The therapeutic potential of exosomes derived from different cell sources in liver diseases

Abstract

Exosomes are small nanovesicles with a size of approximately 40–120 nm that are secreted from cells. They are involved in the regulation of cell homeostasis and mediate intercellular communication. In addition, they carry proteins, nucleic acids, and lipids that regulate the biological activity of receptor cells. Recent studies have shown that exosomes perform important functions in liver diseases. This review will focus on liver diseases (drug-induced liver injury, hepatic ischemia-reperfusion injury, liver fibrosis, acute liver failure, and hepatocellular carcinoma) and summarize the therapeutic potential of exosomes from different cell sources in liver disease.

INTRODUCTION

The morbidity and mortality associated with liver disease have increased, and more than 2 million people die from liver disease worldwide each year, accounting for ∼4% of all deaths globally (1). Despite the availability of multiple treatment methods for liver disease, acute and chronic liver disease continue to pose major healthcare challenges. Moreover, drug treatment is often ineffective in end-stage liver disease. Although many patients with end-stage liver disease benefit from liver transplantation, the shortage of suitable organ donors remains an ongoing challenge (2–5). Therefore, new liver disease treatment methods are urgently needed.

Extracellular vesicles (EVs) are membranous granules secreted by all types of cells. EVs are found in blood, urine, milk, and tissues (6, 7). On the basis of their biogenesis, EVs are generally categorized as exosomes, microvesicles, and apoptotic bodies (8). Since EVs are present in almost all bodily fluids, they also show great potential as disease biomarkers (9), as do exosomes (10, 11). Exosomes are small vesicles that are between 40 and 120 nm in size and wrapped in phospholipid bilayers (12). In this article, we focus on the therapeutic potential of exosomes derived from different cell sources in liver disease.

EXOSOMAL TREATMENT OF DRUG-INDUCED LIVER INJURY

Drug-induced liver injury (DILI) occurs during clinical medication and is induced by various chemical drugs, biological agents, traditional Chinese medicines (TCM), natural medicines (NM), health products (HP), dietary supplements (DS), and their metabolites. DILI is a common cause of clinical liver disease (especially acute liver injury) and can result in drug research and development failure or drug withdrawal from the market (13–16). DILI can be further classified into liver injury caused by the drug itself and idiopathic liver injury (17). The former occurs due to the liver damage caused by the direct toxicity of drugs or their metabolites to the liver, which can be shown in animal models (18). However, most DILIs are idiopathic, and conventional animal toxicology assessments cannot predict the subsequent clinical toxicity risk (19).

Treatment of DILI with Exosomes Derived from Mesenchymal Stem Cells

Mesenchymal stem cells (MSCs) are a type of pluripotent stem cells that possess all the typical characteristics of stem cells, namely, self-renewal and multidirectional differentiation. Simultaneously, they can also regulate immunity. In comparison with MSC transplantation therapy, treatment with MSC-derived exosomes is less risky, since it avoids the potential for carcinogenicity (20), promotion of tumor growth (21, 22), cell rejection (23), thrombi formation (24), unexpected differentiation (25), or spread of infection. A study by Tamura et al. (26) showed that exosomes derived from bone marrow mesenchymal stem cells (hBM-MSCs-Ex) have an inhibitory effect on immune-mediated liver injury. In a mouse model, exosome treatment was shown to increase the production of regulatory T cells (Tregs), upregulate the expression of transforming growth factor β (TGF-β) and hepatocyte growth factor mRNA, downregulate the expression of the proinflammatory cytokine interleukin (IL)-2, and reduce hepatocyte apoptosis and promote tissue regeneration, thereby inhibiting liver cell damage. The findings obtained after three injections of exosomes confirmed that the protective effects of exosomes on the liver are dose dependent and depend on the characteristics of the cells that secrete them because the exosomes derived from fibroblasts did influence the regulation of immunity in the concanavalin-A-mediated liver injury model. A study by Jiang et al. (27) showed that in a carbon tetrachloride (CCl₄)-induced acute liver injury model, in comparison with the commonly used hepatoprotective agent bidentate (DDB), exosomes derived from umbilical cord mesenchymal stem cells (huc-MSCs-Ex) exhibited powerful antioxidant and hepatoprotective activities by reducing liver inflammation and increasing liver cell regeneration.

Treatment of DILI with Hepatocyte-Derived Exosomes

Hepatocyte-derived exosomes may show therapeutic activity by promoting liver immune tolerance (28). In addition, exosome mimetic nanovesicles from hepatocytes, similar to hepatocyte-secreted exosomes, promote hepatocyte proliferation (29). However, Nojima et al. (30) found that only exosomes derived from primary hepatocytes can promote hepatocyte proliferation, and their effect is dose dependent. In contrast, exosomes derived from Kupffer cells show inhibitory effects on hepatocyte proliferation.

EXOSOMAL TREATMENT OF HEPATIC ISCHEMIA-REPERFUSION INJURY

Liver oxygen supply is related to the blood oxygen content perfused to the liver and the total blood flow. Liver ischemia occurs when the oxygen supply and demand are out of balance (31). In vitro experiments and animal models have suggested that due to their higher metabolic rate, hepatocytes are more susceptible to hypoxic damage than other types of liver cells such as sinusoidal endothelial cells, Kupffer cells, or bile duct cells at physiological temperature (32).

Nutritional status also has an important influence on the sensitivity of liver cells to hypoxia. In comparison with customarily fed rats, fasted rats were shown to be less prone to hepatic ischemia. One possible reason is that the liver glycogen stores of fasted rats are reduced, resulting in reduced substrates of anaerobic glycolysis and decreased intracellular lactate production (33). Ischemia also promotes the conversion of cytosolic xanthine dehydrogenase to xanthine oxidase, which causes the accumulated xanthine to produce superoxide and hydrogen peroxide (34). Reperfusion of the ischemic liver will further promote liver damage since ischemic liver cells exposed to reoxygenation produce reactive oxygen species, which leads to liver cell damage through lipid peroxidation (32). In addition, Kupffer cells produce cytokines during ischemia, including tumor necrosis factor-α, which triggers the recruitment and activation of polymorphonuclear leukocytes. Thus, the treatment of hepatic ischemia-reperfusion injury (IRI) remains a clinical challenge. In recent years, some progress has been made in the treatment of hepatic IRI by using exosomes.

Treatment of Hepatic IRI with Exosomes Derived from MSCs

Du et al. (35) showed that exosomes secreted by human pluripotent stem cells (hiPSC-MSCs-Ex) have a significant dose-dependent protective effect on hepatic IRI. The hiPSC-MSCs-Ex can significantly reduce the levels of the hepatocyte injury markers aspartate aminotransferase (AST) and alanine transaminase (ALT) and significantly increase the levels of the hepatocyte proliferation markers proliferating cell nuclear antigen (PCNA) and phosphohistone H3 (pHH3) in the ischemia-reperfusion model. HiPSC-MSCs-Ex activates the hepatocyte sphingosine kinase and sphingosine-1-phosphate signaling pathways to promote cell proliferation and reduce hepatic IRI. In addition, exosomes derived from BM-MSCs have been shown to reduce hepatic IRI by inhibiting liver cell apoptosis and stimulating liver cell regeneration (36). This effect can be enhanced by glycyrrhetinic acid (37). Another study showed that miR-20a, an exosome derived from umbilical cord mesenchymal cells (huc-MSCs-Ex), plays a protective role in hepatic IRI. The mechanism of action underlying this finding was shown to involve inhibition of hepatocyte apoptosis (38). A similar mechanism of action has also been attributed to exosomes from adipose-derived mesenchymal stem cells (ADSCs-Ex). Researchers have confirmed in a rat ischemia-reperfusion model that ADSCs-Ex can regulate mitochondrial dynamics and biogenesis, maintain mitochondrial stability, and inhibit liver cell apoptosis (39). Liang et al. (40) found that exosomes released by MSCs can regulate immune response and promote liver self-repair during ischemia-reperfusion. In addition, Xie et al. (41) demonstrated that exosomes derived from huc-MSCs transport miR-1246 through the Wnt/β-catenin pathway mediated by GSK3β, inhibiting hepatocyte apoptosis and reducing inflammation, and thereby reducing hepatic IRI.

Treatment of Hepatic IRI with Hepatocyte-Derived Exosomes

In the IRI model, Nojima et al. (30) showed that only exosomes derived from primary hepatocytes could induce dose-dependent hepatocyte proliferation in vivo and in vitro. The primary mechanism involves the direct fusion of hepatocyte exosomes with target cells, transportation of neutral ceramidase and sphingosine kinase 2 (SK2), and increased levels of sphingosine-1-phosphate (S1P) in target cells to promote the proliferation of liver cells (30).

Treatment of Hepatic IRI with Exosomes Derived from Dendritic Cells

In the study by Zheng et al. (42), exosomes produced by bone marrow-derived dendritic cells (DCs) mainly regulated the PI3K/mTOR axis through Hsp70, maintaining a balance between Treg differentiation and Th17 cells, thereby alleviating IRI.

EXOSOMAL TREATMENT OF LIVER FIBROSIS

Fibrosis is a wound-healing reaction in which the damaged area is wrapped by an extracellular matrix (ECM) or scar. Almost all patients with chronic liver injury have fibrosis, but the rate of fibrosis occurrence is variable (43–46). The occurrence of liver fibrosis usually requires continuous damage over several months to years. Liver fibrosis can be reversed in the initial stage, whereas progressive liver fibrosis can lead to liver cirrhosis. However, the exact time point of the development of irreversible liver fibrosis is still unclear. Nevertheless, increasing evidence shows that early hepatic cirrhosis may be reversible (47–49). Therefore, treatment of liver fibrosis and early cirrhosis is crucial. Recent studies have investigated the utility of exosomes in the treatment of liver fibrosis.

Treatment of Liver Fibrosis with Exosomes Derived from MSCs

Rong et al. (50) used CCL₄ to induce hepatic fibrosis in rats. By administering exosomes derived from human bone marrow mesenchymal cells (hBM-MSCs-Ex) in vivo, they found that hBM-MSCs-Ex can effectively alleviate liver fibrosis. Furthermore, they confirmed that the mechanism involves inhibition of the Wnt/β-catenin pathway that reduces the expression of α-smooth muscle actin (SMA) and collagen I. These exosomes reduce collagen accumulation, inhibit inflammation, and increase liver cell regeneration, thereby improving liver function and reducing liver fibrosis. In the in vivo experiments conducted by Li et al. (51), huc-MSCs-Ex transplantation was shown to significantly protect liver function and reduce liver fibrosis, which primarily manifested as a significant recovery of serum AST activity. In addition, the expression of ALT and TGF-β1 decreased, and the phosphorylation of Smad2 decreased. At the same time, huc-MSCs-Ex can also improve CCL₄-induced liver fibrosis by inhibiting epithelial-to-mesenchymal transition (EMT). Moreover, in the CCL4-induced mouse liver fibrosis model, exosomes from ADSCs modified by miR-181-5p were shown to inhibit the STAT3/Bcl-2/Beclin 1 pathway of HST cells to increase autophagy and reduce the liver fibrosis induced by TGF-β1 (52). Similarly, the exosomes derived from ADSCs modified by mmu_circ_0000623 can also prevent liver fibrosis by activating autophagy (53).

Exosomes derived from amniotic MSCs have been shown to significantly reduce the activation of Kupffer cells and hepatic stellate cells in the liver of nonalcoholic fatty liver rat models, resulting in decreased inflammation, hepatic fibrosis, and fiber formation (54). Furthermore, Povero et al. (55) confirmed that exosomes derived from human induced pluripotent stem cells could also inhibit liver fibrosis through the mechanisms mentioned earlier.

Treatment of Liver Fibrosis with Exosomes Derived from Myeloid Cells

Myeloid cell-specific IL-6 can induce macrophages and neutrophils, release mir-223-enriched exosomes, transfer to hepatocytes to inhibit liver fibrosis, and provide a target for treating nonalcoholic fatty liver (56). Wang et al. (57) found that exosomes derived from natural killer (NK) cells (NK-92MI) can inhibit the proliferation and activation of hepatic stellate cells induced by TGF-β1 and reduce CCL4-induced liver fibrosis. In addition, NK cells can induce hepatic stellate cell apoptosis, thereby mediating antifibrotic effects (58, 59).

Treatment of Liver Fibrosis with Exosomes Derived from Hepatocytes

Exosomes derived from human liver stem cells (HLSCs) can improve liver fibrosis and inflammation by recoding liver gene expression. In a mouse model of nonalcoholic fatty liver, treatment with exosomes derived from HLSCs significantly reduced the percentage of liver fibrosis and significantly increased the expression of the anti-inflammatory, immune-regulatory cytokine IL-10 (60). In addition, another study (61) reported that exosomes derived from HLSCs can increase the expression of miR-146a-5p in a human hepatic stellate cell line (LX-2). miR-146a-5p is an antifibrotic RNA that can weaken the activation phenotype of hepatic stellate cells and reduce liver fibrosis. The miR-146a-5p-regulated fibrosis-related pathways identified to date include Wnt/β-catenin (62), TGF-β/SMAD (63, 64), PTPRA-SRC (65), and LPS/TLR4/NF-κB (66–68).

EXOSOMAL TREATMENT OF ACUTE LIVER FAILURE

Acute liver failure (ALF) is characterized by the development of severe acute liver injury with encephalopathy and impaired synthetic function [international normalized ratio (INR) ≥ 1.5] in a patient without cirrhosis or preexisting liver disease (69, 70). It is also referred to as fulminant hepatic failure, acute hepatic necrosis, fulminant hepatic necrosis, or fulminant hepatitis. The prognosis of untreated ALF is poor (71). With acute care clinical management, over 55% of patients survive without the need for liver transplantation (72). The management for ALF includes treatment of the underlying causes and complications, including metabolic abnormalities, hepatic encephalopathy, and cerebral edema. For patients who show deterioration, liver transplantation offers the best chance for recovery and survival. In the United States and Europe, patients with ALF requiring transplantation are given high priority on the transplantation list. However, a lack of available donor livers remains the biggest challenge in the field of liver transplantation (73). Attempts have been made to develop artificial hepatic assist devices (74–77), but the results in patients using these systems have been disappointing, and they are not widely available (78, 79).

Exosomes may be sensitive markers to predict early DILI, which remains an unmet need (80–82). They may also play a role in treatment of ALF, and further studies are required to determine if they can serve as a promising treatment approach.

Treatment of ALF with Exosomes Derived from MSCs

In a mouse model induced by lipopolysaccharide and d-galactosamine (LPS/GalN), adipose-derived mesenchymal stem cell (AMSC)-derived exosomes colocalized with liver macrophages after intravenous administration, and they inhibited the activation of inflammasomes to reduce the secretion of inflammatory factors. In addition, the abundant exosomal shuttle miR-17 inhibits NLRP3 inflammasome activation by targeting TXNIP, thereby reducing serum ALT and AST levels (83). A recent study constructed quercetin- and vitamin A-loaded AMSC-derived exosomes to treat ALF in mice, showing that quercetin enhanced the therapeutic efficacy and vitamin A enhanced the liver targeting of exosomes. The loaded exosomes could reduce the rapid senescence-like response induced by ALF (84).

Exosomes derived from human umbilical cord mesenchymal stem cells (huc-MSCs) pretreated by TNF-α could reduce serum ALT, AST, and proinflammatory cytokine levels and inhibit the activation of NLRP3 inflammation-associated pathway proteins in the ALF mouse model (85). Moreover, another study showed that hucMSC-derived exosomes could downregulate the expression of NLRP3, caspase-1, IL-1β, and IL-6 in the ALF mouse model, which reduced ALT and AST levels (86). In addition, ERK1/2 and IGF-1R/PI3K/AKT signaling pathways were also reported to be involved when hucMSC exosomes inhibited oxidative stress-induced apoptosis both in vitro and in vivo (87).

BM-MSC-derived exosomes increased the expression of the autophagy marker proteins LC3 and Beclin-1 and decreased the expression of the proapoptotic proteins Bax and cleaved caspase-3, promoting autophagosome formation and reducing hepatocyte apoptosis in the ALF mouse model (88). In addition, these exosomes could also confer cytoprotective effects, reduce the oxidative stress generated due to injury conditions in vitro, and improve liver regeneration and recovery from liver injury in rats (89).

In an ALF rat model induced by bile duct ligation (BDL), exosomes from placenta-derived mesenchymal stem cells (PD-MSCs) were involved in liver regeneration. The C-reactive protein (CRP) in exosomes excreted by PD-MSCs regulated the Wnt pathway and vascularization in BDL rat hepatocytes by interacting with endothelial cells (90).

Exosomes from human menstrual blood-derived stem cells provide the advantage of being easy to collect and isolate. In the ALF mouse model, MSC exosomes markedly improved liver function, enhanced survival rates, and inhibited apoptosis of liver cells at 6 h after administration (91).

Treatment of ALF with Exosomal Nanoparticles Derived from Plants

Dietary exosome-like nanoparticles (ELNs) have been also hypothesized to be involved in disease mechanisms, representing a promising class of agents with the potential to ameliorate ALF. An animal experiment found that shiitake mushroom-derived ELNs (S-ELNs) could substantially inhibit NLRP3 inflammasome activation by preventing inflammasome formation in primary macrophages. In addition, S-ELNs also suppressed the secretion of IL-6 and both protein and mRNA levels of the IL-1b gene (92).

EXOSOMAL TREATMENT OF HEPATOCELLULAR CARCINOMA

Hepatocellular carcinoma (HCC) is still a global health challenge, and its incidence is increasing worldwide (93–95). Risk factors for HCC include hepatitis B virus infection (96, 97), hepatitis C virus infection (98–100), fatty liver (101, 102), alcoholic cirrhosis (103), smoking, diabetes, obesity (104–106), and various dietary exposures. Transarterial chemoembolization (TACE) combined with sorafenib is currently used for treating advanced liver cancer, but the overall results obtained with this regimen are modest. Liver transplantation is offered for stage II HCC at most transplant centers, but donor liver organ availability is a rate-limiting factor for the application of this technique.

Therefore, there is an urgent need to find new treatments for liver cancer.

Treatment of HCC with Exosomes Derived from Hepatocytes/Tumor Cells

Rao et al. (107) demonstrated that exosomes derived from HCC cells could be used as broad-spectrum antigen carriers to induce DC-mediated tumor rejection in mouse models of ectopic and orthotopic liver cancer. Yang et al. (108) reported that a tumor-derived exosome (TEX) combined with CpG could be used to promote specific antitumor immunity by increasing CD4+ and CD8+ T cell tumor infiltration. Cheng et al. (109) treated exosomes derived from liver cancer with melatonin and found that the treated exosomes can attenuate the tumor-induced immunosuppressive state of macrophages through the STAT3 pathway. A subsequent study (110) demonstrated that p120-catenin (p120ctn), a membrane protein abnormally expressed in many solid tumors, plays a critical role in regulating the progression of HCC. The expression of p120ctn is downregulated in liver cancer tissues, and p120ctn derived from tumor cells inhibits liver cancer cell invasion, migration, and proliferation by inhibiting STAT3 signal transduction. Cho et al. (111) found that exosomes derived from DILI mice can promote the cytotoxicity of primary hepatocytes or hepatocellular carcinoma cells, induce receptor cell apoptosis, and promote liver damage in animals.

Treatment of HCCs with Exosomes Derived from MSCs

Most primary liver cancers are at an advanced stage by the time they are diagnosed, and they show a poor prognosis and limited response to chemotherapy (112). However, one study demonstrated that exosomes (122-Exo) secreted by AMSCs modified by miR-122 could effectively improve the chemotherapy sensitivity of HCC (113). Factors such as ADAM10, IGF1R, and CCNG1 affect tumor development and drug sensitivity and participate in the antitumor activity of sorafenib in vivo (114, 115). In vitro experiments have shown that when combined with sorafenib, 122-Exo can improve the chemotherapy sensitivity of HCC cells by downregulating the expression of target genes inducing cell apoptosis and cell cycle arrest (116). However, in vivo experiments have confirmed that 122-Exo alone cannot inhibit tumor growth. This may be because the effect of 122-Exo is based on the presence of a sorafenib growth-arresting effect on HCC.

Treatment of HCC with Exosomes Derived from Dendritic Cells

Lu et al. (117) injected the exosomes of α-fetoprotein (AFP)-expressing dendritic cells (DEXAFP) into an HCC mouse model and found that DEXAFP induced a solid antigen-specific immune response. The number of CD8+ T lymphocytes expressing interferon (IFN) increased significantly at the tumor site, and the levels of IFN and IL-2 in mice also increased significantly. Conversely, the levels of CD25 + Foxp3+ Tregs, IL-10, and transforming growth factor-β (TGF-β) at the tumor site reduced. Thus, DEXAFP significantly limited tumor growth in a murine liver cancer model and prolonged survival. In this study, the choice of the AFP antigen was crucial, but one limitation was that AFP is only expressed by ∼50% of human HCCs (118).

Conclusions

Exosomes have significant potential in treating liver disease and have also been shown to be useful as adjuvants. For example, exosomes used as an adjuvant with recombinant hepatitis B antigen enhanced the Th1 response to hepatitis B antigen (119). DNA vectors generating engineered exosomes have been shown to serve as hepatitis B vaccine candidates by activating cytotoxic T lymphocytes (120, 121). In addition, exosomes can also participate in drug delivery, increase delivery efficiency, and treat diseases (122, 123).

Interestingly, one study found that in a variety of mouse models of pancreatic cancer, genetically engineered exosomes (iExosomes) targeted oncogenic KRAS, inhibited tumor growth in mouse models, and significantly improved the overall survival rate (124). Although exosomes show promise for the treatment of liver diseases, much research is needed to overcome the current obstacles, including reproducibility, lack of separation methods, low extraction efficiency, and standard detection methods. Furthermore, distinction of the various extracellular vesicle subgroups and the functional differences among the subgroups is challenging (125). Therefore, future research should focus on the application of exosomes in clinical practice and defining scientific and practical separation methods as well as detection methods to provide new strategies for treating clinical liver diseases.

GRANTS

This study is supported by the Natural Science Foundation of Shanghai Science and Technology Commission Grant 18ZR1400100, the Natural Science Foundation of China Grant 82072647, the Natural Science Foundation of China Grant 81970565, the Basic Research projects of Shanghai Municipal Science and Technology Commission Grant 18JC1414000), the Shanghai Pujiang Program Grant 18PJ1409300, the Basic Research projects of Shanghai Municipal Science and Technology Commission Grant 18JC1414000, the National Institutes of Health (NIH) LTCDS N276201200017C (to D.A.G.), and NIH NIDDK Pittsburgh Liver Research Center P30 Grants 1P30DK120531-01 (to D.A.G.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Y.P., W.T., and M.Y. conceived and designed research; W.T. and M.Y. prepared figures; Y.P. and W.T. drafted manuscript; Y.P. edited and revised manuscript; Y.P., W.T., M.Y., J.L., and D.A.G. approved final version of manuscript.