Terminating hepatocyte proliferation during liver regeneration: The roles of two members of the same family (CCAAT‐enhancer‐binding protein alpha and beta) with opposing actions

Potential conflict of interest: Nothing to report. See Article on Page 315 Liver regeneration is triggered by acute loss of a large number of hepatocytes after surgical resection (modeled by two‐thirds partial hepatectomy [PH] in rats and mice) or acute chemical injury (modeled by injection of CCl4 or acetaminophen). Most of the studies in the last three decades have focused on signaling mechanisms at the beginning of regeneration, associated with triggering of hepatocyte proliferation.1 The preponderance of evidence has shown that proliferating hepatocytes generate signals that trigger proliferation of hepatic nonepithelial cells. Regeneration is a “give and take” arrangement in which hepatocytes provide mitogenic signals for endothelial cells and macrophages (secrete vascular endothelial growth factor and granulocyte‐macrophage colony‐stimulating factor, respectively) and receive hepatocyte growth factor (HGF), interleukin‐6, and tumor necrosis factor alpha in return. Stellate cells produce HGF; fibroblast growth factor 1/2, generated by replicating hepatocytes, is mitogenic for stellate cells. Hepatocytes are the main functional cells of the liver, and the objective of regeneration is to result in a happy ending in which the number of hepatocytes returns to a value sufficient to assure normal hepatic contributions to body homeostasis. In view of the above, it is not surprising that hepatocytes would also play an important role at the other end of regeneration: the termination stage. Liver regeneration studies, from the earliest by Higgins and Anderson in rats to later studies in mice, have shown that most hepatocyte proliferation is over by days 6‐8, whereas liver mass is mostly restored by day 15.4 Study of the signaling involved in initiation of liver regeneration is easier because initiation of regeneration is associated with an acute event (hepatectomy, chemical injury, and so on). The termination phase is not associated with a precisely timed event, and this makes it difficult to pinpoint the associated signals. Adding to the complication is evidence from studies showing that more hepatocytes than needed may be generated by the end of regeneration, and a small wave of apoptosis may be needed to correct the imbalance.5 Several extracellular signals to hepatocytes have been considered as playing a role in the termination stage. Transforming growth factor beta 1 (TGF‐β1), an inhibitor of hepatocyte proliferation, has been often implicated in termination signaling. Immunoreactive TGF‐β1 is gradually removed from periportal toward centrilobular hepatocytes during regeneration.6 However, overexpression of TGF‐β1 does not significantly affect liver regeneration and deletion of its receptors does not extend regeneration unless activin receptors are also involved.7 Extracellular matrix (ECM) is inhibitory to proliferation of hepatocytes in culture. ECM is remodeled and partially degraded during regeneration and is gradually rebuilt at the end of regeneration.1 Disruption of ECM signaling by hepatocyte‐targeted elimination of integrin‐linked kinase results in hepatic enlargement, decrease in hepatocyte ploidy, and increased deposition of ECM by stellate cells. PH in these mice terminates in a liver size much larger than the already larger‐than‐normal liver size observed before hepatectomy.9 Glypican‐3 (GPC3), overexpressed in hepatocellular carcinomas, appears to be a growth suppressor for liver and hepatocytes. Expression of GPC3 increases toward the end of regeneration, and GPC3 transgenic mice have smaller livers at the end of regeneration than the initial size, suggesting that GPC3 and its binding partner, CD81, may play a role in regeneration termination.11 Metabolic changes controlled by the liver, such as plasma glucose levels, may also define liver size and play a role at the beginning and end of regeneration.12 More studies have focused on intracellular signaling mechanisms in hepatocytes. The paper by Jin et al., in this issue of Hepatology,13 provide evidence for a complex interaction between two members of the same family, namely, transcription factors CCAAT‐enhancer‐binding protein (C/EBP)α and C/EBPβ. These two family members appear to have a “difficult” relationship during liver regeneration. Earlier studies by Geenbaum et al. had shown that C/EBPα decreases after PH, whereas there is a pronounced increase in C/EBPβ, with the β/α ratio increasing 7‐fold during the early proliferative stage of regeneration and being restored toward the end of the regenerative process.14 The study by Jin et al. utilizes “knock‐in” mice, in which both the normal genes for C/EBPα are replaced by genes in which serine 193 (whose phosphorylation is required for binding of C/EBPα to the cofactor, P300, and further binding to specific chromatin sites) is replaced by alanine, thus rendering C/EBPα unable to perform its functions in gene regulation. Levels of C/EBPα are, in part, controlled by an autoregulation loop in which C/EBPα binds to its own promoter. In the absence of active and functioning C/EBPα in the knock‐in used in this study, the “evil twin,” C/EBPβ, takes over and prevents the reemergence of C/EBPα toward the end of regeneration. This has serious consequences, because C/EBPβ, bound to histome deacetylase 1, also represses expression of mutiple intracellular regulators of hepatocyte biology, including sirtuin 1, peroxisome proliferator‐activated receptor‐gamma coactivator 1 alpha, farnesoid X receptor, p53, and telomerase reverse transcriptase. In the context of an imbalance between a normally operating C/EBPβ and a deactivated C/EBPα, hepatocytes make an earlier entry into the cell cycle and undergo prolonged and sustained proliferation after PH, at relatively high levels up to and beyond the fourth day of regeneration. The net result is a liver of a larger size (approximately 130%) than the original (Fig. 1). Of interest, the earlier entry of hepatocytes into the cell cycle is also observed after CCl4 chemical injury of the knock‐in mice, but the liver size does not increase beyond normal at the end of repair of the injury.Figure 1: In a normal liver regeneration, C/EBPβ and α exert opposing effects, with C/EBPβ favoring hepatocyte proliferation and C/EBPα having an opposite effect. The ratio of β to α increases 7‐fold after PH as a result of enhanced expression of C/EBPβ and decreased expression of C/EBPα. The ratio is restored to normal toward the end of liver regeneration. If C/EBPα becomes inactivated, the effect of C/EBPβ is unopposed, resulting in prolonged liver regeneration and liver size exceeding that of the original (prehepatectomy) level.The importance of the functions of C/EBPα and β for liver and hepatocytes appears during embryogenesis. Suppression of C/EBPα signaling in periportal hepatoblasts is essential for normal biliary development.15 The opposing effects of C/EBPα and β are focused on regulation of expression of TGFβR2.17 Germline elimination of C/EBPα causes immediate (within 8 hours) postnatal mortality as a result of severe hypoglycemia, resulting from an inability of the liver to synthesize and store glycogen.18 At the other end of the life spectrum, in aged mice, glycogen synthase kinase 3 beta (GSK3β) controls and regulates the effects of C/EBPα. The decrease in expression of GSK3β in aging mice, resulting in increased effectiveness of C/EBPα, has been associated with the decreased regenerative capacity of the liver observed in aged mice.19 Overall, the article by Jin et al. expands the importance of C/EBPα and β in hepatocyte and liver biology, extending now beyond metabolic functions into regulation of termination of liver regeneration and regulation of liver size, a complex phenomenon known as “hepatostat.” The interactions between the C/EBP “twins” are likely to be shown to have a major role to play in our understanding of hepatic contributions to systemic homeostasis.