Statistics from Altmetric.com
Liver fibrosis is a leading cause of morbidity and mortality worldwide, and the prevalence of chronic liver disease is expected to increase in parallel with obesity and type 2 diabetes mellitus epidemics.1 As there is currently no Food and Drug Administration (FDA)-approved drug treatment for liver fibrosis, a liver transplant is the only curative treatment for a minority of individuals who develop cirrhosis-associated complications such as liver cancer and liver failure. In the face of an acute organ shortage, there is a growing impetus to better understand the precise cellular and molecular mechanisms of adult liver repair and regeneration, so that clinically relevant therapeutic targets can be identified.
In adults, after an acute liver injury, ‘normal’ regeneration occurs through the proliferation and expansion of remaining healthy hepatocytes and cholangiocytes, and there is minimal involvement of liver progenitors. By contrast, repair during chronic liver injury involves the proliferation of progenitor cells and cells with certain stem-like properties to differentiate into both hepatocytes and cholangiocytes, and thus support the restoration of liver function.2 3 What are the identity and/or origin of these liver progenitors? Multiple progenitor markers have been proposed, but none are completely specific as recent studies demonstrate that liver progenitors are actually a heterogeneous population that contains a variety of cell types, ranging from more primitive progenitors to more differentiated cells (eg, hepatocyte-like). Indeed, in adult mice, progenitors include biliary-like cells (also known as ‘oval cells’ in rodents) which arise in the ductal region, as well as progenitors around the central vein, de-differentiated hepatocytes or cholangiocytes, as well as multipotent mesenchymal stem cells (MSC) which also exhibit immunomodulatory properties.3 4 Similar heterogeneity has now been described in humans: EpCam+/NCAM+bi-potent progenitors reside the ductal plate of the human fetal liver and are subsequently retained in the Canals of Herring of the adult liver, and these latter cells co-express genes associated with both hepatic and biliary lineage (eg, SOX9, HNF1B, K19, CD24, CD133, K7 and OPN).4 During the injury, both mature human hepatocytes and cholangiocytes may also ‘transdifferentiate’ to give rise to these hybrid bi-potent progenitors, thus resulting in a substantial overlap in the ‘marker’ expression between various liver cell types. The consequential lack of antibody specificity and limited resolution of immuno-labelling and fate mapping strategies could explain in part, the apparent contradictory reports of the identities and origins of adult liver cells. The advent of single-cell transcriptomics would hopefully allow us to more accurately define these progenitor populations5; for example, although hepatic stellate cells (HSCs) have hitherto been considered the major source of myofibroblasts (MFs) (which secrete collagen) during liver fibrosis, recent single-cell RNA-sequencing studies in a rodent model of liver fibrosis suggest that MFs are actually a heterogeneous population and could potentially corroborate previous reports that HSCs co-express markers of MSC and that HSC-derived MFs are multipotent progenitors which contribute to liver regeneration.6 These novel approaches could also help elucidate the roles of endothelial and bone-marrow-derived progenitors in fibrosis and repair.
In Gut, Dai and colleagues have extended our understanding of ‘progenitor-associated repair’ by elegantly showing that an (artificially induced) expanded liver progenitor cell population could ameliorate liver fibrosis in two murine models.7 Specifically, they showed that HSC-derived growth differentiation factor (GDF) 11 directly increased the number of leucine-rich repeat-containing G-protein coupled receptor 5 (Lgr5)+progenitors, which co-expressed other stem cell markers such as CD133 and Epcam, as well as the cholangiocyte marker Keratin19, and the hepatocyte marker hepatocyte nuclear factor (HNF) 4A, and thus resembled bi-potent liver progenitors recently described by others.4 They further noted that the addition of Lgr5+progenitors to MFs significantly repressed fibrogenesis in vitro and reduced liver fibrosis in vivo when Lgr5+progenitors were injected into CCl4 and DDC-treated mice and that recombinant GDF11-treated or GDF11-overexpressing Lgr5+progenitors further inhibited fibrogenesis. Conversely, in vivo depletion of the Lgr5+progenitor population-enhanced liver fibrosis (figure 1).
While compelling, these data throw up some important questions, namely what are the underlying mechanisms by which Lgr5+progenitors reduce liver fibrosis? Furthermore, the ductular reaction observed during chronic human liver injury (also known as the liver progenitor response in mouse) is predictive and correlates with liver fibrosis, and Coombes et al had previously reported that Sox9+OPN+K19+liver progenitors could secrete OPN that enhanced HSC activation into MF.8 Consistent with these, Machado and Diehl had also demonstrated in a series of reports that both liver progenitors and HSC–MF could secrete hedgehog ligands to autoregulate and amplify the fibrogenic phenotype.9 How could we reconcile these apparent differences in liver outcomes? One obvious explanation would be that the progenitor populations used in these studies are distinct from one another and Dai and colleagues had alluded to the possibility that ‘other facultative stem cells’ could be playing an antifibrotic role7 (figure 1).
Another factor that dictates liver outcomes is the microenvironment milieu, that is, what other cellular and molecular signals are present or absent during injury and repair, as individual progenitor subsets may be differentially regulated. The transforming growth factor beta (TGF-β) superfamily of proteins plays a critical role in the development and wound healing, and GDF11 is one such member. GDF11 has previously been reported to regulate progenitor cell growth in the developing retina, pancreas, and endothelium, and in this study by Dai et al, forced overexpression of GDF11 protected against liver fibrosis.7 However, members of the TGF-β family often exhibit apparent dichotomy in functionality. For example, bone morphogenic protein (BMP)-9 is downregulated during earlier stages of liver injury and is upregulated at later stages, during the re-differentiation of tissue, and the inhibition of BMP-9 in models of fibrosis ameliorated disease progression.10 Another study further suggested that BMP-9 induced liver fibrosis in part, by negatively regulating the oval cell response.11 In contrast, studies using another strain of mice (129/Ola) showed that BMP-9 inhibited the capillarisation of liver sinusoidal endothelial cells.12 Outside of the TGF-β superfamily, OPN exhibits similar context-dependent, Janus-like effects. For example, OPN knockout mice developed more fibrosis after chronic CCl4, but less fibrosis in dietary-induced nonalcoholic steatohepatitis (NASH) and the biological functions of OPN may also differ in early versus late disease.8 It is therefore, conceivable that forced overexpression of GDF11 in HSC–MF (that overwhelms OPN, hedgehog, BMPs and other profibrogenic ligands) leads to an expansion of a specific progenitor cell population that is more ‘restorative’ (ie, reduces fibrosis). Because liver progenitors secrete an array of chemokines (including CXCL16, MCP1, CCL17 and CCL20) that recruit immune cells to orchestrate a repair-associated inflammatory response,8 9 it is also possible that a GDF11-(over)enriched microenvironment could stimulate progenitor cell secretion of chemokines which preferentially recruit and retain scar-resolving immune subsets (eg, Ly6C int-low macrophages) (figure 1).
In conclusion, harnessing the potential of liver progenitors to treat liver fibrosis is an exciting and promising strategy, but a lot of work remains. In addition to identifying various liver progenitor subsets with greater precision, we will also need to better understand how cellular and molecular signals integrate under homeostatic and diseased conditions to regulate repair and regeneration. This is particularly relevant to humans as chronic liver disease processes are generally protracted and the role of liver progenitors in human chronic liver disease remains unclear.
Contributors Both KB and W-KS contributed equally to this manuscript.
Funding KB was supported by funding from the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation—project no.:393225014; 394046768-SFB1366).
Competing interests None declared.
Patient consent for publication Not required.
Provenance and peer review Commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.