Liver stem cells and prospects for liver reconstitution by transplanted cells


  • Potential conflict of interest: Nothing to report.


Although it was proposed almost 60 years ago that the adult mammalian liver contains hepatic stem cells, this issue remains controversial. Part of the problem is that no specific marker gene unique to the adult hepatic stem cell has yet been identified, and regeneration of the liver after acute injury is achieved through proliferation of adult hepatocytes and does not require activation or proliferation of stem cells. Also, there are differences in the expected properties of stem versus progenitor cells, and we attempt to use specific criteria to distinguish between these cell types. We review the evidence for each of these cell types in the adult versus embryonic/fetal liver, where tissue-specific stem cells are known to exist and to be involved in organ development. This review is limited to studies directed toward identification of hepatic epithelial stem cells and does not address the controversial issue of whether stem cells derived from the bone marrow have hepatocytic potential, a topic that has been covered extensively in other recent reviews. (Hepatology 2006;43:S89–S98.)

What Is a Stem Cell, and How Is This Different From a Progenitor Cell?

During mammalian embryonic development, stem cells originate from the inner cell mass of the blastocyst and are pluripotent. These cells, termed embryonic stem cells, give rise to somatic stem cells that differentiate further into multipotent tissue-specific stem cells.1–3 The properties of somatic stem cells have been established from studies in tissues with high turnover, such as hematopoietic cells of the bone marrow and epithelial cells of the skin and intestinal mucosa, in which continuous proliferation and differentiation of stem cells maintains tissue mass at a steady state.4–7

Stem cells are generally considered to exhibit the following specific properties: (1) capacity for self-renewal or self-maintenance (generally slowly cycling); (2) multipotency (capable of producing progeny in at least 2 lineages); (3) functional, long-term tissue reconstitution; and (4) serial transplantability (Fig. 1). For a pool of stem cells to be maintained in adult tissue, at least some of the cells must undergo division without differentiation or some cells must undergo asymmetric division, such that one of the progeny remains undifferentiated, while the other proliferates and differentiates to generate new tissue mass.8–10

Figure 1.

Properties of somatic (tissue-specific) stem cells.

Progenitor cells are the progeny of stem cells. They comprise distinct subpopulations, some with multilineage potential (early progenitor or stem/progenitor cells) and others (late progenitor cells) that have differentiated further and give rise to progeny in only a single lineage (Fig. 1). In contrast to stem cells or stem/progenitor cells, late progenitor cells do not self-renew; they divide rapidly but are capable of only short-term tissue reconstitution. Progenitor cells cannot be serially transplanted, and they have been termed “transit amplifying cells.”5, 8


uPA, urokinase plasminogen activator; FAH, fumarylacetoacetate hydrolase; 2-AAF, 2-acetylaminofluorine; PH, partial hepatectomy; AFP, α-fetoprotein; Alb, albumin; HNF, human nuclear factor; HSC, hematopoietic stem cell; DDC, 3,5-diethoxycarbonyl-1,4-dihydrocollidine; GGT, gamma-glutamyltranspeptidase; LEC, Long-Evans Cinnamon; CD, choline-deficient; FACS, fluorescence activated cell sorting; α1-AT, alpha1-antitrypsin; ED, embryonic day; H-CFU-C, hepatic colony-forming-unit.

Are Hepatocytes Liver Stem Cells?

In rapidly turning over tissues, stem cells play a major role in normal tissue homeostasis, as well as in tissue repair after acute injury (e.g., wound healing in the skin). However, in the mammalian liver, normal epithelial cell turnover is very slow (up to 1 year or more), but during acute injury [two-thirds partial hepatectomy (PH) or exposure to hepatotoxins], hepatocytes undergo rapid cell division (1-2 cycles) and replace lost cells within days.11, 12 Initially, it was thought that regeneration of the liver by adult hepatocytes represents a very limited capability of hepatocytes to proliferate; however, in the last decade, landmark studies have demonstrated that hepatocytes under specialized circumstances have virtually unlimited proliferative potential. Transplantation of normal (wt) hepatocytes into urokinase plasminogen activator (uPA) transgenic mice, in which host hepatocytes undergo massive and continuous necrosis,13 resulted in 12 or more cell divisions, and most of the host liver was replaced.14 In fumarylacetoacetate hydrolase (FAH) knockout mice, in which there is also extensive and continuous liver injury, the metabolic disorder can be corrected by near total liver repopulation by transplanted wt hepatocytes with restoration of normal lobular structure and function.15, 16 Using this cell transplantation model, Grompe and coworkers demonstrated further that normal adult hepatocytes can be serially transplanted through seven generations of mice, with each transplanted cell undergoing an average of 69 or more divisions16 and, by retroviral marking in one animal, that single FAH+ hepatocytes can be clonally amplified and serially passaged to repopulate the FAH−/−mouse liver.17

In other models of liver repopulation by transplanted cells, host hepatocytes have been rendered incapable of proliferation by DNA damage through treatment with DNA alkylating agents, retrorsine,18 or monocrotaline,19 or x-irradiation20, 21; others have used genetically modified hepatocytes exhibiting increased proliferative activity or resistance to apoptosis, in conjunction with repeated host liver injury.22, 23 To obtain extensive repopulation by adult hepatocytes, all of these models exhibit two features: (1) extensive and continuous liver injury and (2) a strong selective advantage for transplanted cells to survive compared with host hepatocytes. It has also been shown that hepatocytes can differentiate into cholangiocytes and form mature bile ducts, when there is massive bile duct injury.24 Thus, after transplantation under highly selective conditions, hepatocytes fulfill all the listed criteria for stem cells.

Is There a Unique Subpopulation of Hepatic Stem Cells in the Adult Liver?

Studying liver regeneration in mice after severe nutritional injury in the late 1950s, Wilson and Leduc observed that non-parenchymal cells of the distal cholangioles proliferated and, after cessation of injury, differentiated into hepatocytes and possibly into new interlobular bile ducts.25 During the same period, it was established that hepatocytes and bile duct epithelial cells are of common embryonic origin.26, 27 Therefore, it was reasonable to propose that hepatocytes are derived from epithelial precursors in the distal cholangioles. However, specific identification of these cells has not been possible, because unique liver stem cell markers have not yet been found. Nonetheless, numerous studies during the last decade have identified cells both within and outside the liver that exhibit properties of hepatic stem cells and can differentiate into hepatocytes and/or bile duct epithelial cells both in culture and after their transplantation (see “Oval Cells” as Hepatocyte Progenitors).

“Oval Cells” as Hepatocyte Progenitors

The term “oval cells” was first introduced by Farber,28 who found non-parenchymal cells with a characteristic morphological appearance after treatment of rats with carcinogenic agents, ethionine, 2-acetylaminofluorine (2-AAF), and 3-methyl-4-dimethyl aminobenzene. He described these cells as “small oval cells with scant lightly basophilic cytoplasm and pale blue-staining nuclei.”28 In subsequent studies, Farber and coworkers concluded that “oval cells” are not progenitors of hepatocytes.29 However, Thorgeirsson and coworkers clearly showed that “oval cells,” induced in the rat liver by treatment with 2-AAF/PH, proliferate in the periportal region and take up [3H] thymidine, which subsequently accumulates in clusters of basophilic hepatocytes in the mid parenchyma.30 The expression of bile duct markers (CK-7, CK-19, and OV-6) and hepatocytic markers [alpha-fetoprotein (AFP) and albumin (Alb)] in [3H] thymidine-labeled “oval cells” over time also suggested a precursor/product relationship between “oval cells” and hepatocytes.30, 31 Moreover, the temporal pattern of liver-enriched transcription factor expression [human nuclear factor (HNFs) and CCAAT/enhancer-binding protein] in the liver during “oval cell” activation by 2-AAF/PH also mirrored their expression pattern during liver development.32 Finally, studies showing activation of stem cell genes, c-kit,33 CD34,34 flt-3 receptor35 and leukemia inhibitory factor,36 during “oval cell” proliferation suggested their stem cell–like properties.

“Oval cells” are a heterogeneous population of cells activated to proliferate under different pathophysiological circumstances12, 37–39 and among these populations, some cells express 1 or more hematopoietic stem cell genes, such as c-kit, CD34, Sca-1, and Thy-1.33, 34, 40–43 This led to the suggestion that they originate from hematopoietic stem cells (HSC).42, 44 However, several recent studies have shown that “oval cells” in the mouse liver are not derived from HSC45 and that transdifferentiation of HSC into “oval cells” in the rat liver is a very rare event,46 which probably does not have physiological significance.38, 39, 46, 47

Other Models of “Oval Cell” Activation

Cells with the morphological appearance of “oval cells” are also induced to proliferate in rodents by a number of other regimens: administration of a choline-deficient diet supplemented with ethionine,25 treatment with other DNA alkylating agents; 1,4-bis[N,N′-di(ethylene)-phosphamide] piperazine (Dipin),48 or 12,18-dihydroxy-senecionan-11-16-dione (retrorsine) in combination with PH,49 feeding 3,5-diethoxycarbonyl-1,4-dihydrocollidine (DDC),50 phenobarbital/cocaine-induced liver injury,51 acute allyl alcohol toxicity52 or D -galactosamine administration, which temporarily blocks hepatocyte proliferation.53, 54

“Oval cells” activated in all these models display the following characteristics:

  • 1They express markers specific for biliary epithelial cells, such as CK-7, CK-8, CK-18, CK-19, OV-6, glutathione-S-transferase, connexin 43, mouse A6 antigen, and others, markers for immature fetal hepatoblasts, such as AFP, gamma-glutamyltranspeptidase (GGT), muscle pyruvate kinase, and some of these cells express markers for HSC, such as Thy-1, c-kit, CD34, and Sca-1, as well as markers not expressed by other liver cell types, for example, chromogranin A, neural cell adhesion molecule, and Dlk-1.39, 12, 42, 50, 55–58
  • 2By tracing [3H] thymidine transfer from “oval cells” to parenchymal cells and by following “oval cell” differentiation using lineage markers [AFP, GGT, G6P, alpha1-antitrypsin (α1-AT), CK-19, and OV-6], several studies have shown that these cells behave like bipotential progenitor cells and have been termed “facultative or reserve stem cells.”12, 30, 31, 39, 59, 60
  • 3When activated to proliferate, “oval cells” give rise to duct-like structures originating from the periportal region and spreading into the liver acinus. The duct-like structures change appearance after several days, as they transform into clusters of small basophilic hepatocytes and mature bile ducts.
  • 4Activated “oval cells” move in close contact with liver stellate cells,61, 62 which secrete growth factors [transforming growth factor alpha, hepatocyte growth factor, fibroblast growth factor and tumor necrosis factors] that are necessary for “oval cell” growth and proliferation.

A word of caution, however, is that the number of progenitor cells in the normal liver is extremely low63, 64 and their activation and proliferation in response to liver injury (which is required for their isolation in sufficient quantity to study) may change their properties.

Transplantation of “Oval Cells” and “Oval Cell” Lines

Stable “oval cell” lines have been established from normal liver of Fischer (F)-344 rats65 and from the liver of rats fed DL -ethionine66 or allyl alcohol treatment,52 from livers of Long-Evans Cinnamon (LEC) rats, an animal model of Wilson's disease,67 from livers of transgenic rats expressing the Ras oncogene,68 from p53 knockout mice fed a choline-deficient, ethionine-supplemented diet,69 and from human liver.41 The common feature of these “oval cell” lines is their ability in vitro to differentiate into cells expressing hepatocytic or biliary epithelial characteristics, suggesting their bipotential properties.

The most important property establishing “oval cells” as liver stem cells would be their capability to restore liver cell mass after their transplantation.70–72 Several attempts have been made to transplant isolated “oval cells” or “oval cell” lines into the liver of normal, carcinogen-fed or metabolically defective rats or mice. The first successful transplantation of “oval cells” was performed more than 15 years ago, using “oval cells” isolated from the liver of rats fed a choline-deficient (CD)/2-AAF diet.73 It was found that these cells produced “colonies” or clusters in vivo with phenotypes resembling hepatocytes only in the livers of rats fed a CD/2-AAF diet. Transplantation of an epithelial cell line originating from normal liver (WB-344), which exhibits stem cell properties in vitro,74 into the liver of syngeneic animals showed their incorporation into hepatic plates and acquisition of hepatocyte-specific gene expression but limited proliferative capacity.75 “Oval cells” isolated from the liver of D-galactosamine–treated rats also showed engraftment but only very limited proliferation and liver repopulation.76 “Oval cells” isolated from the liver of LEC rats, tagged with β-galactosidase ex vivo and transplanted into the liver of LEC and Nagase analbuminemic rats, also displayed an hepatocytic phenotype and produced albumin,67 but, once again, liver repopulation was low. More recently, “oval cells” were isolated from the livers of DDC-fed mice, and transplanted into FAH null mice. In this model, “oval cells” were at least as efficient as mature hepatocytes in repopulating the liver.45 In another report, an “oval cell” fraction from green fluorescent protein (GFP) transgenic mice, maintained on a DDC diet, was isolated by FACS, using the surface marker Sca-1.77 The “oval cells” were transduced with human α1-AT and transplanted into monocrotaline-treated C57BL/6 mice in conjunction with PH. The authors reported 40% to 50% liver repopulation by GFP expressing cells, of which 5% to 10% also expressed human α1-AT. These results showed that high-level, liver repopulation can be achieved with “oval cells”, but this requires substantial modification of the host liver environment, similar to that required for liver repopulation by mature hepatocytes.

Hepatic Stem Cells During Liver Embryonic Development

During normal liver development in rodents, when the ventral wall of the foregut endoderm becomes positioned adjacent to the developing heart on embryonic day (ED) 8.0 (Fig. 2), endodermal stem cells begin to proliferate and differentiate.26, 27, 78–80 Specification toward the hepatic epithelial lineages occurs at ED8.5 and requires fibroblast growth factor signaling from the cardiogenic mesoderm81 and bone morphogenic protein signaling from the septum transversum mesenchyme.82 Cells expanding from the foregut endoderm can be identified at ED8.5 to ED9.0 as an outgrowth from the ventral wall. By ED9.0 to ED9.5, these cells begin to bud within the immediately adjacent stromal environment and express GATA4 and liver-enriched, nuclear factor HNF4α, as well as liver-specific genes, AFP, followed by Alb. The hepatic-specified cells are now considered to be hepatoblasts and proliferate massively. Cords of hepatoblasts invade the septum transversum mesenchyme, which is the source of stellate cells and sinusoidal endothelial cells that subsequently develop into blood vessels.80

Figure 2.

Schematic representation of rodent fetal liver development.

At ED11, HSC invade the liver bud and form a visible liver structure that is primarily a hematopoietic organ. Hepatoblasts continue to proliferate rapidly and begin to express placental alkaline phosphatase, intermediate filament proteins (CK-14, CK-8, and CK-18), and GGT, and later α1-AT, glutathione-S-transferase-P and fetal isoforms of aldolase, lactic dehydrogenase, and muscle pyruvate kinase.83–86 Just before ED16, hepatoblasts diverge along 2 lineages, hepatocytes and cholangiocytes.87–89 Differentiation along the cholangiocytic lineage is promoted by Notch signaling pathways and is antagonized by hepatocyte growth factor, which in conjunction with oncostatin M promotes hepatocytic differentiation.90 After ED16, a massive change occurs in the gene expression profile of rat fetal liver epithelial cells to a more differentiated phenotype,91 and the percentage of bipotent cells, that is, those expressing genes in both the hepatocytic and cholangiocytic lineages (e.g., AFP or Alb and CK-19, respectively), is markedly reduced.88, 89, 92, 93 At this point, most of the cells are unipotent and irreversibly committed to either the hepatocytic or cholangiocytic lineage,88, 89, 92 although they continue to proliferate (i.e., unipotent or late progenitor cells).

As organogenesis proceeds, intrahepatic bile ducts are formed in the vicinity of large portal vein branches, beginning on approximately ED17.94 The basic lobular structure is now completed, although the parenchymal cords do not become fully mature until several weeks after birth.

Bipotential Cells and Cell Lines Generated From Rodent Fetal Liver

Kubota and Reid, 2000 isolated RT1A1−/OX18low/ICAM-1+ single cells from rat ED13 fetal liver that were clonogenic hepatoblasts, exhibiting bipotential differentiation in vitro.95 Sorting of murine fetal liver cells at ED11.5, 13.5, 16.5, and 18.5 and neonatal liver for a Ter119/CD45/c-kit-/CD29+/CD49f+ cell fraction (Alb/CK-19), Suzuki et al.96 isolated highly proliferative cells, termed hepatic colony-forming-units (H-CFU-C), that showed hepatocytic and biliary lineage markers in low-density cell culture.96 In subsequent studies, Suzuki et al. isolated H-CFU-C as single cells by FACS sorting and demonstrated their bipotential differentiation capacity in vivo.97 Furthermore, they showed “self-renewal” of H-CFU-C in culture by subcloning through serial single-cell sorting. These cells also showed pancreatic, gastric, and intestinal epithelial differentiation in vitro and in vivo and therefore were regarded as endodermal stem cells.97 In the same year, specific surface markers for hepatoblasts were identified: Liv2,98 E-cadherin,99 and Dlk-1,100 and all of these markers identified the same population of AFP+/Alb+/Pan-CK+/Sca-1+/c-kit/CD34/CK19 hepatoblasts at ED12.5.101 E-cadherin and Dlk-1 were also shown to be useful for purification of fetal liver epithelial cells using immunomagnetic beads or FACS sorting.99–101 Cantz et al.102 and Nierhoff et al.101 have also demonstrated liver repopulation with mouse fetal liver epithelial cells in uPA transgenic/Rag2 null and retrorsine/CCl4-treated mice, respectively.

Cell lines have been established from mouse embryonic foregut at ED9.5.103 One of these lines, HBC-3, grew well at low cell density and exhibited bipotential differentiation properties. This cell line also identified the importance of the Wnt signaling pathway during hepatocytic differentiation.104 Using transgenic mice expressing a constitutively active truncated human Met receptor (c-met), met murine hepatocyte cell lines were established from mouse ED14.5 fetal and neonatal livers.105 These met murine hepatocyte cell lines were subsequently subcloned, producing cells with an epithelial morphology expressing liver-enriched transcription factors (HNF1α and HNF4), as well as E-cadherin and Zo-1, but not expressing Alb, transthyretin (TTR), or β-fibrinogen, and cells with a fibroblastic appearance, termed palmate cells, that were essentially negative for all of these markers, but gave rise to epithelial clusters or bile duct–like structures when cultured under selective differentiation conditions.106 Following the same methodology, Strick-Marchand and Weiss107 isolated bipotential cell lines from non-transgenic mouse embryonic liver that exhibited bipotential capacity in vitro107 and participated in liver regeneration in uPA/SCID mice in vivo.108 Bipotential liver epithelial cells have also been isolated and cultured from porcine,109 monkey,110 and human fetal liver.111, 112 Thus, essential steps have been taken to define epithelial cells in the fetal liver as stem-like cells by demonstrating: (1) clonality, (2) bipotential differentiation, (3) high proliferative capacity as cell lines, and (4) ability to differentiate into mature hepatocytes and bile duct cells in vivo. However, a major difficulty with cultured cells and cell lines is very low engraftment efficiency, and such cells appear to be poorly responsive to regenerative signals, even in the massively injured uPA transgenic mouse liver.102, 108

In Vivo Stem Cell Properties of Primary Rat Fetal Liver Epithelial Cells

As stated previously, the ultimate test for a liver stem cell is to demonstrate its ability to self-renew in vivo and to functionally repopulate the liver.70–72 A dilemma that arises in studies conducted to date is that under the same liver injury and selection conditions used with cultured rodent liver cells and cell lines, including “oval cells”, primary adult hepatocytes can fully repopulate the liver. A more definitive approach would be to transplant putative stem cells under conditions in which primary adult hepatocytes will not significantly repopulate the liver. This has been achieved with primary fetal liver cells isolated from wt Fischer (F)-344 rats and transplanted into DPPIV mutant F344 rats subjected to two-thirds PH.

Isolated rat ED14 fetal liver cells contain a subpopulation showing dual expression of AFP and CK-19.92, 93 Liver repopulation by transplanted rat ED14 fetal liver epithelial cells increased progressively over 6 months, and the bulk of repopulating clusters contained both hepatocytes and mature bile duct epithelial cells (“mixed” clusters).93 The transplanted cells were integrated into the host parenchyma and formed hybrid bile canaliculi with host hepatocytes, and in initial studies, 6.6% ± 2.8% liver repopulation was achieved.93 In more recent studies, using 4× to 5× greater numbers of unfractionated rat ED14 fetal liver cells, (containing approximately 1 × 106 AFP+/CK19+ hepatoblasts), 23.5% ± 2.1% replacement of liver mass has been achieved.113 The progeny of transplanted rat ED14 fetal liver cells differentiate into hepatocytes that are morphologically and functionally indistinguishable from neighboring host hepatocytes. These cells express normal levels of many liver-specific genes, such as albumin, the asialoglycoprotein receptor, glucose-6-phosphatase, and UDP transferase, at levels comparable to host hepatocytes at all locations within the hepatic lobules.113 Thus, transplanted rat ED14 fetal liver epithelial cells exhibited major properties of liver stem cells, (1) extensive proliferation, (2) apparent bipotency, (3) liver-specific function, and (4) long-term repopulation in vivo. Because studies were not conducted to demonstrate self-renewal or serial transplantability of rat ED14 fetal liver epithelial cells, we refer to these cells as fetal liver stem/progenitor cells. It should also be noted that derivation of repopulating hepatocytes and bile duct epithelial cells in “mixed” clusters from a common precursor was not formally demonstrated, because this would require retroviral marking of transplanted cells and identification of the same viral DNA integration site in progeny exhibiting each of these phenotypes. After ED16, the ability of fetal liver epithelial cells to repopulate the normal adult rat liver decreased dramatically,93 although repopulation capacity in retrorsine/PH-treated liver remained very high (Sandhu et al., unpublished observations). Most of the cells after ED16 were unipotent, expressing either AFP or CK-19 but not both markers and were considered to represent late progenitor cells.93

Mechanism by Which Fetal Liver Stem/Progenitor Cells Replace Hepatocytes in the Adult Liver

Transplanted fetal liver stem/progenitor cells exhibit higher proliferative capacity and a lower apoptotic rate than endogenous host hepatocytes.113 They repopulate the liver by inducing apoptosis of neighboring host hepatocytes that are proliferating more slowly, while maintaining total liver mass at a fixed size by mechanisms that have not yet been established. This process, referred to as cell–cell competition,113 represents the mammalian counterpart of a mechanism described originally in Drosophila during wing development. In chimeric Drosophila, more rapidly proliferating wing epithelial cells out-compete neighboring, more slowly growing cells of an otherwise identical phenotype by inducing their apoptosis.114, 115 Recapitulating this developmental paradigm in the adult mammalian liver by transplanting fetal liver stem/progenitor cells to achieve functional organ reconstitution, has broad therapeutic implications.

The Liver Stem Cell Niche

Evidence cited previously supports the view that stem cells or at least their immediate progeny, stem/progenitor cells, exist in the fetal liver. Whether cells with such properties continue to exist in the adult liver still remains to be established. If such cells do exist, the question remains as to where do they reside (i.e., where is their “niche”) (Fig. 3).

Figure 3.

Schematic diagram depicting liver epithelial cell lineage progression.

The original idea of a stem cell “niche” evolved from the concept that stem cells reside in tissues within an “inductive microenvironment” that directs their differentiation.116, 117 More recently, the stem cell niche has been described as “a specific location in a tissue where stem cells can reside for an indefinite period of time and produce progeny cells, while self-renewing”.118 Local stromal cells and other extracellular environmental factors attract stem cells to these niches and affect their behavior (gene expression program, proliferation or differentiation), and such niches have been identified in the bone marrow, brain, skin, and intestinal mucosa.119–123

The most likely candidate for a liver stem cell niche is the canal of Hering (Fig. 3). The canals of Hering were original identified more than 100 years ago as luminal channels linking the hepatocyte canalicular system to the biliary tree.124 These channels contain small undifferentiated epithelial cells125, 126 that are in direct physical continuity with hepatocytes at one membrane boundary and bile duct cells at another boundary to form duct-like structures, enclosing a lumen (i.e., the canal). In most models of liver structure, the canals of Hering are depicted as very limited structures confined to the portal space. However, recent studies in humans have demonstrated that the canals of Hering are much more elaborate than previously thought and extend far into the hepatic parenchyma.127, 128 These structures contain epithelial cells with dual expression of bile ductular and fetal hepatocytic markers (AFP, HepPar1, CK-19) and were thus considered to represent “facultative hepatic stem cells.”60, 127

Recent studies in the rat 2-AAF/PH model have demonstrated that proliferating “oval cells” are indeed located in the canals of Hering.129 The proliferating cells are thought to pass into the parenchyma through discontinuities in the ductal basement membrane, continue to proliferate in the parenchyma in conjunction with stellate (Ito) cells, and then differentiate into hepatocytes129 or undergo apoptosis.130 Because the very first cell to undergo proliferation and subsequent differentiation into hepatocytes is found in the canals of Hering, this structure is thought to contain both stem cells and their initial progeny, “oval cells.” A separate population of less well-differentiated progenitor cells has also been identified in the periductular space, and these cells proliferate after allyl alcohol–induced liver injury.52 It has been suggested that these cells constitute a second progenitor or stem cell compartment, and they may originate from hematopoietic stem cells.131 Thus, the liver may have several distinct hepatic stem cell niches.

Human Studies

The human counterpart to “oval cell” activation has been identified in what has been termed the “ductular reaction.”132, 133 In patients with extensive chronic liver injury or submassive hepatic necrosis, regenerative structures have been identified containing epithelial cells with the morphological appearance and immunohistochemical markers consistent with “oval cells.”41, 56, 57, 60, 127, 134–138 Cells with similar morphological and immunohistochemical properties have also been identified in the human fetal liver beginning at 4 weeks gestation with expression of CK-19 and HepPar1 and later CK-14. Expression of these genes increases progressively between 8 and 14 weeks' gestation, with subsequent loss of CK-14 and CK-19 expression in hepatoblasts after 14-to 16-week gestation.139 A number of investigators have isolated, cultured, and/or passaged human fetal liver epithelial cells with bipotent properties,111, 112, 140–143 and several of these studies have demonstrated their subsequent differentiation into hepatocytes after transplantation into SCID or nude mice.111, 142 These studies suggest that the human counterpart to rodent fetal liver stem/progenitor cells exists and in the near future these cells will become available for liver repopulation.

Questions and Prospects for the Future

Although we have made substantial progress during the past 10 to 15 years concerning the possibility of liver repopulation by transplanted cells, much remains to be learned. What factors govern engraftment of transplanted cells into the liver and their homing to the correct niche, what factors regulate proliferation and differentiation of transplanted cells into specific phenotypes required for organ function, and what are the specific host conditions under which effective liver replacement can be achieved? Probably the best starting point for therapeutic liver repopulation will be a genetic disorder with some underlying liver disease to increase proliferative activity in the host environment by transplanted cells. One example would be Wilson's disease, in which the transplanted cells might also have a modest selective advantage, because they will not store high levels of copper. Other examples might include phenylketonuria and hemophilia B, in which there is no underlying liver disease and in which even low levels of liver repopulation might be effective in treating these disorders, as well as Crigler-Najjar syndrome type 1, where 7% to 10% repopulation should abrogate the need for phototherapy.

Clearly, we have been able to repopulate the rodent liver with hepatic cells and produce normal functional liver tissue. However, for cell transplantation to become a viable therapeutic procedure, we will need to achieve this outcome with cultured and expanded cells. Whether we will be able to repopulate the human liver with embryonic or other stem cell lines also remains to be determined. Hopefully, answers to these questions and effective, therapeutic liver cell transplantation will be achieved within the near future.


The authors thank their many colleagues at the Albert Einstein College of Medicine and at other institutions who have contributed to these studies and to Anna Caponigro for typing this manuscript and all of its predecessors.