SEARCH

SEARCH BY CITATION

Keywords:

  • differentiation;
  • hepatic progenitor cells (HPCs);
  • HPCs niche;
  • liver regeneration

Abstract

  1. Top of page
  2. Abstract
  3. Background
  4. Hepatocyte-mediated liver regeneration
  5. Hepatic progenitor cell-mediated liver regeneration
  6. Immunophenotype of HPCs
  7. Mechanisms of hepatic progenitor-cell activation
  8. Differentiation of hepatic progenitor cells and the role of HPC niche
  9. Contribution of bone marrow cells to the liver regenerating epithelial compartment
  10. Concluding remarks
  11. Acknowledgements
  12. References

When there is a massive loss of hepatocytes and/or an inhibition in the proliferative capacity of the mature hepatocytes, activation of a dormant cell population of resident hepatic progenitor cells (HPCs) occurs. Depending on the type of liver damage HPCs generate new hepatocytes and biliary cells to repopulate the liver placing them as potential candidates for cell therapy in human liver failure. Liver injury specific mechanisms through which HPCs differentiate towards mature epithelial cell types are recently become understood. Such new insights will enable us not only to direct HPCs behaviour for therapeutic purposes, but also to develop clinically feasible methods for in vivo differentiation of other stem cell types towards functional hepatocytes. This review aimed to provide the current improved knowledge of the role of HPCs niche and its signals in directing the behaviour and fate of HPCs and to translate this basic knowledge of HPCs activation/differentiation into its clinical applications.


Background

  1. Top of page
  2. Abstract
  3. Background
  4. Hepatocyte-mediated liver regeneration
  5. Hepatic progenitor cell-mediated liver regeneration
  6. Immunophenotype of HPCs
  7. Mechanisms of hepatic progenitor-cell activation
  8. Differentiation of hepatic progenitor cells and the role of HPC niche
  9. Contribution of bone marrow cells to the liver regenerating epithelial compartment
  10. Concluding remarks
  11. Acknowledgements
  12. References

With respect to the pivotal role of the liver in body homeostasis, it is not surprising that severe liver diseases are often fetal. A wide variety of viral, toxic, metabolic, immunological and genetic insults can affect the liver leading to development of steatohepatitis, fibrosis, cirrhosis, cancer and finally death. Liver disease is one of the most important causes of death world-wide. Currently, the only curative option for end-stage liver disease is liver transplantation. However, there is a growing mismatch between the number of patients who needs liver transplantation and the number of available livers, leading to the death of many patients on waiting list for transplantation. Furthermore, liver transplantation requires lifelong immunosuppression which results in long-term side effects and complications including renal, lymphoproliferative and cardiovascular diseases [1-3]. This highlights the urgent need to develop alternative or new therapeutic strategies for the treatment of end-stage liver diseases. Hepatocyte transplantation has been assessed in many clinical studies, but their long-term efficacy remains unclear and more importantly the shortage of hepatocytes is a major limitation for using this method [4]. Therefore, the shortage of donor livers has prompted the search for the other cell types that can be used for this purpose. It seems that stem/progenitor-cell transplantation is a more promising alternative approach in this setting [5-7]. Our improved knowledge on the understanding of stem cell plasticity has led to the concept of ‘regenerative medicine’. Potential therapeutic effects of different types of extra hepatic stem cells such as fetal/embryonic stem cells, induced pluripotent stem cells, multipotent adult progenitor cells, mesenchymal stem cells and bone marrow stem cells (BMSCs) in liver diseases have been investigated by many researchers and are out of the scope of this review. Resident hepatic progenitor cells (HPCs), capable of differentiating towards both hepatocytes and cholangiocytes, are another potential suitable source for liver cell replacement. Mechanisms underlying bidirectional differentiation capability of HPCs are poorly understood, however, recent advances in biotechnology allowing in vitro and in vivo manipulation of these cells and their niche, have provided important clues. This review aimed to provide the current improved knowledge of the role of HPCs niche and its signals in controlling the fate of HPCs and to make a link between current advances in basic knowledge of HPCs activation/differentiation and its possible clinical applications.

Hepatocyte-mediated liver regeneration

  1. Top of page
  2. Abstract
  3. Background
  4. Hepatocyte-mediated liver regeneration
  5. Hepatic progenitor cell-mediated liver regeneration
  6. Immunophenotype of HPCs
  7. Mechanisms of hepatic progenitor-cell activation
  8. Differentiation of hepatic progenitor cells and the role of HPC niche
  9. Contribution of bone marrow cells to the liver regenerating epithelial compartment
  10. Concluding remarks
  11. Acknowledgements
  12. References

The liver is placed in a toxic rich-environment; consequently, it is required to tolerate frequent exposure to toxins. Furthermore, the liver can be affected by a wide variety of toxins, infections, tumours and disorders like viral diseases, genetic or immunological disorders. Therefore, the liver has evolutionary adapted to deal with such insults [8]. Indeed, the liver has an amazing capacity to regenerate after different types of injury. Liver regeneration is an extremely complex process that requires activation of multiple complex pathways, which do not act independent of each other and involves, based on the type and severity of the injury, activation/proliferation of cells of different lineages [9-11]. Hepatocyte-mediated liver regeneration, which occurs during acute mild to moderate liver injury, is the quickest and the most effective way to generate new hepatocytes. In this model of liver regeneration, the constant divisions of the remaining healthy mature hepatocytes will account for liver repopulation. The adult normal liver is usually in a non replicating state. Quiescent hepatocytes are in a stage known as G0, which indicates that the cells are not cycling. After acute mild to moderate hepatic injury, hepatocytes enter the cell cycle (G1 phase), progress to DNA replication (S phase), and then undergo mitosis (M phase) with cell division completing this process [9]. This highly regulated process is simultaneously mediated by different growth factors and cytokines like tumour necrosis factor (TNF)-a, interleukin (IL)-6, which prime hepatocytes to respond to growth factors like epidermal growth factor (EGF) and hepatocyte growth factor (HGF). Other key factors which are released from different glands include insulin, norepinepherine and triodothronine. Joint signals from aforementioned factors result in hepatocytes cell cycle progression which in turn leads to DNA synthesis and ultimately proliferation of hepatocytes [8].

Hepatic progenitor cell-mediated liver regeneration

  1. Top of page
  2. Abstract
  3. Background
  4. Hepatocyte-mediated liver regeneration
  5. Hepatic progenitor cell-mediated liver regeneration
  6. Immunophenotype of HPCs
  7. Mechanisms of hepatic progenitor-cell activation
  8. Differentiation of hepatic progenitor cells and the role of HPC niche
  9. Contribution of bone marrow cells to the liver regenerating epithelial compartment
  10. Concluding remarks
  11. Acknowledgements
  12. References

When there is a massive loss of hepatocytes (e.g. in acute liver failure, characterized by massive necrosis of hepatocytes), and/or an inhibition in proliferative capability of mature hepatocytes (e.g. in cirrhosis caused by diverse aetiologies), activation of a dormant cell population of resident progenitor cells, histologically observed as ductular reactions (DR), occurs [12-14]. Accordingly, proliferation of these adult progenitor cells has been identified in a wide variety of human liver diseases, in which a certain degree of hepatocyte loss and damage with impaired regeneration of remaining hepatocytes and/or bile duct epithelial cells exist [7, 8, 15]. Of interest that a more recent work from Diehl's group has found that liver regeneration also depends upon fibroblast growth factor-inducible 14 (Fn14)+ progenitors after partial hepatectomy (PH), which is believed to be a classic model of hepatocyte-mediated liver regeneration. The authors showed that loss of Fn14 signalling impaired proper liver regeneration and led to liver failure after PH in mice [16].

Sequential histological observations of the native liver in a patient with acute liver failure and 95% loss if hepatocytes who underwent auxiliary partial orthotopic liver transplantation (APOLT) showed full regeneration of the liver from resident HPCs/ductular hepatocytes [17]. These observations were also reported and confirmed by other groups in which the full regeneration of the native liver led to the removal of the graft after stopping immunosuppressive therapy [18-21]. Taken together, aforementioned studies have proved that in the setting of acute massive loss of hepatocytes, HPCs are capable of producing new full functional hepatocytes if positioned in the appropriate microenvironment.

In human cirrhosis, indirect evidence of a link between HPCs of DR and regenerative nodules, using mitochondrial DNA mutations as clonal markers for tracing cell lineages, has been shown [22]. However, a precursor-offspring relationship cannot be established by this approach. Direct evidences indicating a lineage link between HPCs of DR and mature hepatocytes, although being scarce, exist [23]. Yoon et al., in their studies on liver biopsy of patients with hepatitis B induced cirrhosis, have shown gradual telomere shortening from DR to epithelial cell adhesion molecule (EPCAM)+ hepatocytes and to EpCAM(−) hepatocytes indicating that EPCAM positive hepatocytes are derived from slow cycling cells of the ductular reactions [23].

Apart from HPCs, there has been a great debate over the potential contributions of mesenchymal cells types (like hepatic stellate cells (HSCs)/fibroblast) in regenerating epithelial cell compartment of the liver through mesenchymal to epithelial transition (MET) and vice versa (EMT) [24-30]. Potential contribution of alpha-smooth muscle actin (αSMA)-expressing cells to liver epithelial cells have been shown in animal models [24]. However, it is questionable if α-SMA, whose expression in chronic fibrotic liver diseases is often found in myofibroblasts of different origins, is a suitable lineage marker. It has been shown that epithelial cells can express intermediate filaments like α-SMA in the presence of tissue damage; therefore, expression of mesenchymal markers may reflect epithelial injury [25]. Even in the defining EMT this marker is widely used to show cell function of myofibroblasts [26], while this marker cannot denote functionality, because it has no role in the deposition of ECM. A few, but not all, lineage tracing studies in which alpha-smooth muscle actin (α-SMA) or glial fibrillary acidic protein (GFAP) have been used to fate trace HSCs/fibroblast have also shown contribution from these cells up to 24% of newly formed hepatocytes [27, 28]. Consideration should be given to the fact that GFAP is not a specific mesenchymal marker and also marks biliary epithelial cells limiting the interpretation of these studies. Employing novel fate tracing method and using more than one mesenchymal marker concurrently can at least, partly overcome these apparent discrepant results. A more recent study, utilized a novel Lecithin-retinol acyltransferase (Lrat)Cre-transgenic mouse that marked 99% of HSCs [29]. The authors showed that in a wide range of toxic, biliary and fatty liver injuries, HSCs do not function as epithelial progenitor cells [29]. To test whether HSCs give rise to epithelial cells in adult liver, a further study determined the hepatic lineages of HSCs and portal fibroblasts (PFs) using MesP1Cre and Rosa26mTmGflox mice. Genetic cell lineage tracing revealed that the MesP1 + mesoderm gives rise to HSCs and PFs, but not to hepatocytes or cholangiocytes, in the adult liver. Upon carbon tetrachloride injection or bile duct ligation surgery-mediated liver injury, HSCs and PFs differentiate into myofibroblasts but not into hepatocytes or cholangiocytes. Furthermore, differentiation of the mesodermal mesenchymal cells into progenitor (oval) cells was not observed [30].

Immunophenotype of HPCs

  1. Top of page
  2. Abstract
  3. Background
  4. Hepatocyte-mediated liver regeneration
  5. Hepatic progenitor cell-mediated liver regeneration
  6. Immunophenotype of HPCs
  7. Mechanisms of hepatic progenitor-cell activation
  8. Differentiation of hepatic progenitor cells and the role of HPC niche
  9. Contribution of bone marrow cells to the liver regenerating epithelial compartment
  10. Concluding remarks
  11. Acknowledgements
  12. References

Hepatic progenitor cells (oval cells in rodents), in the non-diseased human liver are located in the bile ductules and the canals of Hering that are localized in the portal tract and the periportal parenchyma (Fig. 1) [14]. Since HPCs are hardly recognizable on routine histochemical staining such as Haematoxylin and Eosin, immunohistochemistry or electron microscopy is required to detect these cells. HPCs have distinct morphology; they are small epithelial cells (about 9 μm) with a large oval shape nucleus and scanty cytoplasm which is immunoreactive for neural cell adhesion molecule (NCAM), prominin1 (CD133), telomerase and EPCAM. HPCs also express markers of both biliary epithelial cells (K7, K19 and K14) and hepatocytes lineages (K8, K18, C-met and albumin) and are able to differentiate into both cholangiocytes and hepatocytes, depending on the epithelial cell type (hepatocyte versus bile duct epithelial cell) which is damaged the most [31]. Proliferating activated HPCs form duct like structures which emanate from the portal aria and expand into the parenchyma in the close proximity with hepatic stellate cells and macrophages as their niche cells. Hepatocytic differentiation of these migrating cells leads to the formation of intermediate hepatocytes (Fig. 1), which are polygonal cells with a size and phenotype intermediate between those of progenitor cells and mature hepatocytes [21, 32]. Intermediate hepatocytes in very early stages of differentiation are positive for K19 and K7 in a sub membranous pattern, but they lose immunoreactivity for K19 much earlier in comparison with that of K7. Finally the full maturation of intermediate hepatocytes into hepatocytes is characterized by disappearance of these biliary markers. Cholangiocytic differentiation of HPCs can occur through the formation of immature (small) cholangiocytes. Libbrecht et al. found patchy or diffuse NCAM positivity in interlobular and septal bile ducts in patients with both biliary and parenchymal cirrhosis with a higher proportion of NCAM-positive bile ducts in biliary cirrhosis in comparison with that of parenchymal cirrhosis [33]. In addition, some of the NCAM-positive biliary cells were smaller than the surrounding NCAM-negative biliary cells in the same bile duct. This observations together with the fact that in the liver NCAM is a marker of immature biliary cells (that can differentiate towards biliary lineage) [34], suggest that immature biliary cells may contribute to the repair of damaged bile ducts in chronic liver diseases. By this way HPCs represent a dynamic cell compartment that constantly changes its morphology and its expressed markers, depending on the type of the injury and its consequent differentiation stage. Accordingly, there is not a unique HPCs marker specific for these cells and this is in fact the differentiation stage of these cells which determines the expression pattern of HPCs markers.

image

Figure 1. Canals of Hering (CoH) The canal of Hering, in the periportal area, is a physiological link between hepatocyte-canalicular system and biliary tree. Hepatic progenitor cells are located in the bile ductules and the canals of Hering that are localized in the portal tract and the periportal parenchyma. Lower part of the figure shows small keratin 7-positive hepatic progenitor cells in the periportal area of a biopsy section from a patient with toxic liver damage (A) and keratin 7-positive (in a sub membranous pattern) intermediate hepatocytes (arrows) in a patient with severe liver impairment (B). PT, portal tract; HA, hepatic artery; PV, portal vein; BD, bile duct

Download figure to PowerPoint

Mechanisms of hepatic progenitor-cell activation

  1. Top of page
  2. Abstract
  3. Background
  4. Hepatocyte-mediated liver regeneration
  5. Hepatic progenitor cell-mediated liver regeneration
  6. Immunophenotype of HPCs
  7. Mechanisms of hepatic progenitor-cell activation
  8. Differentiation of hepatic progenitor cells and the role of HPC niche
  9. Contribution of bone marrow cells to the liver regenerating epithelial compartment
  10. Concluding remarks
  11. Acknowledgements
  12. References

The mechanisms controlling the behaviour of human HPCs are partially understood. Most of our current knowledge in this field has gained from animal studies which suggested a pivotal role for inflammatory cytokines and chemokines in initiation the regenerative response [35, 36]. In both acute and chronic human liver diseases, the extent of HPCs activation is correlated with the severity of liver disease [37, 38]. It has also been shown that the severity and localization of the inflammatory infiltrate in chronic viral hepatitis is correlated with the activation and localization of HPCs [37]. These findings suggest that the inflammatory infiltrate may provide paracrine signals from growth and chemotactic factors to initiate the regenerative response. Potentially important elements of the inflammatory response that may stimulate oval cells include IL6 family [leukaemia inhibitory factor (LIF) and oncostatin M (OSM)], lymphotoxin-β, interferon gamma (IFNγ) and TNF superfamily including TNFα and TNF-like weak inducer of apoptosis (TWEAK) [9, 39-41]. Chemokine stromal-cell-derived factor 1 (SDF-1), has also been shown to be upregulated in a wide range of human chronic liver diseases [42]. It binds to its receptor, CXCR4, and plays a variety of roles including recruitment of BM cells to the damaged liver [42, 43]. CXCR4 is expressed by a variety of cells including BMSCs and inflammatory cells and SDF-1 attracts CXCR4 + inflammatory cells, and these cells express TWEAK [43, 44]. TWEAK has recently joined the rank of key HPCs regulators. Despite the other members of TNF family, such as TNFα and lymphotoxin, which play important roles in both HPCs- and hepatocyte-mediated liver regeneration, TWEAK has demonstrated effects only on progenitor-cell compartment and not on the mature hepatocytes [41, 45]. It is produced mainly by monocytes and macrophages and stimulates HPCs proliferation directly through its specific receptor Fn14 [45]. Significant consideration should be given to this issue that TWEAK receptor is also expressed by a subset of myofibroblasts/activated HSCs [45] which suggest a potential role for TWEAK in co-regulation of HPC activation and liver fibrosis. Using recombinant TWEAK to stimulate oval cell expansion, a recent study claimed a progenitor (oval) cell-driven fibrogenic response in fibrotic mice undergoing PH [46]. However, it is possible that TWEAK has parallel effect on myofibroblasts in driving fibrosis [47]. Another possibility is the interactions between lymphotoxin-β producing HPCs and lymphotoxin-β receptor expressing HSCs which can indirectly mediate fibrogenesis through expression of RANTES and ICAM by HSCs [48].

The involvement of Wnt, Notch and Hedgehog (Hh) signalling pathways in HPCs activation in humans has also been reported recently [49-51]. Hh ligands that bind to their receptor Patched (PTC) on oval cells/HPCs are important in the survival of these cells [52]. It has also recently been shown that in patients with alcoholic steatohepatitis and primary biliary cirrhosis, HPCs and hepatic stellate cells (HSC) as a part of HPC niche are capable of both producing and responding to Hh ligands [49, 52]. Taken together, these data suggest that like stem cell niches of the other organs such as brain and intestine [53], in the liver the regulation of HPCs response is multifactorial and this is in fact the complex interplay of aforementioned factors and other yet unidentified factors, produced by HPCs niche, which governs the behaviour of HPCs.

Differentiation of hepatic progenitor cells and the role of HPC niche

  1. Top of page
  2. Abstract
  3. Background
  4. Hepatocyte-mediated liver regeneration
  5. Hepatic progenitor cell-mediated liver regeneration
  6. Immunophenotype of HPCs
  7. Mechanisms of hepatic progenitor-cell activation
  8. Differentiation of hepatic progenitor cells and the role of HPC niche
  9. Contribution of bone marrow cells to the liver regenerating epithelial compartment
  10. Concluding remarks
  11. Acknowledgements
  12. References

The key signals in differentiation of adult HPCs towards a specific epithelial cell type are not well understood in situ. It seems that the interaction between different components of HPCs niche is the key element in regulating the self-renewal/activation/maintenance and differentiation ability of the HPCs. In addition, special microenvironment created in response to a specific liver damage can determine fate decision and differentiation pathways of HPCs. Accordingly, the ability of HPCs to participate in liver regeneration could be associated with the type and the severity of liver disease [38, 50, 51, 53]. These differences are reflected in the expression patterns of many growth factors and regulatory molecules in different type of human liver diseases [50, 51]. In our previous work, activated HPCs and their niche were captured in acute hepatocytic (fulminant hepatitis B and drug induced liver failure), chronic hepatocytic (hepatitis C induced cirrhosis) and chronic cholangiocytic [primary biliary cirrhosis (PBC)] liver diseases by laser microdissection and their gene-expression profiles were investigated by a customized PCR array and immunohistochemistry/-fluorescence [50]. In contrast to hepatocytic diseases, HPCs in cholangiocytic diseases (PBC) showed a high expression of Jag1 and predominant nuclear localization of Notch intracellular domain (NICD) suggesting the importance of Notch pathway in biliary differentiation of HPCs. On the other hand, higher Numb (Notch inhibitor) encoding mRNA was measured in hepatocytic diseases, which indicates that Notch pathway is inhibited in hepatocytic diseases. Although this study revealed an important role for Notch pathway in determining of HPCs fate, neither the contributing cell types nor the mechanisms of their contribution (direct or indirect paracrine signals) to the observed different outcomes in different type of liver diseases were not investigated. Furthermore, a functional role for Notch inhibition in hepatocellular differentiation cannot be established by this study. Recent studies have shed more light on the mechanism by which HPCs differentiate into a specific cell lineage in vivo. Lorenzini et al. have found co-localization of HPCs with laminin, endothelial cells, myofibroblasts and macrophages in patients with chronic hepatitis B and C [54], however, the exact role of these progenitor niche cells were not investigated in this study, leave the question of how niche cell interactions determine the fate and behaviour of HPCs open. Instead, they studied the functional consequences of the laminin rich basement membrane on HPCs behaviour. Culturing oval cells on various matrices they demonstrated that laminin allows maintenance of oval cells in a progenitor/biliary phenotype, and inhibits hepatocyte differentiation. These results were also confirmed by the other groups [55]. Lineage- tracing studies have clearly proved the aforementioned concept. Indeed, during oval cell induced liver regeneration in mice, inhibition of laminin signalling with iloprost, a synthetic analogue of prostacyclin PGI2, resulted in differentiation of oval cells into hepatocytes [56].

To further investigate the functional role of HPCs niche and interactions between niche componenets/cells in delineation of bipotential HPCs fate, Boulter et al. have recently compared accompanying features of HPCs activation/DR in both biliary and hepatocytic human and mice liver injury [51]. Using mouse models of both biliary and hepatocyte regeneration, they digitally reconstructed the activated HPCs niche in three dimensions. They observed that during biliary regeneration, the activated HPCs were surrounded by many α-SMA–positive myofibroblasts which formed close associations with both the emerging HPCs and collagen I. Depositing of a thick layer of collagen around HPCs/DR, excluded macrophages to form close associations with the HPCs. During hepatocyte regeneration, however, the composition of HPCs niche was different, consisted predominantly of macrophages, with fewer myofibroblasts and reduced collagen I. The latter may allow macrophages to associate closely with HPCs. Then the authors went on to find how difference in HPCs niche in aforementioned two types of liver damage can regulate their behaviour. They found high expression of Notch1 and Notch2 receptors in both human disease types; however, their localizations were different being nuclear in biliary type [50]. Furthermore, mRNA expression of Jag1 ligand and Notch targets genes were higher in both human and mouse biliary type diseases compared with those of hepatocellular type. These finding clearly support the previously mentioned role of Notch signalling in driving biliary differentiation. To investigate the previously unknown functional role of Notch inhibition, they first blocked the Notch receptor in vivo, which resulted in biliary differentiation inhibition. If we next consider equal expression of Notch receptors in both types of biliary and hepatocellular injury, this pathway should be somehow inhibited during hepatocytic injury. Indeed an elevated expression of Notch inhibitor Numb was observed in HCV samples compared with PBC and PSC samples as it had been reported previously [50]. In addition, induction of biliary regeneration in mouse resulted in a rapid loss of Numb at both mRNA and protein levels and further confirmed the role of Notch inhibition in hepatocytic differentiation. The authors then continued to investigate the role of Wnt 3a expressing macrophages being more prominent in hepatocellular type of liver injury. Incubating bone marrow derived macrophages with hepatocyte debris induced macrophages to express Wnt3a, which, in turn, stimulated neighbouring HPCs to express Notch pathway inhibitor Numb. By this way, these Numb expressing HPCs turn down Jag1 expressing myofibroblasts resulting in biliary differentiation inhibition. The role of macrophages in HPCs differentiation towards hepatocytes was further confirmed by the authors' observation that ablation of macrophages by liposomal clodronate during hepatocyte regeneration blocked differentiation of HPCs towards hepatocytes; in its place, it stimulated cholangiocytic differentiation of HPCs. The aforementioned results of macrophage ablation have also been shown in other studies in which liver macrophages have been depleted by the same agent [57].

Taken together, these data emphasize the importance of cross talk between local tissue damage and its associated inflammatory niche observed in a special type of liver disease with other extra- and intracellular signalling indicators that direct HPCs behaviour. Understanding such interplay at the HPC niche can result in the development of new therapies aimed to stimulate liver regeneration in different kind of human liver diseases in a controlled manner.

Contribution of bone marrow cells to the liver regenerating epithelial compartment

  1. Top of page
  2. Abstract
  3. Background
  4. Hepatocyte-mediated liver regeneration
  5. Hepatic progenitor cell-mediated liver regeneration
  6. Immunophenotype of HPCs
  7. Mechanisms of hepatic progenitor-cell activation
  8. Differentiation of hepatic progenitor cells and the role of HPC niche
  9. Contribution of bone marrow cells to the liver regenerating epithelial compartment
  10. Concluding remarks
  11. Acknowledgements
  12. References

Bone marrow is a complex microenvironment consisted of different cell types including hematopoietic stem cells (HSCs), non-circulating stromal cells called mesenchymal stem cells (MSCs) and unsorted mononuclear cells. Stem cells from bone marrow have potentially high degree of plasticity which makes them capable of giving rise to a wide range of cell types including hepatocyte-like cells. The question of HPCs being a continuous population with BMSCs was first raised by the following, but not all generally accepted [54, 58] observations: (i) oval cells express hematopoietic markers like CD34, CD45, Thy-1and c-kit [59], (ii) invading bone marrow cells and oval cells both express CXCR4 receptor and respond to SDF-1 in massive liver injury models of oval cell-mediated liver regeneration [43], (iii) presence of Y chromosomes in the hepatocytes of female recipients of male bone marrow transplants [60]. Several groups have investigated the ability of transplanted BMSCs to repopulate the liver (indirectly by giving rise to oval cells, or directly by transdifferentiation to hepatocytes) in different animal models of liver injury [60-68] of which the most notable were study of Petersen [61] in rats treated with 2-acetylaminofluorene and of Laggase [62] in an animal model of tyrosinemia type I [fumarylacetoacetate hydrolase-deficient mouse (FAH-/- mouse)]. In the latter study, HSCs transplantation completely repopulated the liver and rescued the mouse. Follow-up studies of these initial observations, using different populations of H(S)Cs in different models of animal liver injury showed that the aforementioned results were the consequence of recipient hepatocytes and HSCs fusion and that the hepatocyte replacement rate was very low [63]. Although parallel studies have found that generation of hepatocytes from HSCs can occur in the absence of cell fusion [64, 65], regardless of the mechanism, the hepatocyte replacement rate was very low. Contrary to these reports, the study performed by Wagers et al., in which a single GFP+c-kit+Thy1.LinSca-1+ (KTLS) adult BM HSC was transplanted into lethally irradiated non-transgenic recipients, failed to show HSC engraftment to the liver, indicating ‘transdifferentiation’ of HSCs and/or their progeny is an extremely rare event, if it occurs at all [66]. Using a similar cell population of BM Scal+/lin- HSC, Cantz et al. found the similar observations [67]. These finding clearly show that in the absence of selective growth advantage for stem cells, which so far has only been observed in FAH-/- model, the rate of newly stem cell derived hepatocytes is very low and their biological relevance is; therefore, unclear. Furthermore, depending on the source and phenotype of the isolated/transplanted stem cell rich populations and the model and species of liver injury, different outcomes can be observed in terms of number and mechanism of newly derived hepatocytes.

It has been shown that MSCs, like HSCs, can transdifferentiation into hepatocyte-like cells in vitro using specific differentiation media [68, 69]. However, in vivo animal models of MSCs transplantation have shown different somehow conflicting results ranging from profibrogenic potentials [70, 71] to contribution to hepatocytes pool [72, 73]. Following reasons can at least partly explain this controversy: (i) MSCs are heterogeneous populations which do not have a specific marker to be used for their isolation; therefore, they cannot be isolated with high purity. Indeed, different labs have their own protocol for MSCs isolation, culture and differentiation induction and use MSCs in different stages of their differentiation in their experiments resulting in the use of different cell populations while they have all been named MSCs, (ii) the same as HSCs, different models of injury have being used to investigate their regenerative potential, while the aetiology of liver damage has an important role in the observed outcome, (iii) very importantly, functional BM-derived hepatocytes are differently and sometimes incompletely defined by different studies. In a very recent study by Qiang et al. although transplantation of MCSs resulted in better liver function and survival in transplanted animals, the stem cell engraftment rate was very low and could not explain the observed beneficial effects; instead, transplanted MSCs in this study exhibited significant immunomodulatory effects on hepatic stellate cells and also proliferative effect on recipient mature hepatocytes which can explain the mechanism of their action [74]. Another study, using BMMSCs and hepatocyte derived from BMMSCs (MDHs) in a model of mice CCL4 induced fulminant liver failure, showed that both groups of cells rescued liver failure, although transplantation of MSCs showed superior rescuing potentials than MDHs [75]. In this model, although the early engraftment rate of about 4% was shown, differentiation of the MSCs into hepatocytes cannot explain the survival of the mice because engraftments frequencies were reduced to 0.001% and 0.01% at 3 and 6 months post-transplantation respectively. Furthermore, the animals' liver showed compensatory growth (approximately two-thirds of the tissue mass removed was regained) only 3 weeks after PH. These observations, together with superior rescuing potentials in the transplantation of MSCs (not hepatocyte derived MSCs), indicate that different mechanisms other than true differentiation of MSCs into hepatocytes are implicated in the rescue of the animals. Most, but not all, studies of BM transplantation in experimental models of liver injury have demonstrated beneficial effect. The exact mechanisms by which BMMSCs contribute to liver regeneration are not clear, and in particular, they do not account for significant number of new hepatocytes in the damaged liver. It seems that the transplantation of MSCs somehow stimulates the recipient's liver indirectly to repair itself by secreting a wide range of chemokines, cytokines and growth factors that have trophic and/or paracrine effects including stimulation of revascularization and regeneration of endogenous parenchymal cells [76-78]. In addition, it has been shown that MSCs have prominent resistance to oxidative stress which is probably because of enhanced reactive oxygen species (ROS)-scavenging potentials [75].

In humans setting although donor derived hepatocytes have been observed in archival liver biopsies [79], the lack of molecular data have made it difficult to find the mechanism regulating their hepatocytic differentiation from BMSCs, if any [80], and therefore, their biological and clinical relevance still remains to be clarified.

Macrophages are another cell types that significantly contribute to HPCs/oval cells mediated liver regeneration. As was discussed before, during liver injury, bone marrow derived macrophages have been found in close proximity to oval cells/HPCs, capable of influencing their differentiation direction [51, 54]. It has also been suggested that infusion of BM-derived macrophages can improve murine liver function and fibrosis [81]. Thomas et al. in their study tested the therapeutic effects of differentiated BM-derived macrophages (BMMs) and their specific BM precursors on murine liver fibrosis [81]. They found BMMs, unlike their precursors, had anti-fibrotic effect. However, in this study the improved liver function following BMMs therapy was multifactorial and because macrophages have many different roles during injury [82], a direct effect of BM-derived macrophages on HPCs was not shown. Consequent works in undamaged mice liver, in which a single infusion of unfractionated BM cells was used, have found the expansion of a subpopulation of HPCs/oval cells [83]. This study revealed that the macrophages within BM cells were responsible for this effect by producing TWEAK. They demonstrated a direct link between, TWEAK, produced by macrophages and its paracrine effect on HPCs/Oval cells expansion using complimentary TWEAK/Fn14 KO models.

Taken together, these data clearly show the influence of BM upon HPCs behaviour via paracrine signals. Although HPCs are not of BM origin, they are under strong influence of BM which may explain the potential benefit of BM cell therapy, observed in non-randomized and uncontrolled trials of human BM cell therapy in cirrhosis.

Concluding remarks

  1. Top of page
  2. Abstract
  3. Background
  4. Hepatocyte-mediated liver regeneration
  5. Hepatic progenitor cell-mediated liver regeneration
  6. Immunophenotype of HPCs
  7. Mechanisms of hepatic progenitor-cell activation
  8. Differentiation of hepatic progenitor cells and the role of HPC niche
  9. Contribution of bone marrow cells to the liver regenerating epithelial compartment
  10. Concluding remarks
  11. Acknowledgements
  12. References

Currently the only curative option in patients with liver failure is liver transplantation. However, liver transplantation is not widely performed because of the shortcomings such as donor shortage, risk of rejection and lifelong immunosuppression requirement. To compensate for this, development of new regenerative modalities for end-stage liver diseases is an urgent need. The liver's resident progenitor cells are a realistic potential target for producing new hepatocytes. However, there are still obstacles that need to be overcome before their clinical use. Indeed, isolation and culturing of a pure population of progenitor cells from human material and their transplantation have been found to be difficult. A more realistic approach for the next decade, until overcoming these obstacles is, therefore, targeting HPCs in situ by small molecules. To this aim, dissecting the molecular mechanisms underlying HPCs differentiation is of utmost importance and will enable us to control their behaviour towards the desired therapeutic purposes. Furthermore, such clues will help us to develop strategies for differentiation of non-hepatic stem cells into hepatocytes in vivo. In the context of severe parenchymal liver impairment characterized by massive necrosis of hepatocytes, in which a prominent ductular reaction is observed, enhancing hepatocytic differentiation of HPCs might provide a valid future target for drugs or other similar small molecules.

References

  1. Top of page
  2. Abstract
  3. Background
  4. Hepatocyte-mediated liver regeneration
  5. Hepatic progenitor cell-mediated liver regeneration
  6. Immunophenotype of HPCs
  7. Mechanisms of hepatic progenitor-cell activation
  8. Differentiation of hepatic progenitor cells and the role of HPC niche
  9. Contribution of bone marrow cells to the liver regenerating epithelial compartment
  10. Concluding remarks
  11. Acknowledgements
  12. References