Hepatoblastoma (HB) is a greatly malignant pediatric liver neoplasm usually diagnosed before the age of 4 years (1). Although rare, the incidence of HB has increased during the last 3 decades, now being 1/1 million children younger than 15 years of age (2,3). Histologically, HBs can be divided into epithelial and mixed epithelial-mesenchymal subtypes (4). Serum α-fetoprotein is used to support the diagnosis based on the histological assessment of a biopsy or resection specimen. HB only occurs in children, whereas hepatocellular carcinoma (HCC) is the most common malignant liver tumor in adults. In children, the distinction between HB and HCC, the second most common pediatric liver tumor, may be difficult.
In Finland, hereditary tyrosinemia type I (HTT-I), caused by deficiency of the enzyme fumaryl acetoacetate hydrolase, is the most common metabolic disorder leading to abnormal liver function (5). HTT-I ultimately causes liver cirrhosis and increased regeneration of liver tissue; the regeneration nodules are preneoplastic and have a high risk of transforming into HCC if early liver transplantation is not performed.
GATA-4 belongs to a 6-member family of zinc finger DNA-binding proteins that regulate cell growth and differentiation in various tissue types. GATA-1, GATA-2, and GATA-3 are essential regulators of hematopoiesis, whereas GATA-4, GATA-5, and GATA-6 are predominantly expressed in the tissues of endodermal origin (7). GATA-4 is considered crucial for appropriate embryonic folding and proper cardiac morphogenesis (8). Additionally, GATA-4 is thought to be among the first transcription factors promoting endodermal differentiation, and it has an important role in the development of mammalian liver (9). In the murine fetal liver, GATA-4 is detected in the septum transversum during the liver bud formation at e9.5 to e10, after which it is downregulated in hepatocytes and detected postnatally only in the biliary duct epithelial cells (10). In humans, GATA-4 is expressed in hepatocytes during early gestation, but later on it is expressed only in the endothelial cells surrounding the hepatic vessels as well as in Kuppfer cells (11). The expression of GATA-6 in the mouse hepatocytes starts at e9.5 and continues until late gestation beyond the period the cells have ceased to express GATA-4 (10). GATA-6 is essential for visceral endoderm differentiation of the mouse embryo (12), and it is required for liver bud expansion and normal expression of hepatic-specific genes during development (13).
Erythropoietin (Epo) is a hematopoietic growth factor that regulates differentiation of the erythroid progenitor cells into mature blood cells. During embryogenesis, the Epo gene is expressed in the liver, specifically in hepatocytes and Kuppfer cells. The hepatic expression of Epo thus coincides with that of GATA-4, which is suggested to regulate Epo gene expression (11). In addition to its role in erythropoiesis, Epo is suggested to be associated with tumor malignancy and angiogenesis in HCC (14). Of interest, HB cells also have been reported to produce erythropoietin in vitro (15).
GATA-4 expression has been indicated in human cell lines derived from HCCs (hepatomas) (16,17). There are, however, no surveys on GATA-4 or its family members in childhood hepatic tumors. GATA-4 is indicated in the regulation of fetal liver development, and its expression pattern during fetal and postnatal life parallels that of Epo (11). Interestingly, Epo has been related to hematopoietic activity seen in pediatric liver tumors (14). Moreover, GATA-4 (18) and Epo, its putative target gene in the fetal liver (14,19–22), have been connected to tumorigenesis of many human organs. Based on these data, we studied the expression of GATA-4, together with its close family member GATA-6, and Epo, in childhood liver tumors and HTT-I predisposing to early malignant transformation in the liver.
PATIENTS AND METHODS
Human tissue samples from patients with tumors were obtained from Helsinki University Hospital's archives of pathology. The study group included 4 fetuses (gestational age 12–36 weeks postconception) and 38 patients with diagnosed HB, HCC, or hereditary tyrosinemia type I (ages 0.2–75 years) (Table 1). The use of the samples in the present study was approved by the ethical committee of Helsinki University Hospital and the Finnish National Authority of Medicolegal Affairs and Health. The samples were paraffin blocks or fresh frozen tissues stored at −80°C for further use. Paraffin blocks were cut into 5-μm thick slides for immunohistochemistry. Fresh tissue samples, obtained from 5 patients with HB and 2 patients with childhood HCC, were subjected to RNA and protein extraction.
Procedures involving C57BL/6 mice were approved by the institutional animal use and care committee and were in accordance with the National Research Council's Guide for the Care and Use of Laboratory Animals. All of the mice had ad libitum access to water and standard rodent chow and were exposed to 12-hour light and dark photoperiods.
Immunoperoxidase staining of the paraffin-embedded tissue sections followed the protocol previously described (23). A commercially available polymerized reporter enzyme staining system was used to visualize bound antibody (ImmPRESS reagent kit; Vector Laboratories, Burlingame, CA). The primary antibodies were goat anti-mouse GATA-4 immunoglobulin (Ig) G (sc-1237, Santa Cruz Biotechnology, Santa Cruz, CA) at dilution 1:200 (for mouse tissue), goat anti-human GATA-4 IgG (AF2606, R&D Systems, Minneapolis, MN) at dilution 1:200 (for human tissue), rabbit anti-human GATA-6 (sc-9055, Santa Cruz Biotechnology) at dilution 1:200, rabbit anti-human erythropoietin (sc-7956, Santa Cruz Biotechnology) at dilution 1:100, and mouse anti-human KI-67 (sc-15402, Santa Cruz Biotechnology) at dilution 1:50. Primary antibodies were incubated overnight at 4°C. In control experiments, nonimmune serum replaced the primary antibody.
The human HUH6 HB cell line was obtained from Health Science Research Bank, Osaka, Japan. HUH6 cells were cultured as previously described (24).
For adenoviral infections, HUH6 cells were plated on 6-well plates, 96-well plates, or 8-well permanox chamber slides 24 hours before transfection. The cells were infected by incubating them with viruses in 1 mL Dulbecco's Modified Eagle Medium containing 0% fetal bovine serum. The replication-deficient adenoviral constructs expressing wild-type (WT) rat GATA-4 and dominant negative (DN) GATA-4 (engrailed repressor domain, GATA-4 fusion with flag epitope) have been previously described (25,26). Multiplicity of infection used was 40 (WT) or 200 (DN). After 1 hour, Dulbecco's Modified Eagle Medium with 10% fetal bovine serum was added to stop the infection. Cells were harvested or fixed after 48, 96, or 120 hours.
RNA Extraction and Reverse Transcription-Polymerase Chain Reaction (RT-PCR)
Total RNA and protein were extracted from the HUH6 cells and human tissue samples, and genomic DNA was eliminated as previously described (27). First-strand cDNA synthesis was performed from 1 μg total RNA using SYBR Green RT-PCR reagents and random hexamers (Applied Biosystems, Foster City, CA). The primers were designed using the primer software provided by National Center for Biotechnology Information (Bethesda, MD). The primers used were GATA-4: F:(5′-CTCCTTCAGGCAGTGAGAGC-′) and R:(5′-GAGATGCAGTGTGCTCGTGC-3′), GATA-6: F:(5′-ATGACTCCAACTTCCACCTCT-3′) and R:(5′-CAGCCTCCAGAGATGTGTAC-3′).
Western blotting was performed as previously described (27). Primary antibodies used were goat anti-human GATA-4 IgG (sc-1237, Santa Cruz Biotechnology) at dilution 1:1000, rabbit anti-human GATA-6 IgG (sc-9055, Santa Cruz Biotechnology) at dilution 1:500, and goat antihuman β-actin IgG (sc-1616, Santa Cruz Biotechnology) at dilution 1:10000.
Enzyme-linked Immunosorbent Assay For Epo
The EPO enzyme-linked immunosorbent assay detection kit was used for the detection of erythropoietin from cell culture media following the manufacturer's instructions (Bender MedSystems, Vienna, Austria). The samples were analyzed on a 96-well plate with control samples and calibrators provided by the manufacturer. The biotinylated Epo antibody was detected with a peroxidase-labeled antibody. After rinsing the wells in a saline wash concentrate, tetramethylbenzidine was added to the wells. The reaction was terminated with sulfuric acid, and the optical densities were measured spectrophotometrically at 450 nm.
Caspase-3/7 activation was measured using Caspase-Glo 3/7 assays (Promega Corporation, Madison, WI) as previously described (28).
Cell proliferation was assessed by measuring the incorporation of 5-bromo-2′-deoxyuridine (BrdU), a thymidine analog, into the newly synthesized DNA of replicating cells. Staining procedure was performed according to the manufacturer's instruction (Invitrogen, Carlsbad, CA).
GATA-4 Is Not Expressed in Normal Human Fetal Hepatocytes After Week 12 of Gestation
Given that GATA-4 is expressed in murine fetal liver (10), we wanted to examine its expression in human liver during fetal development. In the sections studied, GATA-4 protein was not detected in the liver parenchyma of fetuses ages 12 weeks or older, whereas control sections of fetal heart showed abundant GATA-4 expression. Our results contrast an earlier report on the expression of GATA-4 in human fetal liver during weeks 13 to 18 of gestation (29). Instead, GATA-4 expression was restricted to bile duct epithelial cells and Kuppfer cells in both fetal and normal postnatal liver tissue (Fig. 1A, B). This expression pattern is analogous to that in mouse, in which the initial GATA-4 expression in liver parenchyma diminishes after e11.5 (Fig. 1E–G).
Also, GATA-6 has been shown to have an important role in liver development (13) and it has been detected in fetal murine liver in early development (10). Therefore, we were interested in its expression during human liver development. Similarly to GATA-4, GATA-6 protein was not detected in any of our fetal liver samples collected at 12 weeks of gestation or later.
GATA-4 Is Abundantly Expressed in Childhood HBs and HCCs
GATA-4 expression was examined by immunohistochemistry in human liver tumor samples. Eleven of 14 HBs and 3 of 4 childhood HCCs showed strong nuclear GATA-4 positivity (Table 1), but only 1 of 10 adult HCCs was weakly GATA-4 positive; the remainder were negative (Fig. 2A). The histological subtype of the HBs studied was not related to the degree of GATA-4 positivity. Normal liver tissue adjacent to the tumor was GATA-4 negative except for the Kuppfer cells.
To assess the connection between GATA-4 expression and cell proliferation in tumors, we performed KI-67 antigen immunostaining on HB tumor samples. We found that KI-67 expression was higher in the areas where GATA-4 was expressed (data not shown). Both GATA-4 and KI-67 reactivities were especially high on the peripheral areas of the tumors, suggesting that GATA-4 may be associated with cell proliferation and tumor growth.
In addition, the presence of GATA-4 protein was detected in liver tumors using Western blotting. We were able to demonstrate the presence of GATA-4 in all 5 HBs and in 1 of 2 childhood HCCs examined (Fig. 2B). mRNA extracted from tumor samples of children and adults was subjected to RT-PCR. Expression of GATA-4 mRNA was detected in 2 of 5 HBs and 1 of 2 childhood HCCs (Fig. 2C). The discrepancy between the results obtained in the same samples is likely caused by differences in the sensitivities of the methods used.
Immunohistochemical staining for GATA-6 was positive in 5 of 14 childhood liver tumors (Table 1, Fig. 2A). There was no correlation between GATA-4 and GATA-6 positivity. All of the adult liver tumors were GATA-6 negative. Interestingly, GATA-6 protein was detected in 4 of 5 HB samples and in both HCC samples studied by Western blotting (Fig. 2B). Accordingly, GATA-6 mRNA was demonstrated in all 7 samples (5 HBs and 2 HCCs) (Fig. 2C). The higher frequency of positivity in Western blotting may be the result of better sensitivity of this method compared with immunohistochemistry.
As HTT-I predisposes to early HCC, we wanted to evaluate GATA-4 and GATA-6 expression of biopsies obtained from patients with HTT-I. We found that GATA-4 and GATA-6 were present in regenerative liver nodules in 6 of 10 and 4 of 10 liver samples, respectively (data not shown). There is an increased rate of cell division in these preneoplastic nodules, and it is possible that GATA-4 and/or GATA-6 are associated with increased cell proliferation in these premalignant lesions.
Amount of Functionally Normal GATA-4 in a Human HB Cell Line Does Not Affect Apoptosis or Cell Proliferation
First, we evaluated the expression of endogenous GATA-4 in human HUH6 HB cells using immunocytochemistry (Fig. 3A), Western blotting, and RT-PCR (data not shown). All of these methods revealed abundant expression of this transcription factor in untreated cells. Subsequently, HUH6 cells were infected with recombinant adenovirus to overexpress either WT GATA-4 or DN mutant GATA-4. Neither GATA-4 overexpression with WT GATA-4 nor disturbing its function with DN GATA-4 affected the amount of cleaved caspases 3 and 7, the executor caspases of apoptotic pathway (Fig. 3A, C). These treatments also do not lead to any statistically significant changes in the ratio of proliferating and resting cells as measured by BrdU staining (Fig. 3B, C).
Erythropoietin Is Expressed in HBs
Epo is known to be expressed in human fetal liver (30). We here examined Epo protein expression in fetal liver and detected Epo in all 4 fetal liver samples (12–31 weeks of gestation). Epo protein was also detected in most (10/14) of the HBs but only in a minority of childhood (1/3) or adult (4/8) HCCs (Table 1, Fig. 4A).
Subsequently, we wanted to evaluate the expression of Epo in human HUH6 HB cells and found that they produced EPO in vitro (Fig. 4B). Subsequently, we investigated whether the amount of GATA-4 in HUH6 cells affects the EPO production by these cells. For this purpose, HUH6 cells were infected with adenoviruses carrying WT GATA-4, DN GATA-4, or with a LacZ control virus. Epo protein concentrations were measured by enzyme-linked immunosorbent assay from the cell medium collected after 48 (Fig. 4B) or 96 hours of incubation (data not shown). These experiments revealed that GATA-4 overexpression or disturbing its function does not affect the amount of Epo secreted by HUH6 cells into the medium.
HB and HCC are the most frequent malignant pediatric liver tumors characterized by aggressive growth and early metastases. The exact molecular mechanisms leading to these greatly malignant tumors remain, however, incompletely understood. We have evaluated the possible role of GATA-4, transcription factor implicated in normal liver development, in pediatric and adult liver tumors. GATA-4 has been shown to play a role in the carcinogenesis of various other types of tumors, and its expression is altered in malignant adrenocortical (31), ovarian (32), testicular (33), esophageal (34), and pancreatic tumors (29) as well as in neuroblastomas (35). We demonstrated that GATA-4, absent from late fetal and postnatal liver, is abundantly expressed in childhood liver tumors. GATA-4 was present in both pediatric HB and HCC but absent in adult HCC. Therefore, it cannot be used to segregate these 2 major types of hepatic neoplasms in children. The differential expression of GATA-4 in pediatric versus adult tumors may be related to the differences in the tumor etiologies in various age groups; hereditary factors often predispose to childhood liver cancer, whereas extrinsic factors, such as cirrhosis and hepatitis, are rare in pediatric but common in adult liver tumors (36). We also found that GATA-4 expression is induced in the preneoplastic liver nodules in hereditary tyrosinemia, similar to preneoplastic lesions in the gastrointestinal tract (23), and can thus be involved in the processes leading ultimately to overt malignancy in these patients.
Based on its role in other tissues, GATA-4 can be presumed to inhibit the apoptotic death of the liver tumor cells, advance their proliferation, or regulate tumor angiogenesis, but its exact role in these tumors needs to be clarified. Rather than serving as an independent factor to promote tumor cell survival in liver tumors, GATA-4 is likely involved in complicated molecular pathways. Accordingly, human HB cell lines have been shown to express high amounts of antiapoptotic factors Bcl (B-cell lymphoma)-2 and Bcl-X protecting the cells from commonly used cytotoxic drugs (37,38); however, GATA-4 regulates Bcl-2 and Bcl-X in several tissues (27,39–41). Like GATA-4, these 2 antiapoptotic factors are not, however, expressed in normal postnatal hepatocytes (42).
In the human liver, GATA-4 is expressed in the hepatocytes early in development (11). GATA-4 expression has been reported in human hepatocytes during 13 to 18 weeks of gestation (29). We could not demonstrate GATA-4 expression in the hepatocytes at gestational week 12 or later, but instead found it limited to Kuppfer cells and epithelial cells lining hepatic vessels. GATA-6 was not detected in normal hepatocytes, and was found only in the epithelial cells lining the liver and hepatic vessels. Similar to our findings in the postnatal human liver, GATA-4 is absent from the postnatal murine hepatocytes (43). It is likely that GATA-4 and GATA-6 play important roles during the early fetal liver development in humans.
GATA-4 and GATA-6 regulate many developmentally important genes in the liver including BMP-4 (bone morphogenic protein) and hepatic nuclear factors (HNFs) (10). GATA-4 has been implicated as an early regulator of albumin gene expression in fetal murine hepatocytes (9). GATA-4 acts in concert with HNF-3 to stimulate albumin gene transcription (44), and GATA-4 and GATA-6 cooperate with HNF-1α to activate the liver fatty acid–binding protein gene (Fabpl) (43). Similar to GATA factors, HNFs have been connected to liver tumorigenesis (45–47).
Given that HB tissue has several features typical for the fetal liver, it was interesting to find that GATA-4, which is physiologically expressed in early fetal liver (11), was also abundantly expressed in most of the tumors. We propose that GATA-4 can act as a factor enhancing liver cancer cell survival. In HTT-I, GATA-4 expression seems to be turned on at the same time the cells begin to divide in an uncontrolled manner eventually leading to overt cancer. This finding suggests that GATA-4 may induce tumor cell proliferation in the diseased liver. As discussed above, GATA-4 can act as an antiapoptotic factor (39,40); however, disturbing the normal GATA-4 function can induce apoptosis (48). Herein we examined the role of GATA-4 in apoptosis and proliferation in a human HB cell line, but could not demonstrate alterations in proliferation or apoptosis of these cells after manipulations of the amount or function of GATA-4. The endogenous GATA-4 production in these cells may well be so high that no changes in cell survival can be observed by further increasing the GATA-4 content or by partially abolishing the normal GATA-4 function. It is also possible that GATA-4 protects cells from various toxins such as shown for doxorubicin in cardiomyocytes (39). Because doxorubicin is a key chemotherapeutic agent in childhood liver cancer, it will be of interest to test whether similar results can be obtained also in malignant liver cells.
Based on what is known about GATA-4 and Epo expression patterns in fetal liver, our finding on the expression of Epo in childhood liver tumors was not unexpected. The biological significance of Epo in childhood liver tumors remains, however, to be elucidated. Regulation of Epo by GATA-4 has been suggested, because silencing of GATA-4 with siRNA reduces Epo gene expression in HCC cells (11). In the present study, we found no correlation between GATA-4 expression level and the amount of Epo produced by HUH6 cells. The differential result in these 2 studies may be caused by different methods and cells used.
In the present study, we demonstrated that GATA-4 is expressed in pediatric liver cancer. We postulate that GATA-4 has a role in childhood liver tumors as an antiapoptotic or growth-enhancing factor. Several factors implicated in liver tumorigenesis have been connected to GATA-4 and can be hypothesized to act together with this transcription factor in promoting tumor cell proliferation or survival. The present results provide new information on the molecular mechanisms of pediatric liver neoplasms and can ultimately help to improve their treatment.
We thank Drs Jeffery Molkentin and Mona Nemer for providing the adenoviral constructs. We also thank Drs Päivi Heikkilä and Hannu Jalanko for help, and Ms Taru Jokinen for technical assistance.
1. Meyers RL. Tumors of the liver in children. Surg Oncol 2007; 16:195–203.
2. Ross JA, Gurney JG. Hepatoblastoma incidence in the United States from 1973 to 1992. Med Pediatr Oncol 1998; 30:141–142.
3. Litten JB, Tomlinson GE. Liver tumors in children. Oncologist 2008; 13:812–820.
4. Ishak K, Goodman Z, Stocker T. Tumors of the Liver and Intrahepatic Bile Ducts. Washington, DC:American Registry of Pathology; 2001.
5. Ashorn M, Pitkanen S, Salo MK, et al. Current strategies for the treatment of hereditary tyrosinemia type I. Paediatr Drugs 2006; 8:47–54.
6. Deleted in proof.
7. Burch JB. Regulation of GATA gene expression during vertebrate development. Semin Cell Dev Biol 2005; 16:71–81.
8. Brewer A, Pizzey J. GATA factors in vertebrate heart development and disease. Expert Rev Mol Med 2006; 8:1–20.
9. Bossard P, Zaret KS. GATA transcription factors as potentiators of gut endoderm differentiation. Development 1998; 125:4909–4917.
10. Nemer G, Nemer M. Transcriptional activation of BMP-4 and regulation of mammalian organogenesis by GATA-4 and -6. Dev Biol 2003; 254:131–148.
11. Dame C, Sola MC, Lim KC, et al. Hepatic erythropoietin gene regulation by GATA-4. J Biol Chem 2004; 279:2955–2961.
12. Morrisey EE, Tang Z, Sigrist K, et al. GATA6 regulates HNF4 and is required for differentiation of visceral endoderm in the mouse embryo. Genes Dev 1998; 12:3579–3590.
13. Zhao R, Watt AJ, Li J, et al. GATA6 is essential for embryonic development of the liver but dispensable for early heart formation. Mol Cell Biol 2005; 25:2622–2631.
14. Ribatti D, Marzullo A, Gentile A, et al. Erythropoietin/erythropoietin-receptor system is involved in angiogenesis in human hepatocellular carcinoma. Histopathology 2007; 50:591–596.
15. von Schweinitz D, Schmidt D, Fuchs J, et al. Extramedullary hematopoiesis and intratumoral production of cytokines in childhood hepatoblastoma. Pediatr Res 1995; 38:555–563.
16. Liang SH, Hassett C, Omiecinski CJ. Alternative promoters determine tissue-specific expression profiles of the human microsomal epoxide hydrolase gene (EPHX1). Mol Pharmacol 2005; 67:220–230.
17. Ivanov GS, Kater JM, Jha SH, et al. Sp and GATA factors are critical for apolipoprotein AI downstream enhancer activity in human HepG2 cells. Gene 2003; 323:31–42.
18. Viger RS, Guittot SM, Anttonen M, et al. Role of the GATA family of transcription factors in endocrine development, function, and disease. Mol Endocrinol 2008; 22:781–798.
19. Ribatti D, Vacca A, Roccaro AM, et al. Erythropoietin as an angiogenic factor. Eur J Clin Invest 2003; 33:891–896.
20. Westenfelder C, Baranowski RL. Erythropoietin stimulates proliferation of human renal carcinoma cells. Kidney Int 2000; 58:647–657.
21. Arcasoy MO, Amin K, Karayal AF, et al. Functional significance of erythropoietin receptor expression in breast cancer. Lab Invest 2002; 82:911–918.
22. Kayser K, Gabius HJ. Analysis of expression of erythropoietin-binding sites in human lung carcinoma by the biotinylated ligand. Zentralbl Pathol 1992; 138:266–270.
23. Haveri H, Westerholm-Ormio M, Lindfors K, et al. Transcription factors GATA-4 and GATA-6 in normal and neoplastic human gastrointestinal mucosa. BMC Gastroenterol 2008; 8:9.
24. Dzieran J, Beck JF, Sonnemann J. Differential responsiveness of human hepatoma cells versus normal hepatocytes to TRAIL in combination with either histone deacetylase inhibitors or conventional cytostatics. Cancer Sci 2008; 99:1685–1692.
25. Charron F, Tsimiklis G, Arcand M, et al. Tissue-specific GATA factors are transcriptional effectors of the small GTPase RhoA. Genes Dev 2001; 15:2702–2719.
26. Liang Q, De Windt LJ, Witt SA, et al. The transcription factors GATA4 and GATA6 regulate cardiomyocyte hypertrophy in vitro and in vivo. J Biol Chem 2001; 276:30245–30253.
27. Kyronlahti A, Ramo M, Tamminen M, et al. GATA-4 regulates Bcl-2 expression in ovarian granulosa cell tumors. Endocrinology 2008; 149:5635–5642.
28. Jaaskelainen M, Kyronlahti A, Anttonen M, et al. TRAIL pathway components and their putative role in granulosa cell apoptosis in the human ovary. Differentiation 2009; 77:369–376.
29. Karafin MS, Cummings CT, Fu B, et al. The developmental transcription factor Gata4 is overexpressed in pancreatic ductal adenocarcinoma. Int J Clin Exp Pathol 2009; 3:47–55.
30. Dame C, Fahnenstich H, Freitag P, et al. Erythropoietin mRNA expression in human fetal and neonatal tissue. Blood 1998; 92:3218–3225.
31. Kiiveri S, Siltanen S, Rahman N, et al. Reciprocal changes in the expression of transcription factors GATA-4 and GATA-6 accompany adrenocortical tumorigenesis in mice and humans. Mol Med 1999; 5:490–501.
32. Laitinen MP, Anttonen M, Ketola I, et al. Transcription factors GATA-4 and GATA-6 and a GATA family cofactor, FOG-2, are expressed in human ovary and sex cord-derived ovarian tumors. J Clin Endocrinol Metab 2000; 85:3476–3483.
33. Ketola I, Pentikainen V, Vaskivuo T, et al. Expression of transcription factor GATA-4 during human testicular development and disease. J Clin Endocrinol Metab 2000; 85:3925–3931.
34. Guo M, House MG, Akiyama Y, et al. Hypermethylation of the GATA gene family in esophageal cancer. Int J Cancer 2006; 119:2078–2083.
35. Hoene V, Fischer M, Ivanova A, et al. GATA factors in human neuroblastoma: distinctive expression patterns in clinical subtypes. Br J Cancer 2009; 101:1481–1489.
36. Cabibbo G, Craxi A. Epidemiology, risk factors and surveillance of hepatocellular carcinoma. Eur Rev Med Pharmacol Sci 2010; 14:352–355.
37. Lieber J, Kirchner B, Eicher C, et al. Inhibition of Bcl-2 and Bcl-X enhances chemotherapy sensitivity in hepatoblastoma cells. Pediatr Blood Cancer 2010; 55:1089–1095.
38. Warmann SW, Frank H, Heitmann H, et al. Bcl-2 gene silencing in pediatric epithelial liver tumors. J Surg Res 2008; 144:43–48.
39. Aries A, Paradis P, Lefebvre C, et al. Essential role of GATA-4 in cell survival and drug-induced cardiotoxicity. Proc Natl Acad Sci U S A 2004; 101:6975–6980.
40. Kobayashi S, Lackey T, Huang Y, et al. Transcription factor gata4 regulates cardiac BCL2 gene expression in vitro and in vivo. FASEB J 2006; 20:800–802.
41. Suzuki YJ, Nagase H, Wong CM, et al. Regulation of Bcl-xL expression in lung vascular smooth muscle. Am J Respir Cell Mol Biol 2007; 36:678–687.
42. Charlotte F, L’Hermine A, Martin N, et al. Immunohistochemical detection of bcl-2 protein in normal and pathological human liver. Am J Pathol 1994; 144:460–465.
43. Divine JK, Staloch LJ, Haveri H, et al. GATA-4, GATA-5, and GATA-6 activate the rat liver fatty acid binding protein gene in concert with HNF-1alpha. Am J Physiol Gastrointest Liver Physiol 2004; 287:G1086–G1099.
44. Cirillo LA, Lin FR, Cuesta I, et al. Opening of compacted chromatin by early developmental transcription factors HNF3 (FoxA) and GATA-4. Mol Cell 2002; 9:279–289.
45. Xu L, Hui L, Wang S, et al. Expression profiling suggested a regulatory role of liver-enriched transcription factors in human hepatocellular carcinoma. Cancer Res 2001; 61:3176–3181.
46. Kalkuhl A, Kaestner K, Buchmann A, et al. Expression of hepatocyte-enriched nuclear transcription factors in mouse liver tumours. Carcinogenesis 1996; 17:609–612.
47. Flodby P, Liao DZ, Blanck A, et al. Expression of the liver-enriched transcription factors C/EBP alpha, C/EBP beta, HNF-1, and HNF-4 in preneoplastic nodules and hepatocellular carcinoma in rat liver. Mol Carcinog 1995; 12:103–109.
48. Kyronlahti A, Kauppinen M, Lind E, et al. GATA4 protects granulosa cell tumors from TRAIL-induced apoptosis. Endocr Relat Cancer 2010; 17:709–717.