Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Hepatology. Author manuscript; available in PMC 2010 December 13.
Published in final edited form as:
PMCID: PMC3001243

Role of Transforming Growth Factor β Signaling and Expansion of Progenitor Cells in Regenerating Liver


Adult hepatic progenitor cells are activated during regeneration when hepatocytes and bile duct epithelium are damaged or unable to proliferate. On the basis of its role as a tumor suppressor and in the potential malignant transformation of stem cells in hepatocellular carcinoma, we investigated the role of key transforming growth factor beta (TGF-β) signaling components, including the Smad3 adaptor protein β2-Spectrin (β2SP), in liver regeneration. We demonstrate a streaming hepatocyte-specific dedifferentiation process in regenerating adult human liver less than 6 weeks following living donor transplantation. We then demonstrate a spatial and temporal expansion of TGF-β signaling components, especially β2SP, from the periportal to the pericentral zone as regeneration nears termination via immunohistochemical analysis. This expansion is associated with an expanded remaining pool of octamer 3/4 (Oct3/4)-positive progenitor cells localized to the portal tract in adult human liver from more than 6 weeks posttransplant. Furthermore, disruption of TGF-β signaling as in the β2SP (β2SP+/–) knockout mouse demonstrated a striking 2 to 4-fold (P < 0.05) expanded population of Oct3/4-positive cells with activated Wnt signaling occupying an alpha-fetoprotein (AFP)+/cytokeratin-19 (CK-19)-positive progenitor cell niche following two-thirds partial hepatectomy. Conclusion: TGF-β signaling, particularly β2SP, plays a critical role in hepatocyte proliferation and transitional phenotype and its loss is associated with activation of hepatic progenitor cells secondary to delayed mitogenesis and activated Wnt signaling.

Liver regeneration involves a complex sequence of signaling events to restore liver mass and function. Following two-thirds partial hepatectomy, 95% of differentiated hepatocytes exit G0 and synchronously reenter the cell cycle. DNA synthesis begins within 24 hours and peaks 36-48 hours posthepatectomy in most mouse strains.1 Restoration of liver mass is nearly complete by 5-7 days in rodents and by 3-4 months in humans.2

When hepatocytes and bile duct epithelium are severely damaged or unable to proliferate, a population of hepatic progenitor cells is activated. These progenitors represent a heterogeneous spectrum of cells that express markers corresponding to both the hepatocytic and cholangiocytic lineages and serve as a source of cell replenishment and tissue repair as they differentiate into either cell type, thereby aiding in liver regeneration.3,4 Hepatic progenitor cells have been described to reside and originate from several potential sources including the canals of Hering, intralobular bile ducts, periductal mononuclear cells, and peribiliary hepatocytes.5 To date, progenitor cell activation has been described in rats on the Solt-Farber protocol, in mice fed with choline-deficient, ethionine-supplemented or 3,5-diethoxycarbonyl-1.4-dihydrocollidine (DDC) diets, as well as in numerous hepatic pathologies.6,7 Progenitor cells, however, have been rarely observed in acute liver injury models such as surgical resection or two-thirds partial hepatectomy and little is known about the precise mechanisms by which progenitor cells are activated and then differentiate into mature hepatocytes and bile duct epithelium.

There is ample evidence that transforming growth factor beta (TGF-β) signaling plays a critical role in liver regeneration. To date, the TGF-β signaling pathway is most well known for its antiproliferative effect on hepatocytes and has been shown to reversibly inhibit the proliferative response following partial hepatectomy.8 TGF-β-family messenger RNA (mRNA) and protein are upregulated in quiescent livers in which the majority of cells are in G0 and downstream Smad protein activity as assessed by phospho-Smad2, Smad2, and Smad4 levels are significantly enhanced following partial hepatectomy.2,9,10 TGF-β signaling is inhibited in the early regeneration period by a concomitant up-regulation of TGF-β inhibitory proteins SnoN and Ski and a down-regulation of the TGF-β receptors allowing cell proliferation to transition from G1 to S phase.10 Moreover, experiments with liver-specific conditional knockout mice confirm a key role for TGF-β signaling in hepatocyte mitogenesis and the termination of liver regeneration.11

There is also growing evidence that TGF-β signaling proteins play a role in both the maintenance of cells in their undifferentiated state and in the initiation of differentiation. TGF-β family proteins are thought to play a role in the maintenance of embryonic stem (ES) cell identity12 and mediate key decisions specifying germ layer identification, including hepatoblast development from endoderm.13 In addition, TGF-β signaling has also been implicated in the maintenance and differentiation of somatic stem cells, particularly of the gastrointestinal tract, and in mediating the stem cell niche.12,14,15

We have previously demonstrated the role of a nonplekstrin homology (PH) domain β-general-spectrin, β2SP (also known as Embryonic Liver Fodrin, ELF, or Spectrin β, nonerythrocytic 1 isoform 2) as a Smad3/4 adaptor protein that regulates TGF-β signaling.16 β2SP is also a key suppressor of tumorigenesis in hepatocellular carcinoma (HCC)17,18 and, recently, we identified putative hepatic progenitor cells in human HCC specimens expressing the stem cell markers signal transducer and activator of transcription 3 (STAT3) and octamer 3/4 (Oct3/4) that are strikingly negative for β2SP and the TGF-β type II receptor (TBRII), suggesting that loss of β2SP may play a role in the malignant transformation of hepatic progenitor/stem cells.19

This led us to evaluate the potential role of β2SP in mouse and human liver regeneration and, specifically, the activation of hepatic progenitor cells. Our initial analysis reveals that β2SP expression demonstrates a clear spatial and temporal variation as regeneration proceeds and has a reciprocal relationship with the expression of several progenitor cell markers. Reduced β2SP is also associated with an expanded population of hepatic progenitor cells following two-thirds partial hepatectomy that are likely activated by impaired hepatocyte proliferation and activated Wnt signaling in β2SP+/- mutant mice.

Materials and Methods

Human Liver Biopsy Specimens

Formalin-fixed and paraffin-embedded human postliving donor transplant liver biopsy specimens were obtained from the Department of Pathology, Georgetown University Medical Center, Washington, DC. Liver biopsies from 10 living donor transplant recipients were collected at 1 week (two specimens), 6 weeks (five specimens), and 12-16 weeks (three specimens) post-transplant as part of a standardized protocol to rule out liver pathology following living donor transplantation. Zero specimens were collected to evaluate for suspected rejection. All human tissue procedures were approved by the Institutional Review Board of Georgetown University Medical Center, Washington, DC.

Animals and Experimental Protocols

Wild-type and β2SP+/- 129 SvEv Black Swiss mice 8-16 weeks of age were subjected to two-thirds partial hepatectomy as described by Mitchell and Willenbring20 and then sacrificed at 0, 24, 48, 72, and 168 hours after hepatectomy (n ≥ 3). Liver tissue was then collected for immunohistochemical, protein, and RNA analysis. Generation of β2SP+/- knockout mice was as described.16

Protein Extraction and Western Blots

Whole-cell lysates were prepared from pooled livers from each experimental group with a radioimmunoprecipitation assay buffer (Sigma) containing fresh protease and phosphatase inhibitor cocktails. The primary antibodies used in this study were rabbit anti-β2SP (1:1000) and rabbit anti-actin (1:2500). Details of anti-β2SP antibody have been described.16 Horseradish peroxidase-conjugated secondary antibodies (Santa Cruz Biotechnology) were used at 1:5000 dilution. Blots were subjected to fresh chemiluminescent substrate and visualized by autoradiography. Once scanned, densitometric analysis was performed with SigmaGel software for quantitative analysis.

Microarray Analysis

Custom-designed 44K human 60-mer oligo microarrays (Agilent Technologies) were used for the array experiments. Total RNA was extracted from mouse liver using RNeasy kit (Qiagen).

Histochemical Studies and Confocal Immunofluorescence

Sections from human liver biopsies and mouse liver following partial hepatectomy were prepared and processed for immunohistochemistry. Slides were then incubated overnight at 4°C with primary antibody against β2SP, the TBRII (Santa Cruz Biotechnology), Oct3/4 (Abcam), AFP (Santa Cruz Biotechnology), and CK-19 (Chemicon), Ki-67 clone TEC-3 (Dako), and β-catenin (Santa Cruz Biotechnology). Biotinylated secondary antibody and signal enhancement were then performed using the Vectastain ABC kit (Vector Labs). Signal was then visualized by 3,3′-diaminobenzidine chromogen and substrate buffer (Vector Labs). The labeling index was calculated by dividing the number of positive labeling cells by the total number of cells/hpf (high powered field) averaged over 10 fields. Given the localization of Oct3/4-positive cells, the labeling index was calculated by dividing the number of positive labeling cells within a 50-μm radius of the portal tract by the total number of cells per radius.

Confocal Immunofluorescent Labeling of Human Regenerating Liver Specimens

Colocalization studies were performed with anti-β2SP, -TBRII, p-Histone, and -Oct3/4 antibodies using methods described.19 Primary antibodies were visualized with tetramethyl rhodamine isothiocyanate (TRITC)-conjugated goat antirabbit IgG or FITC-conjugated goat antimouse immunoglobulin G (IgG). Samples were analyzed with a Bio-Rad MRC-600 confocal microscope with an ILT model 5470K laser as the source of the krypton-argon ion laser beam.

Statistical Analysis

Results are expressed as the means ± standard deviation (SD) or ± standard error of the mean (SEM). Student's t test was used for comparison between groups. P values <0.05 were considered statistically significant.


TGF-β Pathway Member Expression Displays a Spatial and Temporal Variation in Regenerating Adult Human Liver

To assess whether TGF-β signaling pathway members and, specifically, β2SP plays a functional role in regenerating human liver, we studied liver biopsy tissue from 10 recipients of living donor liver transplantation. The surgical procedure involves resection and transplantation of the right or left lobe or left lateral segment of the liver, representing 55%-60%, 40%, or 25% of original donor liver mass, respectively, into a recipient. The donor graft then regenerates to ≈85% of the recipient liver mass by 3 to 4 months postsurgery.21 We assessed liver biopsy tissue procured as part of a standardized institutional protocol to evaluate liver regeneration at 1 week (n = 2), 4 weeks (n = 2), 6 weeks (n = 3), 12 weeks (n = 1), and 16 weeks (n = 2) posttransplant and initially focused on the expression of β2SP by immunohistochemical labeling. β2SP labeling was present in all specimens at all timepoints. The areas of most intense labeling, however, varied as a function of time following transplantation. Using the functional classification of hepatic lobules into zones 1 (periportal), 2, and 3 (pericentral veins), in specimens taken from 1 to 6 weeks posttransplant, β2SP labeling was concentrated in zone 1 (19% of cells labeled positively), expanding outward into zone 2 (34%), and diminishing into zone 3 (7%) (Fig. 1A). In specimens procured from 6 to 16 weeks posttransplant, however, β2SP labeling was markedly expanded, particularly in zone 3 (42% of cells labeled positively), and nearly uniform between zones (Fig. 1B). The overall mean percent of positively labeled cells for β2SP increased from 37% in specimens from 1 to 6 weeks to 74% in specimens from 6 to 16 weeks.

Fig. 1
Expression of β2SP and progenitor cell markers Oct3/4, AFP, and CK-19 in human liver tissue biopsies postliving donor transplant. (A) β2SP expression is localized to zone 1 (periportal) and zone 2 at 1 week and then (B) expands to all ...

Given the role of β2SP as a Smad3/4 adaptor protein, we also assessed the expression of other important mediators of this pathway, such as TBRII. TBRII, like β2SP, was present in all specimens at all timepoints. The labeling pattern was similar to that of β2SP, with an increased percent of positive-labeling cells in zone 1 in specimens from 1 to 6 weeks (26%) and a marked increase in labeling, most significantly in zone 3, in specimens from 6 to 16 weeks posttransplant (41%). Like β2SP, by 6 to 16 weeks TBRII labeling was nearly uniform between zones (Fig. 1 Table and Graph; Supporting Fig. 1). Overall, the mean percent of TBRII-positively labeled cells increased from 17% to 41% by the end of liver regeneration.

The increased labeling for TBRII and β2SP over time is consistent with the known role of the TGF-β signaling pathway in the termination of liver regeneration. The spatial variation in labeling over time, however, was unexpected and, per our knowledge, previously unreported.

Progenitor/Stem Cell Marker Expression Displays a Spatial and Temporal Variation in Regenerating Human Liver

Given our previous identification of STAT3/Oct3/4-positive labeling putative progenitor cells in human HCC that do not express β2SP or TGF-β signaling components, we then assessed the expression of known progenitor cell markers in liver biopsy specimens following living donor transplantation. Using immunohistochemical labeling, we labeled specimens for Oct3/4, AFP, and CK-19. Oct3/4 is a transcription factor in pluripotent ES cells and has a key role in the maintenance of an undifferentiated state.22,23 AFP is a marker of the hepatocytic cell lineage in the embryonic liver, whereas CK-19 is a marker of the cholangiocytic lineage.3,4

Oct3/4-positive labeling was observed in specimens from all timepoints posttransplantation. In specimens from 1 week, Oct3/4-positive labeling cells were present in a contiguous streaking manner from the central vein, expanding into zone 2 of the liver lobule and diminishing in the periportal region (Figs. 1C, ,2C).2C). In specimens from 6 to 16 weeks posttransplant the percent of Oct3/4-positive labeling cells in zone 3 significantly decreased to nearly zero (P = 0.004) and became concentrated in the periportal region (Figs. 1D, ,2D).2D). The overall percent of Oct3/4-positive cells decreased from 12% in specimens from 1 to 6 weeks to 8% in specimens from 6 to 16 weeks.

Fig. 2
Colocalization of β2SP and TBRII with Oct3/4 and p-Histone-positively labeled cells. (A) p-Histone positively labels proliferating hepatocytes throughout the liver in specimens from 1 week posttransplant. (B) The number of proliferating hepatocytes ...

Similarly, AFP was expressed in specimens from 1 week and 6 to 16 weeks posttransplantation (overall 22%). Like Oct3/4, AFP-positive labeling cells were present in a streaming pattern through the midzone of the liver (zone 2) (Fig. 1E). By 6 to 16 weeks posttransplant, AFP labeling was completely absent in zone 2 or 3 of the liver and localized exclusively to the portal tract (16%), specifically the periductal region (Fig. 1F). By 16 weeks, only 4% of cells were AFP-positive.

CK-19, interestingly, was also expressed in biopsy specimens from 1 week (overall 12%) and 6 to 16 weeks (overall 8%) posttransplant, but was almost exclusively localized to the portal tract (Fig. 1G,H). Moreover, consecutive serial sections from 12-week biopsy specimens labeled for AFP and CK-19 demonstrate colocalization in periductal cells, thereby likely reflecting a progenitor cell compartment.

The similar labeling patterns of Oct3/4 and AFP raised the question of the nature of these positive-labeling cells. Given the lack of CK-19 labeling of these cells, it is unlikely that they represent an expanded population of bipotential liver progenitor cells. Confocal immunofluorescent labeling subsequently demonstrated colocalization of Oct3/4 and p-Histone, a known marker of cell proliferation, thereby suggesting that the Oct3/4/AFP-positive labeling cells are actually proliferating hepatocytes that express progenitor cell markers. In addition, Oct3/4 and p-Histone colocalized with β2SP and the TGF-β signaling component TBRII at all times (Fig. 2). The spatial and temporal expansion of β2SP and TBRII labeling over time in biopsy specimens following living donor transplantation suggests that β2SP and the TGF-β signaling pathway play a role in the “redifferentiation” of hepatocytes to a more differentiated phenotype (Fig. 2I).

β2SP Expression Displays Temporal Variation Following Partial Hepatectomy

In order to further assess the functional role of β2SP in liver regeneration, we subjected β2SP+/- mice and wildtype mice to two-thirds partial hepatectomy. All mice in the wildtype and β2SP+/- groups survived the procedure and there was zero mortality in each group until sacrifice. No gross morphologic differences were noted between wildtype and β2SP+/- mouse livers either at time of initial surgery or upon sacrifice.

Analysis of β2SP expression in wildtype mice demonstrated a similar temporal pattern as seen in regenerating human livers following living donor transplantation. β2SP expression was significantly decreased from baseline within 24 hours posthepatectomy (P < 0.0001) and then increased as regeneration proceeded to completion, peaking at 72 hours posthepatectomy (Fig. 3A). β2SP expression in our β2SP+/- mice was, as expected, significantly depressed in comparison to wildtype at all timepoints (P < 0.05), suggesting that β2SP plays an important functional role in the response to acute liver injury.

Fig. 3
β2SP and Oct3/4 expression in mouse liver from wildtype and β2SP+/- mice subjected to two-thirds partial hepatectomy. (A) β2SP expression in wildtype mice appears to diminish in the early period following hepatectomy and peaks ...

β2SP+/– Mice Have an Expanded Population of Oct3/4-Positive Progenitor Cells Residing in a Progenitor Cell Niche Following Acute Liver Injury

We then assessed the expression of Oct3/4 in regenerating mouse liver by immunohistochemical labeling. Oct3/4-positive labeling was primarily localized to the portal tract and specifically, periductal and bile duct cells (Fig. 3C-J). Given the small numbers of positively labeling cells, we focused exclusively on positively labeling cells within a 50-μm radius of the portal tract. On average, 3,353 ± 32 cells were counted and specifically in wildtype mice the periportal radius consisted of 333 ± 3 cells/hpf, versus 337 ± 6 cells in β2SP+/- mice. Analysis of the portal tracts demonstrated a marked 2 to 4-fold increase in the number of Oct3/4-positive labeling cells in the portal tracts of β2SP+/- mice as compared to wildtype at nearly all timepoints (Fig. 3B). This difference was statistically significant at 24 (14.58 ± 4.6% vs. 29.19 ± 4.3%) and 48 (8.69 ± 2.5% vs. 21.15 ± 5.0%) hours posthepatectomy (P < 0.05).

To further assess whether Oct3/4-positive cells represent hepatic progenitor cells we evaluated the expression of AFP and CK-19 in consecutive serial tissue sections. Like Oct3/4, AFP and CK-19 labeling was also localized to the portal tract and, more specifically, the periductal region (Fig. 3K-M). Oct3/4-positively labeling cells, therefore, likely reside in a progenitor cell niche and may represent an intermediate hepatic progenitor cell. Moreover, the expanded population of progenitor cells in β2SP+/- mice following acute liver injury suggests that β2SP plays a role in hepatic cell differentiation and its loss likely stimulates activation of the progenitor cell compartment.

β2SP+/– Mice Have a Delayed Mitogenic Response Following Partial Hepatectomy

Given the unusual reciprocal relationship between hepatocyte proliferation and hepatic progenitor cell activation, we then assessed the mitogenic response of wildtype and β2SP+/- mice following partial hepatectomy. There was no significant difference in liver mass/body weight ratio between wild-type (mean of 4.8 ± 0.8% over all time periods) and β2SP+/- (mean of 4.0 ± 0.2%) mice at any measured timepoint. Livers were then subjected to immunohistochemical labeling for Ki-67, a known proliferating nuclear antigen in replicating cells, and a nuclear labeling index was determined for each timepoint. Significantly decreased hepatocyte labeling was observed in the β2SP+/- compared to wildtype mice at 48 hours following hepatectomy (35.36 ± 3.4% vs. 4.96 ± 1.4% positively labeled cells) (P = 0.01), with a striking 7-fold difference detected (Fig. 4A-G). By 72 hours, however, there was no significant difference in hepatocyte nuclear labeling between the two groups (25.52 ± 9% vs. 20.11 ± 5.4%) and both groups returned to baseline proliferation state by 7 days posthepatectomy, suggesting that loss of β2SP delays the mitogenic response following partial hepatectomy. These results clearly demonstrate a key role for β2SP in the response to acute liver injury and suggest that delay in the mitogenic response of hepatocytes may activate the progenitor cell compartment and result in an expansion of hepatic progenitor cells.

Fig. 4
Hepatic cell proliferation and Wnt signaling in wildtype and β2SP+/- mice subjected to two-thirds partial hepatectomy. (A-C) Liver sections labeled for the proliferation marker Ki-67 at (A) baseline, (B) 48 hours post-, and (C) 72 hours posthepatectomy ...

Hepatic Progenitor Cells in β2SP+/– Mouse Livers Display Activated Wnt Signaling

To further assess the mechanism by which β2SP+/- mice demonstrate an expanded population of hepatic progenitor cells, we performed a microarray analysis of wildtype and β2SP+/- mouse livers harvested prior to surgery. Analysis demonstrated a significant fold increase in the mRNA levels of several Wnt-related genes, including LRP6, Wnt3a, and Wnt10a (Fig. 4H). The Wnt signaling pathway has been well described to play a critical role in various aspects of liver biology including development, regeneration, growth, and HCC pathogenesis and has been recently shown to play a key role in the activation and proliferation of adult hepatic progenitor cells.24 Analysis of livers from β2SP+/- and wildtype mice following partial hepatectomy by immunohistochemical labeling demonstrated a striking expression of cytoplasmic and nuclear β-catenin in the periductal and bile duct epithelial cells of β2SP+/- mice. Wildtype mice, however, demonstrated β-catenin labeling localized to the membranes of bile duct epithelium (Fig. 4I,J). Similarly, β-catenin labeling of hepatocytes was localized to the membrane in both wildtype and β2SP+/- mice. These results suggest that loss of β2SP results in an expanded population of hepatic progenitor cells following acute injury via a delay in hepatocyte proliferation and that these cells are activated by an activated Wnt signaling pathway.


Hepatic progenitor cell activation has been observed during liver regeneration typically when hepatocyte proliferation is inhibited. Following acute liver injury, as observed following surgical resection or two-thirds partial hepatectomy, however, hepatocytes are the primary driver of cell replenishment and progenitor cells are rarely observed. Little is known of the mechanisms controlling hepatic progenitor cell activation and its relationship to the mature primary cell types of the liver. The present study demonstrates for the first time an important functional role for β2SP in liver regeneration, specifically in the activation of progenitor cells following acute injury, and suggests a critical role in mediating the reciprocal relationship between hepatocyte proliferation and progenitor cell expansion.

We investigated human liver regeneration following living donor transplantation and demonstrated a spatial and temporal expansion of β2SP expression as regeneration proceeds. Overall, β2SP expression by immunohistochemical labeling increased from liver tissue biopsies taken 1 week posttransplant to those taken 6 to 16 weeks posttransplant, at which time the liver has been restored to nearly 85% of the recipient's liver mass.21 This is not unexpected and was similar to the labeling pattern observed for TBRII and is consistent with the role of β2SP as a TGF-β adaptor protein.

The spatial expansion of β2SP expression, initially from the periportal region and then expanding through the midzone toward the central veins during liver regeneration, however, was unexpected and suggests a unique role in the regenerative process. The proliferation of hepatocytes following liver injury advances as a wave of mitoses from the periportal to pericentral areas of the lobule.1 Similar zonation of gene expression in the liver lobule has been described and reflects the compartmentalization and spatial distribution of hepatic cells. For example, transgenic mice carrying an AFP minigene and the AFP promoter demonstrate highest expression of the AFP gene in centrolobular hepatocytes.25 These compartments, although spatially separated, are highly integrated, however, reflecting that positional signals may differentially modulate activation of transcription factors and signal transduction pathways.26 Moreover, the TGF-β receptor type I (TBRI) has been previously described to increase in intracellular concentration in a wavelike fashion from the periportal to the pericentral region of liver lobules following two-thirds partial hepatectomy.27 The spatial expansion of β2SP during liver regeneration suggests that it plays a critical role in hepatic cell proliferation in response to liver injury.

The spatial and temporal expansion of β2SP expression is most significant, however, when associated with the reciprocal expression of several progenitor cell markers, specifically Oct3/4 and AFP. The finding of Oct3/ 4+/AFP-positive cells in regenerating postembryonic human liver is, to our knowledge, new. Moreover, the absence of CK-19 expression and colocalization of Oct3/4 with p-Histone, β2SP, and TBRII suggests that these cells are proliferating hepatocytes that demonstrate progenitor cell-like characteristics.

There is ample evidence that hepatocytes have a “stem cell”-like clonogenic capacity and animal studies demonstrate that as few as 1,000 hepatocytes are necessary to repopulate the liver. Serial transplantation experiments demonstrate that hepatocytes can divide at least 69 times without loss of function.28,29 It is widely accepted that hepatocytes are not terminally differentiated cells1; therefore, the presence of Oct3/4/AFP-positive hepatocytes in regenerating human liver following living donor transplantation likely reflects the progenitor-like character of hepatocytes. More important, however, the colocalization and contraction of this progenitor cell marker expression with β2SP expansion suggests a critical role in hepatic cell differentiation. Loss of β2SP appears to promote expression of a less differentiated phenotype (Fig. 2I).

This hypothesis is confirmed in experiments with our β2SP+/- knockout mice. Following hepatic resection via two-thirds partial hepatectomy, a similar temporal pattern of β2SP expression was observed with diminished levels within the first 24 hours and increasing toward 72 hours posthepatectomy. β2SP+/- mice also demonstrated a strikingly expanded population of Oct3/4-positive cells localized to bile duct and periductal cells in the portal tract at 24-72 hours posthepatectomy. Moreover, these Oct3/4-positive cells share a niche with AFP- and CK-19-positive cells, suggesting that they may reflect an intermediate bipotential hepatic progenitor cell.

Hepatic progenitor cells are rare quiescent cells that share an unusual reciprocal relationship with the proliferation of hepatocytes. There presence in β2SP+/- mice and even following surgical resection suggests that β2SP plays a critical role in progenitor cell activation. Progenitor cells have only been described to become activated and proliferate in contexts in which hepatocyte proliferation is inhibited.3,30 The mechanism underlying this reciprocal relationship, however, has yet to be elucidated. Evidence for the activation of hepatic progenitor cells is seen with our microarray analysis and immunostaining for β-catenin. Up-regulation of several Wnt-related genes and clear cytoplasmic and nuclear β-catenin expression suggest an activated Wnt signaling pathway. Activated Wnt signaling has recently been shown to promote expansion of the progenitor cell population and occurs preferentially within the progenitor cell population.24,31

Evidence that loss of β2SP not only expands hepatic progenitor cells, but also results in a delayed mitogenic response of hepatocytes, suggests that β2SP may also play a critical and non-TGF-β-mediated role in the hepatocyte-progenitor cell interaction. Although inactivation of TGF-β signaling via the type II receptor resulted in an accelerated mitogenic response in conditional knockout mice, loss of β2SP results in an opposite effect. There is no evidence, however, of accelerated apoptosis or significant loss of hepatocyte function, as all mutant mice survived with no significantly discernable morbidity. In fact, hepatocyte proliferation was merely delayed and rapidly corrected in β2SP+/- mice, as there was no evidence of a significant difference in liver mass: body weight ratio 1 week posthepatectomy. Therefore, it is likely that reduced β2SP disrupts the health and proliferative capacity of hepatocytes following acute liver injury, thereby initiating activation of a progenitor cell compartment tasked with aiding the regeneration process (Fig. 5). The lack of complete β2SP loss, however, affords sufficient reserves to allow hepatocyte proliferation to proceed following a delay and allowing for the differentiation of activated progenitor cells to mature hepatocytes as regeneration terminates.

Fig. 5
Schematic representation of the role of β2SP on liver regeneration following hepatic resection. (A,B) Loss of β2SP results in an impaired mitogenic response to liver injury and subsequent expansion of Oct3/4/AFP/CK-19-positive hepatic ...

An important implication of this work is demonstration of the key functional roles of TGF-β signaling and, specifically, β2SP as a mediator of cell proliferation and differentiation. β2SP is a key TGF-β adaptor protein and possesses tumor suppressor function, particularly in HCC. It is clear from the present study, however, that β2SP regulation of liver proliferation, differentiation, and ultimately tumorigenesis is not so straightforward. There is substantial presumptive evidence suggesting that loss of β2SP may promote hepatic progenitor cell activation. This progenitor cell population, on repeated activation following repeated injury, may be more prone to malignant transformation and subsequent tumorigenesis. Definition of the mechanisms by which loss of β2SP regulates hepatic cell proliferation and differentiation may provide valuable insight into the development of novel therapeutics for cancer and future strategies for hepatic cell transplants.


We thank Drs. Yi Tang, Varalakshmi Katuri, and Rupen Amin for excellent technical expertise and help with immunohistochemistry. We also thank Drs. Zhixing Yao, Zhongxian Jiao, and Wilma Jogunoori for critical review and article preparation.

Supported by the National Institutes of Health grants PO1 CA130821 (to L.M.), RO1 CA042857 (to L.M.), RO1 CA106614 (to L.M.), a Veterans Administration Merit Award (to L.M.), the Benn Orr Scholar Award (to L.M.) and the Georgetown Department of Surgery Huffnagel Resident Research fund (to A.T.).


embryonic stem
hepatocellular carcinoma
octamer 3/4
plekstrin homology
signal transducer and activator of transcription 3
TGF-β type II receptor
transforming growth factor beta
tetramethyl rhodamine isothiocyanate


Potential conflict of interest: Nothing to report.


1. Michalopoulos GK, DeFrances MC. Liver regeneration. Science. 1997;276:60–66. [PubMed]
2. Taub R. Liver regeneration: from myth to mechanism. Nat Rev Mol Cell Biol. 2004;5:836–847. [PubMed]
3. Fausto N, Campbell JS. The role of hepatocytes and oval cells in liver regeneration and repopulation. Mech Dev. 2003;120:117–130. [PubMed]
4. Dorrell C, Grompe M. Liver repair by intra- and extrahepatic progenitors. Stem Cell Rev. 2005;1:61–64. [PubMed]
5. Kuwahara R, Kofman AV, Landis CS, Swenson ES, Barendswaard E, Theise ND. The hepatic stem cell niche: identification by label-retaining cell assay. Hepatology. 2008;47:1994–2002. [PMC free article] [PubMed]
6. Paku S, Nagy P, Kopper L, Thorgeirsson SS. 2-Acetylaminofluorene dose-dependent differentiation of rat oval cells into hepatocytes: confocal and electron microscopic studies. Hepatology. 2004;39:1353–1361. [PubMed]
7. Wang X, Foster M, Al-Dhalimy M, Lagasse E, Finegold M, Grompe M. The origin and liver repopulating capacity of murine oval cells. Proc Natl Acad Sci U S A. 2003;100(Suppl 1):11881–11888. [PubMed]
8. Russell WE, Coffey RJ, Jr, Ouellette AJ, Moses HL. Type beta transforming growth factor reversibly inhibits the early proliferative response to partial hepatectomy in the rat. Proc Natl Acad Sci U S A. 1988;85:5126–5130. [PubMed]
9. Braun L, Mead JE, Panzica M, Mikumo R, Bell GI, Fausto N. Transforming growth factor beta mRNA increases during liver regeneration: a possible paracrine mechanism of growth regulation. Proc Natl Acad Sci U S A. 1988;85:1539–1543. [PubMed]
10. Macias-Silva M, Li W, Leu JI, Crissey MA, Taub R. Up-regulated transcriptional repressors SnoN and Ski bind Smad proteins to antagonize transforming growth factor-beta signals during liver regeneration. J Biol Chem. 2002;277:28483–28490. [PubMed]
11. Romero-Gallo J, Sozmen EG, Chytil A, Russell WE, Whitehead R, Parks WT, et al. Inactivation of TGF-beta signaling in hepatocytes results in an increased proliferative response after partial hepatectomy. Oncogene. 2005;24:3028–3041. [PubMed]
12. Watabe T, Miyazono K. Roles of TGF-beta family signaling in stem cell renewal and differentiation. Cell Res. 2009;19:103–115. [PubMed]
13. Zaret K. Genetic programming of liver and pancreas progenitors: lessons for stem-cell differentiation. Nat Rev Genet. 2008;9:329–340. [PubMed]
14. Haramis AP, Begthel H, van den Born M, van Es J, Jonkheer S, Offerhaus GJ, et al. De novo crypt formation and juvenile polyposis on BMP inhibition in mouse intestine. Science. 2004;303:1684–1686. [PubMed]
15. He XC, Zhang J, Tong WG, Tawfik O, Ross J, Scoville DH, et al. BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-beta-catenin signaling. Nat Genet. 2004;36:1117–1121. [PubMed]
16. Tang Y, Katuri V, Dillner A, Mishra B, Deng CX, Mishra L. Disruption of transforming growth factor-beta signaling in ELF beta-spectrin-deficient mice. Science. 2003;299:574–577. [PubMed]
17. Mishra L, Derynck R, Mishra B. Transforming growth factor-beta signaling in stem cells and cancer. Science. 2005;310:68–71. [PubMed]
18. Kitisin K, Ganesan N, Tang Y, Jogunoori W, Volpe EA, Kim SS, et al. Disruption of transforming growth factor-beta signaling through beta-spectrin ELF leads to hepatocellular cancer through cyclin D1 activation. Oncogene. 2007;26:7103–7110. [PubMed]
19. Tang Y, Kitisin K, Jogunoori W, Li C, Deng CX, Mueller SC, et al. Progenitor/stem cells give rise to liver cancer due to aberrant TGF-β and IL-6 signaling. Proc Natl Acad Sci U S A. 2008;105:2445–2450. [PubMed]
20. Mitchell C, Willenbring H. A reproducible and well-tolerated method for 2/3 partial hepatectomy in mice. Nat Protoc. 2008;3:1167–1170. [PubMed]
21. Akamatsu N, Sugawara Y, Tamura S, Imamura H, Kokudo N, Makuuchi M. Regeneration and function of hemiliver graft: right versus left. Surgery. 2006;139:765–772. [PubMed]
22. Niwa H, Miyazaki J, Smith AG. Quantitative expression of Oct-3/4 defines differentiation, dedifferentiation or self-renewal of ES cells. Nat Genet. 2000;24:372–376. [PubMed]
23. Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell. 2006;126:663–676. [PubMed]
24. Yang W, Yan HX, Chen L, Liu Q, He YQ, Yu LX, et al. Wnt/beta-catenin signaling contributes to activation of normal and tumorigenic liver progenitor cells. Cancer Res. 2008;68:4287–4295. [PubMed]
25. Emerson JA, Vacher J, Cirillo LA, Tilghman SM, Tyner AL. The zonal expression of alpha-fetoprotein transgenes in the livers of adult mice. Dev Dyn. 1992;195:55–66. [PubMed]
26. Jungermann K, Kietzmann T. Oxygen: modulator of metabolic zonation and disease of the liver. HEPATOLOGY. 2000;31:255–260. [PubMed]
27. Jirtle RI, Carr BI, Scott CD. Modulation of insulin-like growth factor II/mannose 6-phosphate receptors and transforming growth factor-β1 during liver regeneration. J Biol Chem. 1991;266:22444–22450. [PubMed]
28. Zaret KS, Grompe M. Generation and regeneration of cells of the liver and pancreas. Science. 2008;322:1490–1494. [PMC free article] [PubMed]
29. Overturf K, al-Dhalimy M, Ou CN, Finegold M, Grompe M. Serial transplantation reveals the stem-cell-like regenerative potential of adult mouse hepatocytes. Am J Pathol. 1997;151:1273–1280. [PubMed]
30. Nguyen LN, Furuya MH, Wolfraim LA, Nguyen AP, Holdren MS, Campbell JS, et al. Transforming growth factor-beta differentially regulates oval cell and hepatocyte proliferation. Hepatology. 2007;45:31–41. [PubMed]
31. Apte U, Thompson MD, Cui S, Liu B, Cieply B, Monga SP. Wnt/beta-catenin signaling mediates oval cell response in rodents. Hepatology. 2008;47:288–295. [PubMed]