Combatting Fibrosis: Exosome-Based Therapies in the Regression of Liver Fibrosis.

Hepatic fibrosis results from chronic injury and inflammation in the liver and leads to cirrhosis, liver failure, and portal hypertension. Understanding the molecular mechanisms underlying hepatic fibrosis has advanced the prospect of developing therapies for regression of the disease. Resolution of fibrosis requires a reduction of proinflammatory and fibrogenic cytokines, a decrease in extracellular matrix (ECM) protein production, an increase in collagenase activity, and finally, a disappearance of activated myofibroblasts. Exosomes are nanovesicles of endocytic origin secreted by most cell types. They epigenetically reprogram and alter the phenotype of their recipient cells and hold great promise for the reversal of fibrosis. Recent studies have shown that exosomes function as conduits for intercellular transfer and contain all the necessary components to induce resolution of fibrosis, including the ability to (1) inhibit macrophage activation and cytokine secretion, (2) remodel ECM production and decrease fibrous scars, and (3) inactivate hepatic stellate cells, a major myofibroblast population. Here, we discuss the research involving the regression of hepatic fibrosis. We focus on the newly discovered roles of exosomes during fibrogenesis and as a therapy for fibrosis reversal. We also emphasize the novel discoveries of exosome-based antifibrotic treatments in vitro and in vivo.

α‐smooth muscle actin

connective tissue growth factor

clusters of differentiation

collagen α1(I)

dendritic cell

extracellular matrix

glucose‐regulated protein 78

hepatic stellate cell

heat shock protein

interleukin

lysyl oxidase‐like 2

microRNA

matrix metalloproteinase

messenger RNA

mesenchymal stem cell

membrane‐type 1

mitochondria peroxisomal membrane protein 70 kDa

natural killer

pathogen‐associated molecular pattern

platelet‐derived growth factor

portal fibroblast

peroxisomal membrane protein 70 kDa

signal transducer and activator of transcription

transforming growth factor β

tissue inhibitors of metalloproteinase

tumor necrosis factor α

tumor necrosis factor‐related apoptosis‐inducing ligand

Hepatic fibrosis is caused by the excessive production and accumulation of insoluble collagen and extracellular matrix (ECM) components following sustained chronic injury in the liver. Various chronic liver diseases, such as hepatitis B virus, hepatitis C virus, alcoholic liver disease, and nonalcoholic steatohepatitis (NASH), result in fibrosis.1 If the death rate from cirrhosis continues to increase as projected, cirrhosis will become the twelfth leading cause of death by 2020.2

Essential mechanisms have been identified for the circuitous nature of the pathogenesis and resolution of hepatic fibrosis due to chronic liver disease. Transforming growth factor β (TGF‐β) is a central regulator in chronic liver disease; it contributes to all stages of disease progression from initial liver injury through inflammation and fibrosis.3 Liver damage‐induced levels of active TGF‐β stimulate an increase in expression levels of many growth factors and cytokines involved in fibrogenesis, including platelet‐derived growth factor (PDGF), connective tissue growth factor (CCN2), interleukins (ILs [IL‐1α, IL‐β, and IL‐6]), and tumor necrosis factor α (TNF‐α).4, 5, 6, 7, 8 Increased levels of active TGF‐β enhance hepatocyte destruction and mediate hepatic stellate cell (HSC) and fibroblast activation, resulting in a wound‐healing response that includes myofibroblast generation and ECM deposition.3 Overexpression of CCN2 in concert with signaling pathways associated with development of liver fibrosing injury can lead to the initiation or exacerbation of fibrosis.8 IL‐1β exerts a stimulatory effect on the synthesis of ECMs,9 IL‐6 induces hepatic inflammation and collagen synthesis,10 and TNF‐α is required for cholestasis‐induced liver fibrosis.11

A key event during liver fibrosis is the activation of myofibroblasts, which originate from fibroblasts, including HSCs, portal fibroblasts (PFs), and fibrocytes. Depending on the ECM composition, fibroblasts maintain quiescence or activate into myofibroblasts.12 Due to chronic insult, fibroblasts subjected to extracellular stress caused by abnormal ECM (e.g., fibronectin, collagen type I and III) proliferate and obtain a myofibroblast‐like phenotype. Activated myofibroblasts secrete ECM and form stress fiber‐induced cell‐matrix junctions, which further facilitate ECM remodeling. Excessive ECM deposition and significant changes in topographic distribution of ECM components increase expression of tissue inhibitors of metalloproteinases (TIMPs).13 Following resolution of the injury, liver fibrosis can be reversed after the withdrawal of the underlying cause of disease. This is associated with a significant reduction of myofibroblasts due to apoptosis, induction of senescence and killing apoptosis of senescent HSCs by natural killer (NK) cells, or phenotypic reversion to the quiescent‐like phenotype. Meanwhile, a reduction in collagen production as well as decreased TIMP‐1 expression and an increase in hepatic collagenase and elastase activity result in ECM degradation and remodeling.14, 15, 16, 17, 18

Over the past 3 decades, the drive to discover the mechanisms underlying the critical events during fibrogenesis has been fundamentally relevant to the development of antifibrotic strategies. However, even with this push to understand the mechanisms of disease progression, there are currently no antifibrotic treatments. Therefore, the need to continue to uncover the mechanisms of fibrogenesis and discover potential targets for treatment is essential in drug development.

Exosomes are cell‐derived vesicles that are present in eukaryotic fluids. They are either released directly from the plasma membrane or from the cell when multivesicular bodies fuse with the plasma membrane. They contain proteins and other molecules that reflect the transcriptional and/or translational activity of the cell of origin.19 The differential contents of RNAs, proteins, lipids, and metabolites in exosomes are distinct to the cell type of origin. Following their release into the intercellular space, exosomes bind to recipient cells and deliver their informative cargo. The recipient cells may then undergo epigenetic reprogramming and subsequent phenotypic alterations according to the molecular information received. The presence of specific components (protein, microRNA [miRNA or miR], or messenger RNA [mRNA]) in different types of exosomes results in different functional properties in the recipient cells. For instance, lipotoxic fatty acid‐injured hepatocytes produce exosome‐like vesicles, which are then taken up by HSCs, leading to fibrogenic activation.20, 21 Additionally, the use of human mesenchymal stem cell (MSC)‐derived exosomes allows an MSC‐like therapeutic payload to be delivered to the liver, followed by a subsequent reduction of liver fibrosis, thus protecting hepatocytes.22 Further, fibrogenic signaling in HSCs is suppressed by exosomes shuttled between quiescent and activated HSCs.23, 24, 25

This review covers some of the most important functions of different exosomes during liver fibrogenesis and the regression of liver fibrosis. We emphasize both the established mechanism of regression of liver fibrosis and the new developments in novel exosome‐based antifibrotic strategies. We also highlight the emerging consensus about rodent models of fibrosis regression, which demonstrate a return of normal or near‐normal liver histology and function.

Exosomes in Liver Fibrosis

The Pathogenesis of Epithelial Injury

Recurrent epithelial injury is a prominent driving factor in the pathogenesis of progressive fibrosis26 and results in hepatocyte dysfunction, which can occur through apoptosis.26, 27 Hepatocytes, in response to the hostile environment, undergo apoptosis through an extrinsic death receptor‐mediated pathway, or alternatively, intracellular stress can activate the intrinsic pathway of apoptosis. Both pathways target the mitochondria, and mitochondrial dysfunction is a prerequisite for hepatocyte apoptosis.28

Hepatocytes can produce exosomes that contain caveolae (Caveolin‐1), early endosome (Eaa1), endoplasmic reticulum (glucose‐regulated protein 78 [Grp78]), peroxisome (peroxisomal membrane protein 70 kDa [Pmp70]), or mitochondria (Prohibitin1 and mtPmp70),29 and these exosomes are able to communicate with other hepatocytes or other cell types throughout the body. Interestingly, the proteins that are found in hepatic‐derived exosomes have been demonstrated to play a role in metabolizing lipoproteins, endogenous compounds, and xenobiotics. Further, exosomes derived from injured hepatocytes are enriched with cytochrome P450s that serve important roles in the cellular detoxification of endogenous toxic substances.29 Cytochrome P450 2E1 (CYP2E1) generates reactive oxygen species that can produce superoxide anion radicals, hydrogen peroxide, and powerful oxidants, such as the hydroxyl radical, in the presence of iron catalysts. Elevated levels of CYP2E1 under a variety of pathophysiologic conditions lead to hepatic apoptosis through mechanisms of oxidative stress.30 Therefore, it is speculated that injured hepatocyte‐derived exosomes containing P450s participate in the development of steatosis, increased fibronectin expression, and hepatocyte apoptosis.

In response to lipid injury, hepatocytes release exosome‐like vesicles containing tumor necrosis factor‐related apoptosis‐inducing ligand (TRAIL) and clusters of differentiation (CD)40 ligand, inducing the production of inflammatory‐type macrophages.31 Hepatocytes injured by lipotoxic fatty acids produce exosome‐like vesicles enriched in miR17‐92 clusters, which are taken up by HSCs, leading to fibrogenic activation.20, 21

The exosomes derived from CCl4‐treated hepatocytes include diverse types of self‐RNAs and recognize an activator of toll‐like receptor 3, which increases the production of IL‐17A production in hepatic γδ T cells. The increased levels of proinflammatory cytokines are tightly associated with HSC activation.32 In agreement, Il‐17‐produced T cells regulate production of TGF‐β1 in Kupffer cells and can directly activate collagen type I production by HSCs, the major source of fibrogenic myofibroblasts in fibrotic liver.33

Exosomes released from epithelial cells carry information that can activate fibroblasts and initiate and perpetuate fibrosis. For example, injured epithelial cells produce increased numbers of exosomes containing information sufficient to activate fibroblasts. When released by injured epithelial cells, these exosomes are taken up by neighboring fibroblasts, resulting in their increased production of α‐smooth muscle actin (α‐SMA) and type I collagen, thus driving liver fibrosis.34

Excessive Deposition of ecm During Development of Liver Fibrosis

The ECM represents a noncellular component in the liver that is mainly composed of proteins and proteoglycans.35 It forms an intricate network that provides a physical scaffold for cellular support while allowing unimpeded transport of solutes and growth factors.36 The ECM undergoes continuous remodeling, particularly during injury and wound healing.37 In response to chronic liver injury, the secretion of types of ECM proteins alters dramatically, resulting in abundant production of type I and III collagens, increased deposition of fibronectin and proteoglycans, as well as other subtypes of collagens.36 During this process, lysyl oxidase‐like 2 (LOXL2) facilitates crosslinking of collagens and elastin by catalyzing oxidative deamination of lysine residues.38 As a consequence, tissue stiffness is also increased. This heavily crosslinked collagen network replaces normal tissue structure and results in a change in the phenotype of normal resident cells as well as pathologic myofibroblasts.

Exosomes play an important role in ECM crosslinking, thus affecting processes such as angiogenesis, fibroblast activation, and premetastatic niche formation.39 LOXL2 has been detected on the exterior of endothelial cell‐derived exosomes, placing it in the direct vicinity of the ECM. Increased LOXL2 levels in both endothelial cells and endothelial cell‐derived exosomes enhance the activity of collagen gel contraction. However, knockdown of LOXL2 in exosome‐producing endothelial cells in both normal and hypoxic conditions reduces exosome activity.40 Thus, ECM crosslinking by endothelial cell‐derived exosomes is mediated by LOXL2.

The Origin of Myofibroblasts in Fibrotic Liver

Hepatic fibrosis is accompanied by the accumulation of increased numbers of myofibroblasts in the liver.41, 42 These myofibroblasts are the source of ECM components necessary for building the fibrous scar tissue surrounding the wound. The production of these stress fibers makes myofibroblasts highly contractible and mobile, enabling their migration throughout the injured tissue and their further secretion of ECM components. Therefore, activation of myofibroblasts is a key mechanism in the development of liver fibrosis.

The origin of myofibroblasts has been well studied. HSCs are the major source of myofibroblasts.1 Hepatic myofibroblasts also originate from PFs and fibrocytes.43 Despite earlier studies,44, 45 recent reports have demonstrated that myofibroblasts do not originate from epithelial cells undergoing an epithelial‐to‐mesenchymal transition.46, 47

HSCs

HSCs are perisinusoidal cells that reside between the hepatocytes and small blood vessels in the liver. They are characterized by the presence of numerous retinoid and lipid droplets48, 49 where they store vitamin A.50 A critical feature of the wound‐healing response during liver injury is the differentiation of HSCs from a “quiescent” state in the normal liver to an “activated” state in the injured liver.51 This transition is characterized by both morphologic and functional changes, including down‐regulation of vitamin A expression; production of α‐SMA, which confers contractility and promotes wound closure; and ECM synthesis.51, 52

TGF‐β is a potent cytokine that activates HSCs into myofibroblasts followed by increased expression of α‐SMA, PDGF, CCN2, type I collagen, and TIMP1,53, 54, 55 all of which result in a wound‐healing response, including myofibroblast generation and ECM deposition.3 CCN2, a fibrogenic molecule synthesized downstream of TGF‐β, is tightly associated with fibrogenic pathways in activated HSCs.56 It has recently been found that activated HSC‐derived exosomes contain CCN2 or CCN2 mRNA, each of which increases in concentration during HSC activation and amplified fibrogenic signaling.57 The induction of CCN2 expression in activated HSCs is due to decreased expression of miR‐214, which otherwise inhibits CCN2 expression by directly binding to the CCN2 3′‐untranslated region.23, 57 Further, miR‐214 can be exported from HSCs through exosomes to neighboring cells, leading to regulation of miR‐214 target genes.23 The dynamic expression of miR‐214 in HSCs is the result of its transcriptional regulation by Twist‐1, which is also exosomally transferred between HSCs where it maintains its ability to induce miR‐214 in recipient cells.25 Thus, a Twist1–miR‐214–CCN2 axis is exosomally shuttled to activate additional HSCs in which fibrogenic signaling is then modulated.

Fibrocytes

Although HSCs are believed to be a major source of myofibroblasts (which produce collagen type I in the fibrotic liver), bone marrow‐derived fibrocytes, as defined by their simultaneous expression of CD45 and collagen type I, are also a potential source of myofibroblasts and are implicated in the pathogenesis of liver fibrosis.58 Fibrocytes also express CD34 and major histocompatibility complex II and secrete TGF‐β, promoting the deposition of ECM.59, 60

Exosomes released from fibrocytes have a concentration‐dependent proangiogenic activity. The recipient cells of fibrocyte‐derived exosomes demonstrated a dose‐dependent increase in the expression of collagen α1(I) [Colα1(I)] and α‐SMA.61 Heat shock protein (HSP)‐90a and total activated signal transducer and activator of transcription 3 (STAT3) are important components of the fibrocyte exosome cargo. Fibrocyte‐derived exosomes are also enriched with miR‐21, miR‐142a, miR‐125b, miR‐126, miR‐130a, and miR‐132, all of which work in tandem to modulate collagen production, resulting in enhanced deposition of mature collagen fibrils in the wound and promotion of wound contraction at an early stage in the wound‐healing process.62

PFs

PFs normally comprise a small population of the fibroblastic cells that surround the portal vein to maintain integrity of the portal tract. They are associated with the pathogenesis of cholestatic liver injury.63 It has been demonstrated that PFs are a major source of myofibroblasts in cholestatic liver injury, contributing to greater than 70% of myofibroblasts at the onset of injury (5 days after bile duct ligation).64 PFs respond rapidly to TGF‐β1, as demonstrated by up‐regulation of Colα1(I), α‐SMA, TIMP1, TGF‐β2, plasminogen activator inhibitor 1, elastin, fibronectin, and CD73 ecto‐enzyme.65, 66, 67, 68 However, unlike HSCs, PFs respond to stimulation with taurocholic acid and IL‐25, leading to an induction of Colα1(I) and IL‐13, respectively.64

Although PFs play a critical role in the pathogenesis of cholestatic liver fibrosis, functional properties of PFs and the mechanism by which PFs contribute to cholestatic fibrosis are not well understood. Additionally, exosomes originated from PFs have not yet been reported.

Macrophages and Immune Cells

Chronic inflammation and fibrosis are inextricably linked through the interactions among immune cells.69, 70, 71, 72 Macrophages play an important role in inflammation and subsequent fibrogenesis.73, 74, 75, 76, 77 To promote fibrosis, macrophages produce specific matrix metalloproteinases (MMPs), such as MMP9, that degrade the basement membrane and allow inflammatory cells and recruited fibroblasts to enter sites of injury. They secrete a variety of profibrotic mediators, including TGF‐β1, PDGF, and many chemokines, that recruit and activate inflammatory cells.78, 79 Macrophages are also tightly associated with collagen‐producing myofibroblasts in vivo and produce cytokines and growth factors that modulate myofibroblast activity. Various types of macrophages, such as M1 (inflammatory), M2a‐like (profibrotic), and Mregulatory/M2c‐like (regulatory),73, 75, 76, 77, 80, 81 are recruited during fibrogenesis, resulting in re‐epithelization, healing, or pathologic scarring. In contrast, macrophages also play a distinct role in the resolution of fibrosis. Macrophages may activate additional stem cell and local progenitor cell populations that participate in repair; thus, macrophages that exhibit an anti‐inflammatory phenotype become the dominant population.16 These macrophages respond to IL‐10 and other inhibitory mediators and secrete a variety of anti‐inflammatory mediators, such as IL‐10 and TGF‐β1, that play major roles in suppressing the immune system and quieting the inflammation.82, 83, 84, 85 Macrophages can also induce myofibroblast apoptosis, remove cellular debris, and stimulate the production of collagen‐degrading MMPs in myofibroblasts.86 Therefore, different phenotypes of macrophages play unique and crucial roles at different stages of tissue repair.

Macrophages can reportedly release exosomes, which contain pathogen‐associated molecular patterns (PAMPs), that lead to the activation of naive recipient immune cells.87 Moreover, these macrophage exosomes can be actively endocytosed into placenta tissue and drive cytokine release.87 Thus, the macrophage‐immune cell exosome pathway represents a novel non–cell‐associated mechanism of antigen transfer between immune cells; this can exert varying effects on naive cells. The uptake of macrophage‐derived exosomes into neighboring immune cells could then result in immunomodulation and alteration of subsequent inflammatory stimuli. However, functional properties of macrophage exosomes in the resolution of fibrosis have not been reported, although it is clear that macrophages exhibit an important role in the mechanisms of fibrosis regression.

NK cells and NKT cells provide the initial defense, invading infectious microbes and neoplastic cells.88 Dendritic cells (DCs) are central to the processes that modulate liver immunity,89 whereas regulation of T cells mediates immune tolerance.90 The role of B cells in the pathogenesis of fibrosis was identified by the reduction in collagen deposition observed in CCl4‐induced fibrosis in B‐cell‐deficient mice.72 It has been shown that DC‐derived exosome‐like vesicles can enhance the antigen‐specific responses of CD4+ and CD8+ T cells and participate in the activation of NK cells.91 Exosomes from IL‐10‐treated DCs suppressed inflammation and collagen‐induced arthritis in mice.92 In addition, miRNAs released from T‐cell exosomes are transferred into DCs in an antigen‐specific manner.93 However, the mechanisms of immune cell‐derived exosomes during liver fibrogenesis are still under investigation. The actions and roles of exosomes in liver fibrosis are summarized in Figure 1 and Table 1.

Figure 1

Exosomes regulate cell functions. 1, Hepatocytes produce exosomes enriched in Caveolin‐1, early endosome (Eaa‐1), endoplasmic reticulum (Grp78), peroxisome (Pmp70), and mitochondria (Prohibitin 1 and mtPmp70), participating in hepatocyte metabolism. 2, Injured hepatocytes enriched in cytochrome P450s promote hepatocyte steatosis and apoptosis. 3, Lipid‐induced injury of hepatocytes enriched in TRAIL and CD40 ligand promote activation of macrophages and HSCs. 4, Injured hepatocytes by lipotoxic fatty acids produce exosomes enriched in miR17‐92 clusters, promoting HSC activation. 5, Endothelial cells release exosomes enriched in LOXL2, enhancing the activity of collagen contraction. 6, Fibrocytes release exosomes enriched in HSP‐90a, activated STAT3, and miRs (21, 142a, 125b, 126, 130a, and 132), participating in ECM remodeling. 7, Activated macrophages produce exosomes enriched in PAMPs, leading to the activation of naive recipient immune cells. 8, miR155‐ and miR125b‐enriched exosomes promote differentiation of M1 macrophages over M2 macrophages. 9, Exosomes enriched with MT1‐MMP, IDE, heparanase, integrins, and LOXL2 lead to collagen cleavage and degradation. 10, HSCs release exosome‐enriched Twist1 and miR214/199‐5a clusters, reducing CCN2 expression in activated HSCs. Abbreviations: IDE, insulin‐degrading enzyme; MT1‐MMP, membrane‐type 1 MMP.

Table 1

Exosomes Regulate Cell Functions

Donor Cells	Exosomal Cargo	Target Cells	Outcomes
Normal hepatocytes36	Caveolin‐1, Eaa‐1, Grp78, Pmp70, and mitochondria (Prohibitin 1 and mtPmp70)	Hepatocytes	Participate in hepatocyte metabolism
Injured hepatocytes37	Cytochrome P450s37	Hepatocytes	Promote hepatocyte steatosis and apoptosis
Steatotic hepatocytes38	TRAIL and CD40 ligand	Macrophages	Promote activation of macrophages
Lipotoxic fatty acid‐induced injury of hepatocytes27, 28	miR17‐92 clusters	HSC	Promote HSC activation
Endothelial cell44, 45	LOXL2	ECM	Enhance activity of collagen contraction
Fibrocytes67	HSP‐90a, activated STAT3, and miRs (21, 142a, 125b, 126, 130a, and 132)	ECM	Participate in ECM remodeling
Activated macrophages84	PAMPs	Immune cells	Lead to activation of naive recipient immune cells
Activated HSC58	CCN2	HSC	HSC activation
Normal HSC30, 31, 32	Twist‐1 and miR214/miR199‐5a	Activated HSC	Reduce CCN2 expression in activated HSC

Exosomes in the Regression of Hepatic Fibrosis

Hepatic fibrosis was considered to be irreversible.94 However, numerous studies have demonstrated that regression of hepatic fibrosis is possible and is dependent on an increase in collagenase activity, a disappearance of activated myofibroblasts, and a suppression of proinflammatory and fibrogenic cytokines, subsequently resulting in a decrease in ECM production and increased ECM degradation.

The emerging field of exosome biology has identified several novel pathways of exosome‐dependent intercellular transfer of biologically active materials that not only facilitate the development of liver fibrosis but can also initiate fibrosis resolution. Exosomes from healthy subjects can transport biologically active antifibrotic molecules, including proteins and nucleic acids, that in turn regulate gene expression and cellular function in target cells. For example, DC‐derived exosomes from mice subjected to immunosuppressive treatments or modified to express immunosuppressive cytokines promoted tolerogenic immune responses, leading to amelioration of inflammatory responses in mice,95 and mRNA‐155‐ and miRNA‐125b‐enriched exosomes promoted differentiation of M1 macrophages over M2 macrophages.96 (Figure 1) Human amnion epithelial cell‐derived exosomes significantly reduced the number of macrophages and macrophage infiltration during liver fibrosis.97

In addition, exosomes from healthy subjects also contain a variety of molecules capable of interacting with and altering ECM components, including enzymes, such as membrane‐type 1 (MT1) MMP, insulin‐degrading enzyme, and heparanase, as well as integrins and LOXL2.39 These enzymes could potentially localize on the surface of exosomes and through their contact with molecules in the ECM and could lead to cleavage of a wide range of substrates, such as collagen, a step necessary for collagen degradation. For example, MT1‐MMP derived from exosomes has been demonstrated to target the ECM and degrade fibrillar collagen (type I, II, and III) as well as other matrix components, including fibronectin and vitronectin, promoting cell migration.98 On the other hand, fibronectin‐enriched exosomes interact with integrins to promote adhesion formation of cells.39 Thus, a dynamic mechanism of forming adhesion interactions and breaking interactions by exosomal enzymes and components of the ECM is involved in wound healing and inflammation. Another enzyme, heparanase, is expressed on the surface of exosomes and has been shown to degrade heparan sulfate within the ECM and to participate in an inflammatory response.99 Further, exosomes from hypoxia‐induced endothelial cells have increased collagen crosslinking activity in the ECM through up‐regulation of LOXL2, whereas knockdown of LOXL2 in endothelial‐derived exosomes in both normal and hypoxic conditions reduced activity of exosomes.40 (Figure 1 & Table 1) Therefore, exosomes have been implicated in the regulation of both inflammation and ECM remodeling.

Studies to elucidate the signaling molecules that contribute to the activated HSC phenotype have identified potential therapeutic targets for antifibrotic therapy.56, 100 One potential target is CCN2, a profibrotic factor that is produced in fibrosing liver tissue. CCN2, a cysteine‐rich matricellular protein, interacts with integrins, low‐density lipoprotein receptor‐related proteins, and heparan sulfate proteoglycan coreceptors,101, 102 thus stimulating adhesion, migration, proliferation, survival, and differentiation of HSCs. CCN2 exhibits strong profibrogenic properties. Overexpression of CCN2 promotes ECM deposition and development of fibrotic lesions. Hepatic levels of CCN2 correlate with the severity of liver disease in patients with liver fibrosis.103, 104 Additionally, overexpression of CCN2 mediates TGF‐β1‐dependent fibrotic pathways in HSCs,105, 106 and TGF‐β1 mRNA transported by injured epithelial‐derived exosomes results in a rapid initiation of activation of myofibroblasts.34 In the exosome‐mediated transfer of activated HSC‐derived exosomes to quiescent HSCs, CCN2 was directly targeted through modulation of a Twist–miR‐214/199 axis, resulting in HSC activation.23, 24, 25

Quiescent HSCs produce exosomes that inhibit activation of HSCs and attenuate pathways of fibrogenesis.23, 24, 25 This results in an exosomal transfer of miR‐214, miR‐199a‐5p, or Twist‐1 into the recipient HSCs and directly inhibits transcription of CCN2, thus suppressing downstream collagen production and reverting HSCs to a more quiescent phenotype (Figure 1 & Table 1). Hepatocytes have also been demonstrated to produce exosomes that cause a reversal of fibrosis‐associated gene expression and ethanol‐induced damage in hepatocytes.107 Hepatocyte‐derived exosomes can bind to activated HSCs or injured hepatocytes through mechanisms that involve heparin‐like molecules and cellular integrin subunits αv or β1, thus mediating therapeutic changes.107 Therefore, hepatic fibrosis is amendable to therapy. Exosomes produced either from quiescent HSCs or normal hepatocytes may be important for a reduction in the progression of fibrosis and therefore have serious potential as antifibrotic therapies.

The Role of Exosomes as a Biomarker of Liver Fibrosis

Clinically, patient management decisions depend on the accurate assessment of the severity and progression of liver fibrosis. Liver biopsy is the “gold standard” and is invasive, expensive, and risky to patients. Because the components of exosomes are a “fingerprint” of the dynamic status of the underlying pathologic condition in patients, they might represent a new biomarker for identifying and assessing molecular signatures associated with liver fibrosis. In addition, exosomal components are protected from proteinase‐dependent degradation and thus can be stably detected in the circulating plasma and serum, making them ideal biomarkers for a number of clinical applications.108, 109 Increased levels of CD10 protein in urinary exosomes from glycine N‐methyltransferase knockout mice have been associated with steatosis, fibrosis, and hepatocellular carcinoma.110 CD81‐enriched serum exosomes of patients with chronic HCV was associated with inflammation and severity of fibrosis.111 Decreased levels of miRNAs (miR‐34c, miR‐151‐3p, miR‐483‐5p, or miR‐532‐5p) were detected in serum exosomes of CCl4‐induced mice or human patients with F3/4 fibrosis.97

MSC‐Derived Exosomes or Other Exosomes as a New Therapeutic Strategy for Experimental Fibrosis Models

The transfer of MSCs has been proposed as a potential therapeutic strategy for the treatment of various diseases and immune disorders, mostly due to their immunoregulatory properties.112 For example, MSCs secrete several antifibrotic molecules, such as hepatocyte growth factor, fibroblast growth factor, epidermal growth factor, insulin, and dexamethasone. They were also reported to mediate cytoprotective, anti‐angiogenic, and regenerative effects in damaged liver.113 Although the exact mechanism remains unknown, MSCs were demonstrated to attenuate liver fibrosis by suppressing activation of T helper 17‐positive immune cells in fibrotic liver.114 A similar effect can be achieved using adoptive transfer of MSC‐derived exosomes to mice with liver fibrosis.115, 116

Exosomes are emerging as effective therapeutic tools for different diseases because these particles can bypass biological barriers and can serve as powerful drug and gene therapy transporters, raising the exciting prospect of “cell therapy without the cells.”117, 118, 119 The administration of MSC‐derived exosomes is a potential strategy for treating liver disease.22, 120, 121, 122 The safety and feasibility observed in early clinical trials using MSCs has resulted in increased interest in the translation of the use of these cells to the clinic.123, 124 Likewise, increasing evidence suggests that MSC‐derived therapeutic effects are mainly mediated in a paracrine manner by extracellular vehicles, such as exosomes.125, 126 In this regard, use of MSC‐derived exosomes will allow the delivery of anti‐inflammatory cytokines and other biologically active proteins to injured livers without administration of heterologous divergent cells. For example, by using a CCl4‐induced liver injury model in Kunming mice, delivery of human umbilical cord MSC‐derived exosomes reduced hepatic fibrosis through inhibition of collagen production.122 Furthermore, the delivered exosomes migrated to the liver, resulting in a significant suppression of the TGF‐β1/Smad pathway and subsequent down‐regulation of collagen type I/III and TGF‐β1 in these mice.22 Moreover, it has been demonstrated that exosomes released from adipose tissue‐derived MSCs inhibited the proliferation and activation of the human LX‐2 cell line as well as primary HSCs from male Sprague Dawley rats.121

In addition to MSC‐derived exosomes, exosomes isolated from serum have also been shown to have a therapeutic effect in mice with liver fibrosis. For instance, hepatic fibrosis was decreased in CCl4‐injured or thioacetic acid‐injured mice treated with exosomes derived from the serum of healthy mice (but not from fibrotic mice)127; mice showed improved liver function, reduced apoptosis of hepatocytes, suppression of an inflammatory response in the injured liver, reduced release of hepatic or circulating proinflammatory cytokines (IL‐2, IL‐4, IL‐5, IL‐6, IL‐10, IL‐12p70, interferon‐gamma, and TNF‐α), reduced inflammatory infiltration, and reduced circulating aspartate aminotransferase/alanine aminotransferase levels.127

Overall, adoptive transfer of exosomes from normal healthy individuals may be beneficial for patients with liver fibrosis. The main mechanism by which “healthy exosomes” support the repair of injured livers may be the release of paracrine factors.128 These exosomes shuttle across the intercellular space and deliver “therapeutic molecules” between different liver cells; this results in the elevation of collagenase activity, the abortion of activated myofibroblasts, and the reduction in proinflammatory and fibrogenic cytokines, which finally results in a decrease in ECM production and regression of liver fibrosis.

Summary

Decades of basic and translational research have brought us to a new era where promising therapies for liver inflammation and fibrosis are being developed. Hepatic fibrosis has been shown to be reversible in patients with chronic liver disease and in experimental models, as demonstrated by a reduction of proinflammatory and fibrogenic cytokines, remodeling of ECM production and decreased fibrous scar, and the inactivation or disappearance of myofibroblast populations. Because exosomes function as conduits for intercellular transfer and have been demonstrated to be sufficient to inhibit liver fibrosis in rodent models, the administration of exosomes as a therapy for hepatic fibrosis in humans holds great promise.

Exosomes can remain hidden in the bloodstream, carry multiple doses, specifically target the cells, and store and administer treatment, and their small size allows them to cross barriers that cells cannot. However, there are some open questions remaining that limit the application of exosome therapy: (1) How should the the characterization and quantification of exosomes be standardized? More effective methods and techniques for large‐scale exosome production are needed; (2) What is the proper source of exosomes for therapy? Careful analysis of the complex information in cell‐derived exosomes or circulating exosomes may result in the identification of unique “molecular signatures” for exosome therapy; (3) How should dosing of exosomes be evaluated? Exosome‐regulated signaling pathways are dose dependent.129 Therefore, the tuning of exosome dose may enable the balancing of potential deleterious and therapeutic effects of exosome administration.

Potential conflict of interest

Nothing to report.