NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Madame Curie Bioscience Database [Internet]. Austin (TX): Landes Bioscience; 2000-2013.

Cover of Madame Curie Bioscience Database

Madame Curie Bioscience Database [Internet].

Show details

Transcriptional Profiling of the Hepatic Growth Response

*.

* Linda Greenbaum-Gastrointestinal Division, Department of Medicine, University of Pennsylvania School of Medicine, 675 CRB, 415 Curie Boulevard, Philadelphia, Pennsylvania 19104, U.S.A. Email: greenbal@mail.med.upenn.edu

The cells in the adult liver retain the capacity to proliferate in response to a loss of liver mass. The partial hepatectomy model in rodents has been extensively used to investigate the mechanisms responsible for hepatic growth and proliferation and is the best in vivo model of synchronous cell cycle progression in mammals. Many of the individual genes involved in the hepatic growth response have been identified using mouse genetic models. However, as a consequence of the complexity of the transcriptional events that occur in response to partial hepatectomy, there remains much to be learned regarding the integration of the products of these genes into regulatory networks that result in the restoration of liver mass and maintenance of homeostasis. Computational approaches have led to the characterization of biologically relevant networks and the identification of transcription factor targets involved in this highly synchronized in vivo growth model.

The molecular analysis of mammalian cellular proliferation in vivo is limited in most organ systems by the low turnover and/or the asynchronous nature of cell cycle progression. A powerful approach to the study of hepatic growth regulation is afforded by the fact that cells in the adult liver retain the capacity to proliferate in response to a loss of liver mass. The partial hepatectomy model in rodents has been extensively used to investigate the mechanisms responsible for hepatic growth and proliferation and is the best in vivo model of synchronous cell cycle progression in mammals. Progenitor cells, often referred to as oval cells, do not play a major role in this growth response, instead, the existing quiescent hepatocytes and nonparenchymal cells (cholangiocytes, Kupffer, endothelial and stellate cells) reenter the cell cycle and divide.1

Within 30 minutes of the surgery, which typically entails removal of 70% of the liver, cytokine and growth factors including TNF-α, IL-6, HGF, and TGF-α are rapidly activated and/or secreted in the remnant liver. These signaling molecules subsequently activate immediate-early growth genes including myc, fos and jun. Mitogen activated protein kinases including extracellular-regulated signal kinase (ERK) and Jun kinase (JNK) are also rapidly activated after partial hepatectomy. These molecules then initiate signaling cascades in hepatocytes that rapidly activate several preexisting transcription factors including NF-κB, STAT3 and C/EBPβ via posttranslational modifications. The activation of both preexisting and de novo synthesized transcription factors is responsible for the induction of a transcriptional program responsible for stimulating normally quiescent hepatocytes and nonparenchymal liver cells to reenter the cell cycle and ultimately restore the liver mass.2-5

Many of the individual genes involved in the hepatic proliferative response have been identified using mouse genetic models (reviewed in refs. 2, 4, 6-9) and more recently through the use of high density oligonucleotide and cDNA arrays.10-16 However, as a consequence of the complexity and robust nature of the transcriptional changes that occur in response to partial hepatectomy, there remains much to be learned regarding the regulatory networks that result in this highly synchronized regenerative response. High-density cDNA and oligonucleotide arrays have made it feasible to simultaneously analyze genome-wide changes in expression of thousands of genes. The challenges resulting from this explosion of information have been to integrate these data into transcriptional networks and global biological programs. The contribution of computational approaches to the analysis of hepatic regenerative response will be the focus of this review.

Identification of Global Programs Important for Homeostatic and Proliferative Response in the Regenerating Liver

Coordinated changes in gene regulatory networks involved in growth or homeostatic functions are essential for restoration of liver mass and survival during the post-hepatectomy period. Cluster analysis has been used in the majority of published array studies to identify global changes in biological functions that occur in the regenerating liver.12-15,17 This approach has demonstrated an overall trend towards increased expression of genes involved in protein synthesis and a reduction in most aspects of intermediary metabolism.13,14 Our laboratory has utilized the recently developed Expression Analysis Systematic Explorer (EASE) program to identify additional biologically relevant themes important for homeostatic and proliferative responses in the regenerating liver during multiple phases of the cell cycle.18 Using the EASE program, we identified a shift in the transcriptional program from genes involved in lipid and hormone biosynthesis in the quiescent liver to those contributing to cytoskeleton assembly and DNA synthesis during the posthepatectomy period16 (Fig. 1). The results of EASE analysis emphasize the importance of coordinated shifts in transcriptional sub-programs that control discrete biological programs in the regenerating liver. The mechanisms responsible for synchronized transcription of genes in the partial hepatectomy model have not been investigated because the genetic analysis of the contribution of transcription factors has been confined to individual regulatory proteins and a small set of individual target genes. The importance of coactivator proteins in other mammalian systems for integration of transcriptional responses to physiologic signals suggests that coactivators and perhaps corepressors may be involved in coordinating transcription of genes that together regulate complex biological processes in the regenerating liver (ref. 19 and references therein). Identification of promoters bound by specific coactivator proteins using arrays containing promoter sequences (see below: Location analysis) promises to provide new insights regarding the contribution of coactivators for coordinating transcriptional programs responsible for either growth or metabolic processes in this system in the future. However, this approach will require the development of new experimental and computational tools, because coactivators and corepressors do not contact DNA directly and therefore do not converge on a single consensus binding site.

Figure 1. Biological themes over-represented during liver regeneration after partial hepatectomy.

Figure 1

Biological themes over-represented during liver regeneration after partial hepatectomy. A schema summarizing important changes in the gene expression of major biological processes during liver regeneration were identified using the software application (more...)

Another important question in the field of liver regeneration is to identify the signals that are responsible for the termination of the regenerative process. It is a striking property of these processes that proliferation ceases once the original liver/body weight ratio has been restored. Experimental evidence supports the notion that normalization of metabolic function is important for cessation of liver growth.2 While it is not known to what extent these signals may be regulated at the transcriptional level, it is conceivable that important new insights regarding the signals that terminate regeneration may be obtained from transcriptional profiling at these later time points. There are no published microarray data available investigating these later events in the hepatectomy model. Application of EASE and similar programs may be useful for identifying shifts in metabolic programs that occur during the termination phase of hepatic regeneration. Understanding the signals that terminate regeneration during physiologic growth may provide important insights regarding how deregulation of these pathways contributes to hepatocarcinogenesis.

Identification of Biologically Relevant Networks

The regenerative response in the partial hepatectomy model can be divided into several phases based on patterns of gene expression and function. The first of these phases, “the priming phase” was initially proposed by Fausto and colleagues as the period during which quiescent hepatocytes are induced to reenter the cell cycle in response to TNF-α and IL-6 stimulation. Many of the genes induced during the priming phase are grouped in the category of “immediate-early genes” based on their rapid induction and the fact that their transcription is independent of de novo protein synthesis. Conventional approaches in the prearray era led to the identification of a number of proto-oncogenes, transcription factors, and metabolic proteins that are rapidly and dramatically induced during this initial phase of hepatic regeneration (reviewed in refs. 2, 4, 6, 7). Schultz and colleagues utilized high density oligonucleotide arrays to characterize the gene expression profiles during this initial phase of hepatocyte proliferation.10 The results of this study not only confirmed the work of previous investigators demonstrating the induction of proto-oncogenes, pro-inflammatory, and anti-apoptotic genes during the priming phase but also identified additional genes not previously known to be activated during this period. These novel genes include several that are involved in cytoskeletal and matrix remodeling and regulation of cell cycle reentry. As predicted, the majority of the genes identified in this study are pro-proliferative. Although previous studies had demonstrated that both pro-proliferative and cell cycle inhibitory genes are induced during the priming phase, this study demonstrated that several genes involved in cell cycle checkpoint regulation and arrest including GADD45, TIS21 (involved in p53 growth arrest) and p21 were induced within 10 minutes of the hepatectomy stimulus and continued to increase throughout the four hour priming phase. These findings suggest that transcriptional control of the proliferative response begins almost immediately after hepatectomy and that checkpoint mechanisms are an important component of the initial stages of hepatic regeneration.

The priming phase is followed by a period in which “delayed early genes” are induced. The activation of these genes is thought to represent a secondary transcriptional response as this phase of gene induction their can be blocked by protein synthesis inhibitor.20,21 Additional stages of gene expression correspond to the late G1/S phase which occurs 24h posthepatectomy in the rat and 36-40h posthepatectomy in the mouse followed by the G2 and M phases, with the peak of M phase occurring approximately 24 hours after S phase. There is much less information available regarding the changes in gene expression that occur at stages of hepatic regeneration beyond the initial priming phase. In addition, very little is known regarding how individual genes are integrated in specific regulatory and signaling networks. The majority of published microarray analyses have not detected changes in expression of many growth factors and DNA binding proteins in the partial hepatectomy model and there have been only a few studies in which differences in cell cycle associated genes have been reported beyond the priming phase.16,17 Computational pathway analysis programs, exemplified by the Ingenuity Pathways analytic tool, a web-delivered application, have facilitated the identification of specific regulatory networks at the G1/S phase transition in the hepatectomy model from lists of differentially expressed genes. Application of this computational tool also helps to overcome two problems of microarray expression profiling, which are its limited sensitivity and the absence of certain genes on a given array platform. For example, application of the Ingenuity program to the analysis of genes differentially expressed during the priming phase, mid-G1 and S phases allowed us to identify genes not previously identified as differentially induced in this model such as MAP kinase phosphatase 3 (early G1) and several proteins involved in cytoskeletal assembly (mid G1). Pathway analysis at the peak of hepatocyte S phase yielded networks involved in DNA replication, mitotic spindle assembly and mitotic checkpoint control16 (Fig. 2). An additional advantage of pathway analysis is that this computational tool can also include genes in a regulatory network that are modified post-translationally or as a result of protein-protein interactions. Obviously, such genes will not appear on any list of differentially expressed genes, as microarray expression profiling can only identify changes in mRNA levels. However, because pathway analysis integrates all published information on thousands of proteins, its use will result in the integration of these additional regulatory proteins. A specific example of this would be the inclusion of a transcriptional regulatory factor that is activated by phosphorylation. If a number of its targets are induced during a biological process, pathway analysis will include the transcription factor as a node of regulation.

Figure 2. Ingenuity pathway analysis identifies networks of genes regulated during S phase in the liver in vivo.

Figure 2

Ingenuity pathway analysis identifies networks of genes regulated during S phase in the liver in vivo. The network is displayed graphically as nodes (genes/gene products) and edges (the biological relationships between the nodes). Edges are displayed (more...)

Rudolph and colleagues also investigated the patterns of gene expression that occur at the G1/S transition in the partial hepatectomy model. Although the constellation of genes detected in their analysis overlapped only partially with those identified in our study, both studies identified differentially induced expression of genes at G1/S time points that encode proteins known to participate in the G2/M transition.17 These findings underscore the requirement for coordinate regulation of cytoskeletal and chromosomal components of mitotic complexes at the mRNA level prior to the G2/M phase transition.

Genome-Wide Location Analysis as a Tool to Identify Transcription Factor Targets

Genome-wide location analysis is a relatively novel functional genomics tool that was pioneered by the lab of Richard Young. In a seminal publication in 2000, his laboratory combined location analysis using a yeast promoter array with expression profiling in order to identify direct transcription factor targets.22 Location analysis involves chromatin immunoprecipitation using an antibody specific to the DNA binding protein of interest. Immunoprecipitated chromatin is then purified, the DNA amplified and fluorescently labeled and hybridized to a microarray containing promoter sequences of multiple genes of interest. In contrast to conventional chromatin immunoprecipitation assays in which primers must be designed corresponding to known promoter sequences, location analysis can identify previously unknown promoters that may be bound by a DNA binding protein of interest. Recently, this approach has been adapted to mammalian promoters, and several thousand proposed targets of hepatic DNA binding proteins have been identified, though not confirmed.23 The combined approaches of location analysis and expression profiling represents a powerful tool with which to identify genes that are both bound by and dependent on a specific DNA binding protein for proper induction. This approach, known as orthogonal analysis, was used to identify biologically relevant targets of the bZIP transcription factor, CCAAT enhancer binding protein beta (C/EBPβ) in the regenerating liver. Using this technology, we identified a series of genes not previously identified as in vivo targets of C/EBPβ. For some genes, both promoter occupancy by C/EBPβ and mRNA expression levels were reduced in C/EBPβ-/- livers posthepatectomy, demonstrating that they are functionally relevant targets of C/EBPβ in this model. However, several of the genes that were bound by C/EBPβ were nevertheless expressed at normal levels in the livers of C/EBPβ null mice, illustrating the redundancy present in complex regulatory networks such as the regenerating liver. The combined analysis of promoter occupancy and expression profiling also led to some unexpected observations. The previous applications of computational sequence analysis to identify C/EBPβ target promoters had been impeded by the fact that significant variations from the optimal C/EBP binding sequence are tolerated in the positional weight matrices employed by programs such as TRANSFAC or TESS. In addition, the C/EBP consensus sequence appears at a high frequency in the mammalian genome, limiting its discriminative power. The high frequency of C/EBPβ-binding sites predicted by computational analysis and in vitro DNA binding assays such as electrophoretic mobility shift assay suggested to us that additional features must distinguish bona fide in vivo C/EBPβ target genes. We speculated that other transcription factors or even nucleosomes bound in the region surrounding the C/EBPβ site might be important for C/EBPβ binding specificity. Our discovery of several genes bound in vivo by C/EBPβ provided an opportunity to identify some of these sites. Using all 674 transcription factor-binding sites in the TRANSFAC data base, we performed a computational search for sites that are bound more frequently in the set of C/EBPβ bound genes than in genes that were neither bound by C/EBPβ in our location analysis nor dependent on C/EBPβ for expression. Using this approach, we identified the IFN-stimulated response element, the peroxisome proliferator-activated receptor direct-repeat 1 site, and the albumin D-site binding protein consensus sites as the highest ranking cis-regulatory elements. Higher-order combinations of a C/EBPβ site and these three cis-regulatory elements were present at a significantly increased frequency in promoters bound by C/EBPβ in vivo relative to the total sequences present on the mouse promoter microarray. Thus, location analysis followed by computational evaluation can be used to determine the binding of multiple additional tissue-specific factors.

Coactivator and corepressor proteins are responsible for regulating transcription of large groups of genes that together are responsible for effecting complex biological processes, such as occurs during the posthepatectomy period.19 Although several coactivators have been linked to transcriptional activation of individual genes that are known to be induced in the regenerating liver,24-26 genome-wide approaches may provide more global information regarding how these coactivators regulate transcription of complex networks of genes involved in proliferative and/or metabolic responses to partial hepatectomy. For example, location analysis may be helpful for identifying target promoters bound by specific coactivators. In addition, computational analysis could be used to search for sites that are bound more frequently in the set of genes bound by a coactivator vs. unbound promoters and in this way, identify candidate transcription factors that may be regulated by a specific coactivator. The highly synchronous nature and complex characteristics of the partial hepatectomy model make it a highly attractive choice for combined functional genomics and computational approaches for the identification of coactivator target genes and associated transcription factors.

Future Directions

One limitation of transcriptional profiling of the entire liver is that the liver is not a homogenous organ, i.e., in addition to the predominant hepatocyte population which comprises approximately 60% of the liver cells, the nonparenchymal cells including endothelial, biliary, Kupffer and stellate cells must also proliferative to restore the original mass of the organ after hepatectomy. Because each of the individual nonparenchymal cell populations constitutes only a small percentage of the total liver, it is likely that the majority of differentially expressed genes detected by transcriptional profiling are hepatocyte in origin. With the advent of laser capture microscopy, it should be possible to isolate these nonparenchymal cell populations in regenerating livers and increase the sensitivity of detecting changes in gene expression in highly enriched mRNA fractions. These studies will be critical for defining the transcriptional changes that are involved in the proliferation and remodeling of blood vessels, biliary structures, stellate cells and immune cells during liver regeneration.

Another areas that requires further investigation is the identification of transcriptional networks in other models of hepatic regeneration such as those involving hepatic progenitor cell proliferation.2,6,27 Progenitor cells may provide an additional source of cells that may restore the liver mass following recovery from fulminant hepatic failure in patients. Therefore understanding the mechanisms that regulate their proliferation and differentiation may be useful for future cell replacement therapies.

References

1.
Higgins GM, Anderson RM. Experimental pathology of the liver. I. Restoration of the liver of the white rat following partial surgical removal. Arch Pathol. 1931;12:186–202.
2.
Fausto N. Liver Regeneration. J Hepatol. 2000;32(suppl 1):19–31. [PubMed: 10728791]
3.
Kim TH, Mars WM, Stolz DB. et al. Extracellular matrix remodeling at the early stages of liver regeneration in the rat. Hepatology. 1997;26(4):896–904. [PubMed: 9328311]
4.
Koniaris LG, McKillop IH, Schwartz SI. et al. Liver Regeneration. J Am Coll Surg. 2003;197(4):634–659. [PubMed: 14522336]
5.
Crissey MA, Leu JI, De Angelis RA. et al. Liver-specific and proliferation-induced deoxyribonuclease I hypersensitive sites in the mouse insulin-like growth factor binding protein-1 gene. Hepatology. 1999;30:1187–1197. [PubMed: 10534340]
6.
Taub R. Liver regeneration: From Myth to mechanism. Nat Rev Mol Cell Biol. 2004;5:837–847. [PubMed: 15459664]
7.
Michalopoulos GK, DeFrances MC. Liver regeneration. Science. 1997;66:60–66. [PubMed: 9082986]
8.
Borowiak M, Garratt AN, Wustefeld T. et al. Met provides essential signals for liver regeneration. Proc Natl Acad Sci USA. 2004;101(29):10608–10613. [PMC free article: PMC490025] [PubMed: 15249655]
9.
Matsuo T, Yamaguchi S, Mitsui S. et al. Control mechanism of the circadian clock for timing of cell division in vivo. Science. 2003;302(5643):255–259. [PubMed: 12934012]
10.
Su AI, Guidotti LG, Pezacki JP. et al. Gene expression during the priming phase of liver regeneration after partial hepatectomy in mice. Proc Natl Acad Sci USA. 2002;99(17):11181–11186. [PMC free article: PMC123230] [PubMed: 12177410]
11.
Li W, Liang X, Leu JI. et al. Global changes in interleukin-6-dependent gene expression patterns in mouse livers after partial hepatectomy. Hepatology. 2001;33(6):1377–1386. [PubMed: 11391526]
12.
Fukuhara Y, Hirasawa A, Li XK. et al. Gene expression profile in the regenerating rat liver after partial hepatectomy. J Hepatol. 2003;38:784–792. [PubMed: 12763372]
13.
Arai M, Yokosuka O, Chiba T. et al. Gene expression profiling reveals the mechanism and pathophsyiology of mouse liver regeneration. J Biol Chem. 2003;278(32):29813–29818. [PubMed: 12782638]
14.
Locker J, Tian J, Carver R. et al. A common set of immediate-early response genes in liver regeneration and hyperplasia. Hepatology. 2003;38(2):314–325. [PubMed: 12883475]
15.
Kelley-Loughnane N, Sabla GE, Ley-Ebert C. et al. Independent and overlapping transcriptional activation during liver development and regeneration in mice. Hepatology. 2002;35:525–534. [PubMed: 11870364]
16.
White P, Brestelli JE, Kaestner KH. et al. Identification of transcriptional networks during liver regeneration. J Biol Chem. 2005;280(5):3715–3722. [PubMed: 15546871]
17.
Satyanarayana A, Geffers R, Manns MP. et al. Gene expression profile at the G1/S transition of liver regeneration after partial hepatectomy in mice. Cell Cycle. 2004;3(11):1405–1417. [PubMed: 15492508]
18.
Hosack DA, Dennis Jr DG, Sherman BT. et al. Identifying biological themes within lists of genes with EASE. Genome Biol. 2003;4(10):R70. [PMC free article: PMC328459] [PubMed: 14519205]
19.
Spiegelman BM, Heinrich R. Biological control through regulated transcriptional coactivators. Cell. 2004;119(2):157–167. [PubMed: 15479634]
20.
Taub R. Expression and function of growth-induced genes during liver regeneration. In: Jirtle RL, ed. Liver Regeneration and Carcinogenesis. San Diego: Academic Press. 1995:71–97.
21.
Du K, Leu JI, Peng Y. et al. Transcriptional up-regulation of the delayed early gene HRS/SRp40 during liver regeneration. Interactions among YY1, GA-binding proteins, and mitogenic signals. J Biol Chem. 1998;273(52):35208–35215. [PubMed: 9857059]
22.
Young SS, Liston P, Xuan JY. et al. Genomic organiziation of the physical map of the human inhibitors of apoptosis: HIAP1, HIAP2. Mamm Genome. 1999;10(1):44–48. [PubMed: 9892732]
23.
Odom DT, Zizlsperger N, Gordon DB. et al. Control of pancreas and liver gene expression by HNF transcription factors. Science. 2004;303(5662):1378–1381. [PMC free article: PMC3012624] [PubMed: 14988562]
24.
Herzig S, Long F, Jhala US. et al. CREB regulates hepatic gluconeogenesis through the coactivator PGC-1. Nature. 2001;413(6852):179–183. [PubMed: 11557984]
25.
Wang XL, Herzog B, Waltner-Law M. et al. The synergistic effect of dexamethasone and all-trans-retinoic acid on hepatic phosphoenolpyruvate carboxykinase gene expression involves the coactivator p300. J Biol Chem. 2004;279(33):34191–34200. [PubMed: 15166231]
26.
Major ML, Lepe R, Costa RH. Forkhead box M1B (FoxM1b) transcriptonal activity requires binding of Cdk/Cyclin complexes for phosphorylation-dependent recruitment of p300/CBP coactivators. Mol Cell Biol. 2004;24:2649–2661. [PMC free article: PMC371108] [PubMed: 15024056]
27.
Fausto N. Liver regeneration and repair: Hepatocytes, progenitor cells, and stem cells. Hepatology. 2004;39(6):1477–1487. [PubMed: 15185286]
Copyright © 2000-2013, Landes Bioscience.
Bookshelf ID: NBK6327

Views

  • PubReader
  • Print View
  • Cite this Page

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...