| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Cell and Tumor Biology |
Knock-in Mice Exhibit Early Liver Glycogen Storage and Reduced Susceptibility to Hepatocellular Carcinoma
Departments of 1 Physiology and 2 Pathology, National University of Singapore and 3 Institute of Molecular and Cell Biology, Singapore, Singapore
Requests for reprints: Nai-dy Wang, Department of Physiology, National University of Singapore, Block MD9, 2 Medical Drive, Singapore 117597, Singapore. Phone: 65-6874-3663; Fax: 65-6778-8161; E-mail: phswnd{at}nus.edu.sg.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(C/EBP) is vital for establishing normal hepatic energy homeostasis and moderating hepatocellular growth. CEBPA loss-of-function mutations identified in acute myeloid leukemia patients support a tumor suppressor role for C/EBP. Recent work showed reductions of C/EBP levels in human hepatocellular carcinoma with the reductions correlating to tumor size and progression. We investigated the potential of reactivating c/ebp expression during hepatic carcinogenesis to prevent tumor cell growth. We have developed a c/ebp knock-in mouse in which a single-copy c/ebp is regulated by one allele of the -fetoprotein (AFP) gene promoter. The knock-in mice are physically indistinguishable from wild-type (WT) controls. However, knock-in animals were found to deposit fetal hepatic glycogen earlier than WT animals. Quantitative real-time PCR confirmed early c/ebp expression and early glycogen synthase gene activation in knock-in fetuses. We then used diethylnitrosamine to induce hepatocellular carcinoma in our animals. Diethylnitrosamine produced half the number of hepatocellular nodules in knock-in mice as in WT mice. Immunohistochemistry showed reduced C/EBP content in WT nodules whereas knock-in nodules stained strongly for C/EBP. The p21 protein was examined because it mediates a C/EBP growth arrest pathway. Nuclear p21 was absent in WT nodules whereas cytoplasmic p21 was abundant; knock-in nodules were positive for nuclear p21. Interestingly, only C/EBP-positive nodules were positive for nuclear p21, suggesting that C/EBP may be required to direct p21 to the cell nucleus to inhibit growth. Our data establish that controlled C/EBP production can inhibit liver tumor growth in vivo. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(C/EBP) belongs to the bZIP family of proteins (1, 2) and is a versatile regulatory factor. C/EBP induces the differentiation of adipocytes and granulocytes (3–5); it transactivates liver genes that establish and maintain normal energy homeostasis in the neonate (6, 7); and C/EBP functions as a potent inhibitor of cell division required for moderating the growth of hepatocytes (8–11).
The growth inhibitory properties of C/EBP have been extensively studied. Early investigations identified an inverse correlation between C/EBP gene expression and cellular proliferation in differentiated adipocytes (3, 4). Observations in rat primary hepatocytes and partial hepatectomies revealed a >80% drop in C/EBP transcripts as cells move into the growth phase (12). Western blot analyses showed that cultured hepatoma cells contain 
5% of the amount of C/EBP as the normal rat liver (13) and proliferation of hepatoma cells and transformed hepatocytes was found to be inhibited by transfection with constitutively expressed C/EBP genes (8, 9).
Inhibition of hepatoma cell growth by C/EBP does not require p53 or Rb (8). Instead, C/EBP curbs proliferation through multiple pathways involving interactions with cell cycle–related proteins (14–16). Notably, C/EBP increases the expression and stability of the cyclin-dependent kinase (Cdk) inhibitor p21/WAF-l/CIP-1 (14) and also directly inhibits two key Cdks, Cdk2 and Cdk4 (15). The growth-inhibiting features of C/EBP give it the characteristics of a tumor suppressor. Indeed, the identification of CEBPA loss-of-function mutations in patients with acute myeloid leukemia has solidified the status of CEBPA as a tumor suppressor gene (17).
Recent work revealed that C/EBP gene expression is reduced in human hepatocellular carcinoma cells and the extent of the decrease correlates with tumor size and progression (18). However, the question of whether or not forced expression of C/EBP can overcome liver cancer formation in vivo has remained unanswered. Our laboratory sought to explore the potential of reactivating C/EBP gene expression in mice during hepatic carcinogenesis to prevent tumor cell growth.
We used the promoter of the -fetoprotein (AFP) gene to drive expression of C/EBP because AFP is frequently up-regulated in liver cancer cells (19). In normal development, the fetal oncoprotein AFP is synthesized primarily in the embryonic liver and yolk sac (20, 21). AFP mRNA is detectable in the hepatic primordia, which form on the 10th day of the 21-day mouse gestation period (21). Expression peaks in the fetal liver by the 19th day and decreases to negligible amounts after birth (20). The prenatal AFP gene expression pattern precedes fetal c/ebp expression in the liver by 2 to 3 days where C/EBP mRNA is detectable on day 13, peaks near the time of birth, and then also drops after birth (22, 23). Transcription of afp normally remains off in the adult; in contrast, c/ebp transcription increases to a maximum again as the animal enters the adult stage of development at 6 weeks of age (7). The increased production of C/EBP in adulthood is consistent with the concurrent decreased rate of hepatocyte proliferation. Interestingly, during instances of coerced hepatic cell division such as in partial hepatectomies and tumorigenesis, c/ebp transcription becomes suppressed whereas afp expression becomes reactivated (11, 12, 19, 24, 25). This inverse relationship in gene activation between c/ebp and afp presents an ideal condition for artificially expressing c/ebp under the controlling elements of the AFP gene to test C/EBP as an inhibitor of tumor growth during chemically induced carcinogenesis. Our experiments used the carcinogen diethylnitrosamine because it is well known to produce hepatocellular carcinoma in animals (26, 27).
Here, we report the development of a C/EBP gene knock-in mouse strain in which a single-copy C/EBP gene is placed under the regulation of one of two endogenous alleles of the mouse AFP gene promoter. We find that c/ebp knock-in mice deposit fetal hepatic glycogen earlier than wild-type (WT) controls and afp promoter-driven c/ebp expression confers reduced susceptibility to diethylnitrosamine-induced hepatocarcinogenesis.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and AFP gene sequences were isolated from a Stratagene 
phage library (Stratagene, La Jolla, CA) containing murine 129Sv genomic DNA. afp promoter (5.5 kb) and upstream sequence were spliced onto the C/EBP gene sequence in a pBlueScript vector; a PGKneo expression cassette was placed downstream of c/ebp; an additional 1 kb of afp sequence starting from inside intron 1 to a portion of intron 3 was added downstream of the PGKneo expression cassette to provide sufficient homologous sequence to promote efficient gene targeting into the AFP gene locus; and MC1tk was placed at the 3' end of the cloned sequence (see Fig. 1). The construct (25 µg/mL) was linearized and electroporated into AB2.2 mouse embryonic stem cells (1.8 x 106 cells/mL). Positive drug selection using G418 (active ingredient, 0.18 mg/mL; Sigma, St. Louis, MO) and negative selection using ganciclovir (2 µmol/L; Roche, Mannheim, Germany) were employed to enrich the yield of gene-targeted clones. Correctly modified stem cells were identified by PCR and verified by Southern blotting. Two populations of targeted clones were then used to generate male, chimeric founder mice carrying the altered allele in their germ cells. The founders were bred to produce heterozygous gene targeted knock-in mice.
|
, 5'-TGGACAAGAACAGCAACGAGTA-3' and 5'-GCAGTTGCCATGGCCTTGA-3' generating a 238-bp product; C/EBPß, 5'-GGTTTCGGGACTTGATGCAATC-3' and 5'-CAACAACCCCGCAGGAACAT-3' generating a 131-bp product; C/EBP
, 5'-CCCAAAGCTATGTGCCTTTC-3' and 5'-CTGGAGGGTTTGTGTTTTCTG-3' generating a 94-bp product; glycogen synthase (GS), 5'-CCATGAACAGCAAGGGTTGTAA-3' and 5'-TGGAAGTGGGCAACCACATA-3' generating a 262-bp product; phosphoenolpyruvate (PEP) carboxy kinase, 5'-CGCTGGCCAAGATTGGTATT-3' and 5'-GGCCCAGTTGTTGACCAAA-3' generating a 178-bp product; glucose-6-phosphatase, 5'-ACTGGTTCAACCTCGTCTTCAA-3' and 5'-AACCACAAGATGACGTTCAAACA-3' generating a 264-bp product; and ß-actin, 5'-ACCCACACTGTGCCCATCTA-3' and 5'-GCCACAGGATTCCATACCCA-3' generating a 342-bp product.
Histologic analysis. Livers were fixed in 10% neutral buffered formalin, paraffin embedded, and cut into 4-µm sections. Stains used include H&E and periodic acid Schiff. Primary antibodies used were C/EBP (14AA, Santa Cruz Biotechnology, Santa Cruz, CA) and p21 (M-19, Santa Cruz Biotechnology). Horseradish peroxidase–conjugated antirabbit antibodies (EnVision System, HRP (DAB), DakoCytomation, Carpinteria, CA) were used as secondary antibodies.
Northern blot and Western blot analyses. Northern and Western blots were quantified with the Molecular Dynamics Typhoon 8600 Variable Mode Imager. Probes for Northern blots were generated by PCR: a 542-bp C/EBP probe was synthesized using primer sequences 5'-CACTATGCTCCCGCCCCACTCAC-3' and 5'-CTCCTTCCCCCAGCCGTTAG-3'; a 299-bp AFP probe used primer sequences 5'-CACTGCTGCAACTCTTCGTA-3' and 5'-CTTTGGACCCTCTTCTGTGA-3'; and a 206-bp 18S probe used primer sequences 5'-CCGGCGGCTTTGGTGACTCTA-3' and 5'-CGCGCCTGCTGCCTTCCT-3'. Antibodies used in Western blot were C/EBP (14AA, Santa Cruz Biotechnology), AFP (C-19, Santa Cruz Biotechnology), p21 (M-19, Santa Cruz Biotechnology), Cdk2 (M-2, Santa Cruz Biotechnology), Cdk4 (C-22, Santa Cruz Biotechnology), cyclin A (C-19, Santa Cruz Biotechnology), cyclin D1 (M-20, Santa Cruz Biotechnology), cyclin E (M-20, Santa Cruz Biotechnology), proliferating cell nuclear antigen (PCNA; PC10, Santa Cruz Biotechnology), glyceraldehyde-3-phosphate dehydrogenase (GAPDH; Chemicon, Temecula, CA), and ß-actin (A5441, Sigma).
Carcinogenic induction. For tumor induction, diethylnitrosamine (Sigma) diluted in tricaprylin (Sigma) to a concentration of 2.5 µg/µL was given to 12-day-old male mice by i.p. injection at 10 µg/g body weight.
5-Bromo 2-deoxyuridine incorporation. Mice were injected via the i.p. route with 5-bromo 2-deoxyuridine (BrdUrd) solution (BD Biosciences PharMingen) at 30 µg/g body weight. Mouse livers were collected 4 hours postinjection and fixed in 10% buffered formalin. The formalin-fixed, paraffin-embedded liver sections were immunostained with the BD PharMingen (San Diego, CA) BrdUrd In Situ Detection Kit according to the instruction of the supplier. BrdUrd-positive and BrdUrd-negative nuclei were scored (at least 1,000 nuclei per slide were scored).
Immunoprecipitation and cyclin-dependent kinase assays. Cdk2 and Cdk4 were immunoprecipitated from whole-cell protein extracts (500 µg) for 2 hours at 4°C with saturating amounts of polyclonal Cdk2 (M-2) and Cdk4 (C-22) antibodies (Santa Cruz Biotechnology) and then with protein A-Sepharose beads (Calbiochem, San Diego, CA) for another 2 hours. For the Cdk2 kinase assay, the immunoprecipitated proteins on beads were washed thrice with 1x PBS and resuspended in 30 µL of kinase buffer [25 mmol/L Tris-HCl (pH 7.5), 5 mmol/L MgCl2, and 5 mmol/L dichlorodiphenyltrichloroethane]. Five microliters of the suspension were incubated with 2 µg histone H1 (Upstate, Charlottesville, VA) as substrate in the presence of 2 µCi of [
-32P]ATP for 50 minutes at 37°C. The reactions were stopped on ice and samples were boiled in 5x electrophoresis sample buffer and separated by PAGE. For the Cdk4 kinase assay, 4 µg of Rb (Cell Signaling Technology, Beverly, MA) were used as substrate. After incubation at 30°C for 30 minutes, the reaction was stopped on ice and samples were boiled in 5x electrophoresis sample buffer for 10 minutes. The reaction products were separated by PAGE and transferred to polyvinylidene difluoride membrane and analyzed by Western immunoblotting for phospho-Rb(Ser-795) (Cell Signaling Technology).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
knock-in mice are phenotypically similar to wild-type mice. Because our gene knock-in strategy replaces afp exon 1 with c/ebp sequence and a neomycin expression cassette (Fig. 1A), homozygous knock-in mice would be fully deficient for AFP. We initially thought that AFP-deficient animals would be embryonic lethal; however, Gabant et al. (28) recently created afp knock-out mice and showed that AFP is necessary for female fertility but not required for embryonic development. We confirm Gabant's observation that AFP-deficient mice can survive to adulthood (results not shown). The majority of our experiments were conducted using heterozygous knock-in mice generated by breeding heterozygous males with WT females. Genotypes were verified by Southern hybridization (Fig. 1B). Of 478 mice genotyped, 53% were WT and 47% were knock-in. These percentages do not deviate significantly from Mendelian expectations (Pearson 
2 test, P = 0.2). Knock-in mice are physically indistinguishable from their WT counterparts and they are fertile.
Early accumulation of CCAAT/enhancer binding protein transcripts in knock-in livers correlates with early glycogen synthase expression and early liver glycogen storage. Because afp promoter activity is high normally only in fetal liver, we verified expression of the C/EBP knock-in gene using quantitative real-time PCR analysis of fetal livers collected on days 13.5, 14.5, and 15.5 of gestation. As expected, C/EBP mRNA levels in knock-in animals were already higher than in WT animals by day 13.5 of gestation (Fig. 2A). C/EBP gene expression was consistently higher in knock-in animals than in WT animals throughout the time period tested. The 2-fold and 3-fold increases on days 13.5 and 15.5, respectively, are statistically significant (Fig. 2A). Whether the excess C/EBP transcript was due to expression of knock-in gene alone or due to a combination of knock-in gene plus elevated expression of the native gene remains to be determined.
|
knock-in mice stained for glycogen at a time point earlier than WT mice. On day 14.5 of gestation, zero of four knock-in and zero of three WT animals showed hepatic glycogen deposits (Fig. 2B). On day 15.5 of gestation, glycogen was distinctly visible in five of six knock-in livers with the sixth liver showing an abundance of glycogen; in contrast, eight of nine WT livers showed no detectable signs of glycogen deposition although trace amounts were found in the ninth liver (Fig. 2B). By day 16.5 of gestation, four of four knock-in livers contained abundant glycogen whereas only one of three WT livers showed abundant glycogen; the remaining two WT livers showed trace amounts of glycogen (Fig. 2B). Real-time PCR revealed GS mRNA amounts to be significantly higher in knock-in mice than in WT controls on day 14.5 of gestation (Fig. 2A). By day 15.5 of gestation, WT GS mRNA levels had increased to levels comparable to those of knock-in animals.
These data show that c/ebp knock-in mice express higher levels of C/EBP mRNA earlier than controls and this expression pattern coincides with the higher expression of glycogen synthase message and the early initiation of hepatic glycogen storage. However, two additional genes important for glucose homeostasis, PEP carboxy kinase and glucose-6-phosphatase, did not show significant deviations from normal on day 15.5 of gestation (Fig. 2C). The glucose-6-phosphatase message was not detectable in the fetal samples but was easily detected in adult controls (not shown). In adult mice, no difference was observed in blood glucose levels (mean ± SD) between WT (9.95 ± 0.89 mmol/L, n = 6) and knock-in animals (9.68 ± 2.35 mmol/L, n = 5).
Also revealed in gestational day 15.5 livers was a 2.5-fold increase of C/EBPß message in knock-in fetuses above WT fetuses (Fig. 2C). C/EBP levels were not significantly different between knock-in and WT fetal livers.
Differences in CCAAT/enhancer binding protein expression levels between knock-in and wild-type mice are highest during gestation and diminish near the time of birth. We assessed Northern and Western blots of liver samples for C/EBP and AFP gene products (Fig. 3A). In concurrence with our real-time PCR data, mRNA values normalized against 18S rRNA showed C/EBP message levels to be higher in knock-in livers than in WT livers on day 19.5 of gestation, the day of birth (Newborn), and at 6 weeks of age (Adult; Table 1). Figure 3B shows the changes over time in the relative fold differences of C/EBP message between knock-in and WT mice from the fetal to adult stages of development. The data indicate that whereas C/EBP mRNA levels had increased earlier in knock-in mice, by the perinatal period, differences in c/ebp expression between WT and knock-in mice were minimal. This likely represents a negative feedback regulation during gestation and we postulate that the native genes are down-regulated to prevent overproduction of C/EBP message. After birth, the small but statistically significantly higher level of C/EBP mRNA in knock-in mice can reasonably be attributed to the continued low-level activity of the AFP promoter (Fig. 3B).
|
|
protein amounts to be essentially the same between WT and knock-in mice from gestational day 19.5 onwards (Table 1), again indicating the effect of knock-in gene during the perinatal period is negligible. Differences in AFP mRNA levels between knock-in and WT mice were not statistically significant although AFP protein levels seemed to be reduced in knock-in mice (Table 1).
CCAAT/enhancer binding protein knock-in livers are less susceptible to hepatocellular carcinoma than wild-type livers. To investigate the potential effect of altered c/ebp expression on liver tumorigenesis, we induced hepatocellular DNA damage using i.p. injections of diethylnitrosamine at 10 µg/g body weight in 12-day-old male pups. Diethylnitrosamine is a carcinogenic agent which alkylates multiple sites on DNA, forming DNA adducts that lead to genetic alterations (27). All male pups from a given litter were injected in a single session. In addition, to preserve unbiased handling of WT and knock-in mice, all injections were done before genotyping of the animals. We used juvenile mice because rapidly proliferating liver cells are required to promote the tumorigenic effects of diethylnitrosamine. At 1 year of age, both WT and knock-in mice were killed and examined for the presence of liver tumors (Fig. 4A). For every size of tumor recorded, knock-in livers displayed fewer numbers than WT livers (Fig. 4B). The average number of surface tumors detected on knock-in livers (19.8 ± 12.6; n = 10) was significantly lower than that detected on WT livers (41.0 ± 23.3; n = 5; Student's t test, P < 0.04). Examination of liver sections revealed a histologic spectrum of hepatocellular nodules of various sizes ranging from dysplastic nodules to frank hepatocellular carcinoma (designated as "tumor"). WT livers tended to exhibit more tumors and dysplastic nodules arising in a background of dysplasia as compared with knock-in livers where nodules seemed to arise in normal-looking liver parenchyma (Fig. 4C). In the absence of diethylnitrosamine treatment, we have not found spontaneous development of hepatocellular nodules in either knock-in or WT animals.
|
correlates with hepatocellular nodules containing nuclear p21 whereas the absence of CCAAT/enhancer binding protein correlates with hepatocellular carcinomas lacking nuclear p21. Immunohistochemistry done on diethylnitrosamine-exposed livers showed positive nuclear staining for C/EBP in the hepatocellular nodules of knock-in mice whereas similarly sized nodules in WT mice had reduced levels of nuclear C/EBP (Fig. 5A). Correspondingly, the nuclei of knock-in nodules were positive for p21 whereas the nuclei of nodules in WT mice showed low levels of p21 (Fig. 5B). The quantity of p21 seems to be less in the cells in knock-in nodules as compared with WT nodules. This is likely due to p21 being consolidated in the nuclear compartment in knock-in nodules. Western blot analysis has verified higher p21 amounts in nuclei of cells from knock-in nodules than in nuclei from WT nodules (see below). The levels of nuclear p21 possibly reflect cell growth in WT nodules and growth arrest in knock-in nodules. In accord with this view, nuclei of the largest knock-in tumors as well as nuclei of tumors of WT animals also were low for C/EBP and p21 (Fig. 5C) reinforcing the idea that reduced C/EBP levels correlate with tumor development (18). Thus far, we have not observed any large tumors in either WT or knock-in animals having high amounts of nuclear C/EBP. Moreover, we detected high levels of cytoplasmic p21 in the large knock-in tumors and WT nodules and tumors (Fig. 5B and C). This finding is consistent with a previous report that cytoplasmic p21 promotes cell growth (29).
|
because only those cells containing C/EBP exhibit nuclear p21 (Fig. 5A-C). This proposal is supported by the observation that p21 is absent in nuclear extracts derived from the livers of C/EBP gene knock-out mice (30). Western blot analysis confirms our immunohistochemistry findings that knock-in hepatocellular nodules contain higher levels of both nuclear C/EBP and p21 than WT nodules (Fig. 5D). Mean densitometric values expressed as a ratio to ß-actin (mean ± SD) show that C/EBP levels in knock-in nodules (0.66 + 0.09, n = 2) are nearly twice as high as those in WT nodules (0.35 ± 0.01, n =2; Student's t test, P < 0.05). Similarly, p21 levels in knock-in nodules (1.44 ± 0.30, n = 2) are >3.5 times higher than those in WT nodules (0.40 ± 0.14, n = 2; Student's t test, P < 0.05).
Cell proliferation rates are low in knock-in nodules and high in wild-type nodules. To determine the difference in growth rates between knock-in and WT nodules, BrdUrd uptake assays were done in age-matched 10-month-old mice (Fig. 6A). In concordance with the C/EBP and p21 results, the percent of WT nodule cells incorporating BrdUrd was nearly nine times higher than that of knock-in nodules (Fig. 6B). The percent BrdUrd uptake by knock-in nodules was not significantly higher than that of quiescent hepatocytes from both knock-in and WT mice not treated with diethylnitrosamine. The data also show that knock-in quiescent hepatocytes grow at a rate similar to WT quiescent hepatocytes.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
for establishing normal hepatic function and maintaining the nontransformed hepatic phenotype has been shown by studies of the C/EBP knockout mouse (6, 11). Recent work has strongly linked C/EBP to tumor suppressor activity (17, 18). However, attempts to arrest cell growth using C/EBP have been done primarily in vitro and questions about the effectiveness of C/EBP at inhibiting tumor growth in vivo have remained unanswered. As far as we are aware, we are the first to show inhibition of tumor cell growth using C/EBP in vivo in a gene knock-in mouse model.
We chose the gene knock-in method to be sure that every cell in our system expresses the inserted C/EBP gene in a temporal and spatial manner essentially identical to AFP. Our C/EBP knock-in mice appear completely normal, showing no obvious adverse effects. Although morphologically indistinguishable from WT animals, knock-in mice nevertheless showed early glycogen synthase expression and early glycogen deposition in accordance to the AFP promoter being activated earlier than normal C/EBP. However, the gluconeogenic enzymes, PEP carboxy kinase and glucose-6-phosphatase, were not up-regulated. This may be explained by the fact that fetal mice do not yet require independent glucose production as their energy needs are more than adequately met by the maternal blood supply. Thus, our results suggest that fetal carbohydrate metabolism is likely normal with the exception of early glycogen storage.
Because C/EBP is an important transcription factor known to transactivate many liver-specific genes, there exists the possibility of unintended and undesired gene activation in knock-in fetal livers. Although unfavorable consequences have not been detected, our observations indicate that accelerated development and maturation of knock-in liver may be a potential outcome. This hypothesis is supported by the early occurrence of glycogen accumulation (a hallmark of the mature hepatocyte) and the presence of elevated levels of C/EBPß mRNA in knock-in fetal liver. C/EBPß is known to be associated with hepatocyte proliferation (31, 32).
Our tumor induction data indicate that knock-in mice develop half the number of nodules and tumors as WT animals. Although the time of onset of nodule/tumor formation is similar for knock-in and WT mice (microscopic foci could be detected in both starting at 6 months of age), the total number of nodules and tumors detected at 1 year of age was significantly less in knock-in mice than in WT mice. Low BrdUrd uptake and low PCNA protein amounts in knock-in nodules show that knock-in nodules do not have the high proliferative capacity of WT nodules. In addition, BrdUrd uptake assays indicate that the cell proliferation rate in knock-in nodules is comparable to that in normal WT quiescent hepatocytes. Additional evidence supporting this inference is provided by kinase assays which show that, first, Cdk2 and Cdk4 activities in knock-in nodules are lower than that of WT nodules and, second, the levels of these kinase activities in knock-in nodules are similar to those in normal quiescent adult hepatocytes. Taken together, these data suggest that knock-in nodules experienced growth arrest during their development. However, the range of nodule/tumor sizes detected in knock-in mice implies that the time of arrest was not uniform, perhaps due to a variability in time of activation of the AFP promoter in cancer cells. Conspicuously, the largest tumors in knock-in mice were low for nuclear C/EBP and p21 (Fig. 4C). These likely represent the subset of tumors in which the AFP promoter escaped activation and thus escaped the growth inhibitory actions of C/EBP. This raises the prospect of exploring additional promoters for use in conjunction with the AFP promoter to drive expression of antitumorigenic genes in a manner that more fully encompasses all liver cancer cells. Furthermore, because our current tumor induction experiments were done in heterozygous knock-in mice, the question remains on whether or not homozygous knock-in animals may be better protected against chemically induced carcinogenesis. Additionally, incorporation of an inducible promoter into a transgene construct is also a possibility for driving higher-level expression of C/EBP.
The mechanism responsible for the arrest in nodule/tumor growth seen in our knock-in mice is likely the one elucidated by Timchenko et al. (14, 30). C/EBP was shown to restrain p21 degradation, allowing p21 to inhibit kinase activity and DNA synthesis. Our data are consistent with this picture as the majority of knock-in nodules were positive for both C/EBP and nuclear p21 whereas WT nodules/tumors were absent for both. Moreover, our results show that nuclear p21 is detected only in the presence of C/EBP. Thus, we propose that, in addition to stabilizing p21, C/EBP may also be necessary for recruiting p21 to the cell nucleus. Whether or not the hepatocellular nodules in knock-in animals are composed of permanently growth-arrested cells is an issue yet to be resolved.
Our results show that C/EBP gene knock-in mice possess greater resistance to diethylnitrosamine-induced liver tumorigenesis than WT mice, establishing for the first time that C/EBP gene expression regulated by the AFP promoter is a viable strategy for inhibiting liver tumor growth in vivo. In addition, we provide evidence that early c/ebp expression accelerates hepatic glycogen storage. Because premature human infants experience life-threatening hypoglycemia, perhaps as a result of incomplete activation of CEBPA and its target genes, creating a process which stimulates C/EBP production to accelerate liver maturation and drive expression of critical energy-related genes is desirable. Similarly, forced expression of the C/EBP gene in liver tumor cells may prove useful in attenuating hepatocellular carcinoma development. Further investigations will elucidate the feasibility of applying controlled C/EBP gene expression for the benefit of preterm infants and hepatocellular carcinoma patients.
|
Acknowledgments |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
We thank Y.C. Lim for helpful discussions and for providing expertly prepared tissue sections. For technical assistance, we thank J. Woo, S. Gopinadhan, and D. Swee. We thank K.C. Wang for critical reading of the manuscript.
Received 1/ 6/05. Revised 8/28/05. Accepted 9/13/05.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Science 1992;256:379–82. is a regulatory switch sufficient for induction of granulocytic development from bipotential myeloid progenitors. Mol Cell Biol 1998;18:4301–14. knockout mice. Science 1995;269:1108–12.: a critical regulator of genes governing integrative metabolic processes. Curr Opin Genet Dev 1995;5:565–70.[CrossRef][Medline]
occurs in many cell types, does not require the presence of p53 or Rb, and is not affected by large T-antigen. Nucleic Acids Res 1995;23:4726–33. identifies a dominant antiproliferative role for this isoform in hepatocytes. J Biol Chem 1996;271:7343–50.. J Biol Chem 1996;271:24753–60. gene expression results in increased DNA synthesis and an increased frequency of immortalization of freshly isolated mouse hepatocytes. Hepatology 1998;27:392–401.[CrossRef][Medline]
) gene expression during hepatocyte proliferation in the regenerating liver and in culture. Mol Cell Biol 1992;12:2553–60. (C/EBP) inhibits cell proliferation through the p21 (WAF-1/CIP-1/SDI-1) protein. Genes Dev 1996;10:804–15. arrests cell proliferation through direct inhibition of cdk2 and cdk4. Mol Cell 2001;8:817–28.[CrossRef][Medline]
triggers proteasome-dependent degradation of cdk4 during growth arrest. EMBO J 2002;21:930–41.[CrossRef][Medline]
(C/EBP), in acute myeloid leukemia. Nat Genet 2001;27:263–70.[CrossRef][Medline]
and ß genes in human hepatocellular carcinoma: a possible prognostic marker. Anticancer Res 2003;23:351–4.[Medline]
-fetoprotein, albumin and fibrinogen polypeptide mRNAs in different organs, developing tissues and in liver during carcinogenesis in rats. Biochim Biophys Acta 1982;699:131–7.[Medline]
-fetoprotein genes in foetal tissues of mouse and rat. Dev Biol 1984;102:51–60.[CrossRef][Medline]
-fetoprotein messenger RNA expression during mouse embryogenesis. EMBO J 1983;2:549–54.[Medline]
-fetoprotein in pregnant, neonatal, and partially hepatectomized rats. Cancer Res 1974;34:865–71.-fetoprotein reexpression in tumors. Semin Cancer Biol 1999;9:95–107.[CrossRef][Medline]
-Fetoprotein, the major fetal serum protein, is not essential for embryonic development but is required for female fertility. Proc Natl Acad Sci U S A 2002;99:12865–70. regulates p21 protein and hepatocyte proliferation in newborn mice. Mol Cell Biol 1997;17:7353–61.[Abstract]
. Mol Cell 1999;4:1087–92.[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  Y. Dewa, J. Nishimura, M. Jin, M. Kawai, Y. Saegusa, T. Harada, M. Shibutani, and K. Mitsumori Molecular Expression Analysis of {beta}-Naphthoflavone-induced Hepatocellular Tumors in Rats Toxicol Pathol, June 1, 2009; 37(4): 446 - 455. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. SEIFEDDINE, M.-C. FULCHIGNONI-LATAUD, and L. MASSAAD-MASSADE Down-regulation of C/EBP{alpha} in Breast Cancer Cells by Hypoxia-Estrogen Combination is Mainly Due to Hypoxia Anticancer Res, April 1, 2009; 29(4): 1227 - 1231. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Koschmieder, B. Halmos, E. Levantini, and D. G. Tenen Dysregulation of the C/EBP{alpha} Differentiation Pathway in Human Cancer J. Clin. Oncol., February 1, 2009; 27(4): 619 - 628. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G.-L. Wang, E. Salisbury, X. Shi, L. Timchenko, E. E. Medrano, and N. A. Timchenko HDAC1 Cooperates with C/EBP{alpha} in the Inhibition of Liver Proliferation in Old Mice J. Biol. Chem., September 19, 2008; 283(38): 26169 - 26178. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Seifeddine, A. Dreiem, E. Blanc, M.-C. Fulchignoni-Lataud, M.-A. L. F. Belda, F. Lecuru, T. H. Mayi, N. Mazure, V. Favaudon, C. Massaad, et al. Hypoxia Down-regulates CCAAT/Enhancer Binding Protein-{alpha} Expression in Breast Cancer Cells Cancer Res., April 1, 2008; 68(7): 2158 - 2165. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Parent, D. Kolippakkam, G. Booth, and L. Beretta Mammalian Target of Rapamycin Activation Impairs Hepatocytic Differentiation and Targets Genes Moderating Lipid Homeostasis and Hepatocellular Growth Cancer Res., May 1, 2007; 67(9): 4337 - 4345. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Datta, S. Majumder, H. Kutay, T. Motiwala, W. Frankel, R. Costa, H. C. Cha, O. A. MacDougald, S. T. Jacob, and K. Ghoshal Metallothionein Expression Is Suppressed in Primary Human Hepatocellular Carcinomas and Is Mediated through Inactivation of CCAAT/Enhancer Binding Protein {alpha} by Phosphatidylinositol 3-Kinase Signaling Cascade Cancer Res., March 15, 2007; 67(6): 2736 - 2746. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G.-L. Wang, X. Shi, E. Salisbury, Y. Sun, J. H. Albrecht, R. G. Smith, and N. A. Timchenko Growth Hormone Corrects Proliferation and Transcription of Phosphoenolpyruvate Carboxykinase in Livers of Old Mice via Elimination of CCAAT/Enhancer-binding Protein {alpha}-Brm Complex J. Biol. Chem., January 12, 2007; 282(2): 1468 - 1478. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. McFie, G.-L. Wang, N. A. Timchenko, H. L. Wilson, X. Hu, and W. J. Roesler Identification of a Co-repressor That Inhibits the Transcriptional and Growth-Arrest Activities of CCAAT/Enhancer-binding Protein {alpha} J. Biol. Chem., June 30, 2006; 281(26): 18069 - 18080. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
). Right, enlarged pictures of boxed areas in left. Dashed lines, tumor (t) and nodule (n) boundaries. Bars, 1 cm and 100 µm.
