalN/LPS-induced ALF mice by migrating to liver tissue, inhibiting liver mononuclear cells and downregulating the apoptotic protein caspase-3 [86]. Jiang et al. showed that MSC-derived exosomes alleviated carbon tetrachloride (CCl4)-induced acute liver injury and liver fibrosis via the inhibition of oxidative stress and hepatocyte apoptosis and further improved the antioxidant and hepatoprotective effects of bifendate treatment [87]. In addition, exosomes more significantly alleviated CCl4-induced liver fibrosis than MSCs, as demonstrated by a reduction in collagen accumulation and enhanced liver functions through a reduction in liver inflammation, upregulation of hepatocyte regeneration, and inhibition of HSC activation through the Wnt/β-catenin pathway [88]. Exosomes from human umbilical cord MSCs significantly protected against liver dysfunction, as shown by decreased ALT levels and TGF-β1 and phosphorylated Smad2 expression. Exosomes also decreased collagen deposition and ameliorated CCl4-induced liver fibrosis by inhibiting epithelial-to-mesenchymal transition (EMT), attenuating hepatic inflammation, and eliminating hepatocyte apoptosis [89].

Exosomes from gene-modified MSCs for treating liver diseases

Intriguingly, gene modification of MSCs provides additional therapeutic effects of exosomes in liver diseases. Liu et al. showed that exosomes from adipose-derived MSCs also improved liver functions through a similar pathway. However, knockout of miR-17 in MSCs abolished the therapeutic effects of exosomes from miR-17-knockout MSCs in ALF mice, which indicated that miR-17 plays a critical role in the anti-inflammatory function of exosomes [90]. A dose-dependent effect was also observed after gavage administration of MSC-derived exosomes. Although 8 mg/kg MSC-derived exosomes failed to rescue ALF animals, 16 and 32 mg/kg MSC-derived exosomes improved the survival rate to 60 and 90%, respectively. In addition, Yan et al. demonstrated that exosomes effectively rescued ALF mice by reducing oxidative stress and apoptosis in liver tissue, and knockdown of glutathione peroxidase 1 (GPX1) in MSCs abrogated the protective effects of exosomes and inhibited the recovery of hepatic oxidative injury in ALF mice [91].

How can the therapeutic effects of MSC-based cell-free strategies be improved?

To further improve the therapeutic effects, modification of the MSC microenvironments and gene modification of MSCs can consequently alter the composition of proteins and vesicular contents of MSC-derived soluble factors (Table 2).

Table 2 Improvement in the efficacy of MSC-derived soluble factors in treating liver diseases via different mechanisms
Full size table

Pretreatment or cotreatment for MSC-based cell-free therapy

The common oxygen level is approximately 21% O2 for standard cell culture, and this oxygen concentration is not the same as that in the in vivo microenvironment. The physiological oxygen concentration in tissues varies from 1% in cartilage and bone marrow to 12% in peripheral blood [92]. To improve MSC multipotency, proliferation, and secretion of cytoprotective molecules, hypoxic conditions have been proven to significantly improve the survival rate of MSCs in the harsh environments of injury sites after transplantation [93, 94]. CM derived from hypoxic (10% O2) MSCs more significantly improved the biochemical parameters and histological results in acetaminophen-induced ALF models by reducing liver inflammation and improving liver regeneration than MSC-CM isolated from a normal oxygen culture environment [95]. CM from MSCs cultured at lower (5%) O2 levels more significantly inhibited the activation of inflammation and attenuated hepatocyte necrosis in liver tissue than CM from normoxic MSCs, and hypoxic MSC-derived CM more robustly downregulated oxygen tension and accelerated the healing process in damaged liver tissue of ALF models than CM from normoxic MSCs. Furthermore, CM derived from MSCs cultured in 1% hypoxia showed higher levels of IL-6, tumor necrosis factor (TNF)-α, hepatocyte growth factor (HGF), and VEGF, and these extremely hypoxic CMs significantly increased the number of proliferating cells in liver tissue via the activation of JAK/STAT3 signaling [96]. Proteins isolated from mildly hypoxic MSCs (5% and 10% O2) showed better effects in treating acetaminophen-induced ALF than those from normal MSCs by inhibiting inflammatory reactions and reducing hepatocyte necrosis in damaged liver. This group further showed that even moderate hypoxia produced discrete changes in the expression of various subsets of proteins responsible for intracellular respiration and cell signaling [97].

To further improve the therapeutic effects of MSC-CM, MSCs were cocultured with hepatocytes in vitro to improve MSC secretion of protective factors. In comparison to MSC-CM and CM from hepatocytes, the cocultured MSC-CM significantly prolonged the survival time of ALF rats by preventing liver injury and promoting liver tissue repair [98]. Cotreatment with melatonin and MSC-derived exosomes markedly decreased the liver injury score and AST level in the liver I/R group. Intriguingly, the cotreatment group showed comparably low levels of inflammatory factors, apoptosis-related proteins, oxidative stress, DNA damage, and mitochondrial damage to those of the normal group [99].

Culture conditions may influence the regenerative and immunomodulatory capacities of MSC-derived soluble factors, and the optimization of culture conditions will contribute to the further regeneration of liver tissues. According to current studies, pretreatment or cotreatment of MSCs with different factors exerts decisive effects on the therapeutic effects of MSC-based cell-free treatment in recipients. However, the study number is limited, and we suggest exploring the detailed treatments of MSCs and their related mechanisms in depth.

Soluble factors from MSCs with gene modifications

Valadi et al. first reported that exosomes effectively transferred functional mRNAs or miRNAs into target cells since they found that there were multiple new mouse-derived proteins in human mast cells after the transplantation of mouse MSC-derived exosomes [100]. Current studies have tried to alter the delivered factors to improve the therapeutic effects of MSC-derived soluble factors.

MSCs were infected with pre-miR-223 and transfected with a miR-223 inhibitor to generate exosomes containing miR-223 and exosomes without miR-223, respectively. Exosomes and exosomesmiR-223(+) significantly reversed autoimmune hepatitis by downregulating cytokines such as NLRP3 and caspase-1, while exosomesmiR-223(−) did not exert protective effects in the experimental model of autoimmune hepatitis. This finding indicated that miR-223 was a key factor participating in the immunoregulation of MSC-derived exosomes to protect against liver injury [101]. In addition, exosomesmiR181-5p(+) delivered miR181-5p to injured liver tissue and inhibited the deposition of collagen I, vimentin, α-SMA, and fibronectin in fibrotic livers after activating autophagy and downregulating STAT3 and Bcl-2 in HSCs [102].

Conclusions

MSC-based cell-free products protect against the potential risks associated with the use of native or engineered MSCs in vivo; moreover, an increasing number of studies have shown that the therapeutic effects of MSC-based cell-free products are equal to those of MSCs. In comparison to MSCs, MSC-derived soluble factors are easier to generate in advance, more stable for storage and transport, and more convenient to control in terms of quality and quantity. As transplantation of MSC-derived soluble factors avoids delivering active cells, infusion of these factors is less immunogenic, less tumorigenic, and less likely to result in exogenic infections, including cytomegalovirus and herpes simplex virus infections [103]. All these advantages make MSC-derived soluble factors promising alternatives to MSC therapy in treating liver diseases. However, current studies have not clarified the detailed factors in MSC-derived soluble factors in vitro and in vivo. The presence of different bioactive molecules mediates the different biological activities of MSC-derived exosomes, which may also influence their final effects in vivo. Furthermore, questions about the characterization, potency, and quantification of MSC-derived exosomes must be addressed before their clinical application is feasible. As MSC-derived soluble factors effectively deliver various cargos in vivo, the identification and purification of these factors should be explored to avoid the administration of potentially harmful vesicles or molecules in the human body. In our opinion, pretreatments or gene modifications may also serve as effective strategies to improve the therapeutic effects of MSC-based cell-free products.