- Email: [email protected]
Effect of simvastatin, a 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitor, on α-fetoprotein gene expression through interaction with the ras-mediated pathway
Journal of Hepatology
Copyright 8 European Association for the Studv of the Liver 1999
1999; 30: 904-9 10
Printed in Denmark All rights reserved Munksgaord
Copenhagen
Journalof Hepatology ISSN 0168-8278
Effect of simvastatin, a 3-hydroxy-3-methylglutarylcoenzyme A reductase inhibitor,on a-fetoprotein gene expression through interaction with the ras-mediated pathway Hiroyuki
Mazume ‘, Keisuke Nakata2, Daisaku Hida’, Keisuke Hamasaki’, Kazuhiko Nakao’, Yuji Kato’ and Katsumi Eguchi’
‘The First Department ~f‘hternal
Shotaro
Tsuruta’,
Medicine, Nagasaki University School of‘A4edicine und ‘Health Research Center, Nagasaki University, Nugasuki, Jnpan
Background/Aims: The ras proto-oncogene encodes a small GTP-binding protein (Ras) which regulates cell growth and differentiation by relaying signals from the cell surface to the nucleus. In the present study, the role of Ras signal transduction pathway in a-fetoprotein (AFP) gene expression was evaluated in HuH7 human hepatoma cells using simvastatin, a 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitor, which blocks Ras function through inhibition of farnesylation, and the rasva’-*’expression vector. Methods: The HUH-~ cells were treated with simvastatin (10 -01/l), or both simvastatin and mevalonate (300 mol/l), and numbers of viable cells were counted after treatment. To elucidate the effects of simvastatin on AFP gene expression and the interactive effect of simvastatin on Ras signal transduction pathway, Northern blotting and transient chloramphenicol
acetyltransferase plasmid transfection assays were performed. Results: Cell growth was inhibited by simvastatin, and this growth inhibition was restored by addition of mevalonate. Levels of AFP mRNA but not albumin mRNA were elevated by simvastatin in a dose-dependent manner (l-10 pal/l). AFP promoter and enhancer activities were stimulated by simvastatin. In contrast, both activities were repressed by transfection with the rasvB’-‘2expression vector. The rasval-l2-mediated repression was restored by simvastatin and returned to the repressed level by simvastatin plus mevalonate. Conclusions: These results indicate that the Ras signal transduction pathway functions to down-regulate the AFP gene transcription in human hepatoma cells.
CL-FETOPROTEIN (AFP) and albumin genes are similar in structure and believed to be derived from a common ancestral gene (1). The two genes are arranged in tandem on human chromosome 4 and expressed at high levels during the fetal stage; after birth, AFP expression decreases rapidly to an almost undetectable level (2). However, AFP gene transcription is reactivated in hepatoma cells (3). Recent studies have led to much progress in characterization of cis- and trans-acting elements regulating AFP gene transcription (46). Many growth factors are now known to
regulate AFP gene expression. Epidermal growth factor suppresses AFP enhancer activity, resulting in the down-regulation of the AFP gene (6). Transforming growth factor-/31 and hepatocyte growth factor repress AFP gene transcription through the reduction of its promoter activity (7,8). Although these growth factors exert their effects through the activation of the specific cell surface receptors, the relationship between the receptor-mediated intracellular signal transduction pathway and AFP gene expression is not fully understood. The ras-mediated pathway plays a key role in relaying signals from growth factor receptor-mediated stimuli. Thus, alterations in its activity should affect AFP gene expression at the transcriptional level. The ras proto-oncogene encodes a small GTP-binding protein (Ras) which relays the receptor-mediated stimuli from the cell surface to the nucleus (9,lO). The
?“”
Received 6 May; revised 27 October; accepted 3 November 1998
Correspondence: Katsumi Eguchi, The First Department of Internal Medicine, Nagasaki University School of Medicine, 1-7-1 Sakamoto, Nagasaki 852, Japan. Tel: 81 95 849 7260. Fax: 81 95 849 7270.
904
Key words: a-Fetoprotein; Ras; Simvastatin.
Effect of simvastatin on AFP expression
activation of Ras is mediated by molecular switches, the GDP-bound inactive and the GTP-bound active forms regulated by guanine nucleotide exchange factors (11-13). Oncogenic Ras mutants are defective in responsiveness to the GTPase activating protein and remain in the GTP-bound active form. This results in the constitutive activation of the Ras-mediated signal transduction pathway, which promotes aberrant cell growth in a variety of tumor cells. Several lines of evidence indicate that farnesyl isoprenoid linkage of Ras to the cell membrane (14) is essential to Ras function (15). Since simvastatin, a 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitor, suppresses the formation of vital cholesterol precursors, including mevalonate (16), in addition to reduction of cholesterol synthesis (17), it can block processing or membrane localization of Ras through inhibition of farnesylation (16). In the present study, the effects of simvastatin not only on cell growth but also on AFP and albumin gene expression in HUH-~ human hepatoma cells were investigated by Northern blotting and transient chloramphenicol acetyltransferase (CAT) plasmid transfection assay.
Materials and Methods Chemicals Simvastatin was a generous gift from Banyu Co., Ltd. (Tokyo, Japan). To convert the inactive lactone form to the active form, the drug was dissolved in ethanol, heated at 50°C in 0.1 N NaOH, neutralized with HCl, and stored at -20°C. Mevalonate was purchased from Sigma Chem., Co. (St. Louis, MO, USA). [a-32P] deoxycytidine triphosphate and D-threo-[dichloroacetyl-1-i4C] chloramphenicol were purchased from Amersham Japan (Tokyo, Japan). Lipofectin reagents were purchased from Gibco (Gaithersburg, MD, USA). Antibodies to the mitogen-activated protein kinase (MAPK) superfamily of enzymes, which react only with active form (active MAPK) and both active and inactive forms (pan ERK) were purchased from Promega Co. (Madison, WI, USA). Cell culture HUH-7 human hepatoma cells were maintained in a chemically defined medium, IS-RPM1 (18). Cell growth was analyzed using 25 cm2 flasks (Falcon Plastics, Los Angeles, CA, USA), and l.5X106 cells were placed in each flask and incubated at 37°C in 5% CO*. One day later, the medium was replaced with fresh medium or fresh medium containing 10 mol/l simvastatin alone or simvastatin plus 300 mol/ 1 mevalonate. The cells were further incubated at 37°C in 5% COz, and numbers of viable cells were counted 24 and 48 h after incubation using the trypan blue dye exclusion method. Northern blot analysis Total cellular RNA was isolated by the acid guanidinium-phenolchloroform method. The extracted RNA (15 pg) was fractionated on a 1% formaldehyde agarose gel, transferred to a nylon membrane, and hybridized with a 32P-labeled cDNA probe. AFP cDNA (pHAF2) (19) and albumin cDNA (palb-7) (2) were used as probes. CATplasmids and ra.@” expression vector The CAT plasmids used in this study were described previously (20,21). The pBR-CAT plasmid contains the CAT coding sequence and the simian virus 40 polyadenylation signal but no upstream regu-
latory sequences. pAFS.l-CAT and pAF0.3-CAT contain 5.1~kilobase pairs (kb) and 0.3-kb of the AFP 5’-flanking sequence, respectively, linked to the CAT gene in pBR-CAT. pSVAF2.4CAT contains 2.4 kb full AFP enhancer region (-5.3 to -2.9 kb) inserted at the 5’ end of the CAT gene in pSVl’-CAT which contains the simian virus 40 TATA box but lacks most of the 72 base pairs (bp) repeat sequence. pAF5.1 [d 2.7]-CAT contains the 2.4 kb full AFP enhancer and promoter regions but lacks the 2.7 kb (-2.9 kb to -169 bp) fragment of pAFS.l-CAT, where the AFP “silencer” region is identified. pSVZCAT (6) plasmid was used as a control plasmid in this study. Based on the coding sequence of c-H-ras, the mutant ra@-I2 sequence which has a G-to-T substitution in codon 12 to encode valine instead of glycine, was synthesized as described previously (22). The fragment was inserted into pRSV dneo (23), by removing the neomycin resistance gene to yield pRSV-rasva’-i2 (24). Cell transfection and CAT assay Transfection was performed using 3 peg @SVAFZ.CCAT and pSV2CAT), 5 ,ug (pAFS.l-CAT and pAFS.l[d 2.71~CAT), 10 ,ug @AF0.3CAT), 2.5 peg (pRSVrasVa’-‘2 and pRSV dneo) of the plasmid DNA per dish (60 cm2) by the lipofection method (25). After transfection, HUH-7 cells were incubated with the fresh medium in the absence or presence of 10 eel/l simvastatin alone, or simvastatin plus 300 GoY 1 mevalonate. Two days later, cells were harvested and lysed by five cycles of freezing and thawing. The lysate was heated at 65°C for 10 min and centrifuged at 10 000 rpm for 5 min, and the supernatant was used for determination of CAT activity as described previously (26). Zmmunoblotting To investigate the alteration in the ras-mediated activation pathway by simvastatin, expression of its downstream activity known as ERK (extracellular signal regulated kinase) or MAPK (mitogen-activated protein kinase) was analyzed by Western blotting. The HUH-7 cells plasmid or pRSV dneo plasmid were transfected with pRSVras va1-12 as a control vector and incubated in the absence or presence of 10 eolil simvastatin. Twelve hours later, cellular protein was extracted with the lysis buffer containing 25 mM Tris, 25 mM NaCl, 10 mM sodium fluoride, 10 mM sodium pyrophosphate, 0.2 mM sodium molybate, 1 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride and 0.5 mM EGTA. Ten micrograms of cellular protein was electrophoresed in a 12% SDSpolyacrylamide gel and electroblotted to a nitrocellulose membrane. Nonspecific binding of the membrane was blocked with 1% bovine serum albumin for 1 h at room temperature. The membrane was then probed with anti-pan ERK as a control or anti-active MAPK for 2 h at room temperature. The signals were then visualized by the enhanced chemiluminescence system (Amersham Life Science).
Results Inhibition of cell growth by simvastatin Numbers of viable cells were counted 24 and 48 h after incubation in the absence or presence of 10 pmol/l simvastatin alone or simvastatin plus 300 pmol/l mevalonate. Cell growth was inhibited by simvastatin (Fig. 1). The simvastatin-treated cells had round shapes on the contrast microscopy as reported previously (27,28). The simvastatin-mediated growth inhibition was almost completely restored by addition of mevalonate. Increase in the level of AFP mRNA but not albumin mRNA by simvastatin Total cellular RNA was extracted from the cells incubated for 48 h with fresh medium or fresh medium containing varying concentrations (1 .O, 5.0, 10 PmoYl)
905
H. Mazume et al.
of simvastatin. Levels of AFP mRNA were dose dependently elevated by simvastatin (Fig. 2A). In contrast, levels of albumin mRNA were not significantly changed by this treatment (Fig. 2B).
0’
0
24
48 Ms.)
Time Fig. I. Ejfect of simvastatin on cell were incubated in the absence ( l) or simvastatin alone (A) or 10 pmol/l ,amol/l mevalonate(A), and numbers counted 24 and 48 h after treatment. meanIflSD (n=4). “p
growth. HUH-~ cells presence of 10 pmol/l simvastatin plus 300 of viable cells were Results represent the
Stimulation of AFP enhuncer and promoter activities by simvastatin through inhibition of Ras function
In an attempt to clarify the regulatory mechanism of AFP gene expression by simvastatin, transient CAT plasmid transfection experiments were performed. CAT expression from pAFS.l-CAT that has the fulllength of the 5’-regulatory sequence of the AFP gene was elevated by simvastatin, and this stimulatory effect was blocked by addition of mevalonate (Fig. 3). CAT expression from pSV2-CAT used as a control plasmid was not affected by these treatments. Co-transfection with pAFS.I-CAT and pRSV-ras”“‘-‘z, a ras”“‘-‘* expression vector, caused reduction of CAT activity, compared with co-transfection with pAFS.l-CAT and pRSV dneo used as a control vector (Fig. 4). However, when the cells co-transfected with pAFS.l-CAT and pRSV-ras”“‘-” were treated with 10 /tmol/l simvastatin, CAT expression from pAF5.1~C4T returned to the control level. In addition, treatment with both 10 pmolll simvastatin and 300 pmolil mevalonate resulted in restoration of the rash’-“-mediated repression. CAT expression from pAFS.l[d2.7]-CAT which contains both the AFP promoter and enhancer elements but lacks “silencer elements” was elevated by simvastatin and suppressed by transfection with pRSV-‘as”“‘-” (Fig. 5). Similarly, CAT activities from pAF0.3-CAT and pSVAF2.4-CAT, each of which contains only the AFP promoter and enhancer region, respectively, were stimulated by simvastatin and repressed by transfection with pRSV-ras”“‘-‘2.
A B
A AFP mRNA
AlbuminmRNA
Fig. 2. Efjects of simvastatin on the levels of AFP and albumin mRNA. HUH-~ cells were treated with vurying concentrations of simvustatin (O-10 ,amol/l). Two days later, total RNA was isolated and analyzedjor AFP mRNA (A) and albumin mRNA (B) by Northern blotting. Lane I, control; lane 2, I pmol/l simvastatin; lane 3, 5 pmol/l simvastatin; lane 4, 10 ,umol/l simvastatin.
906
B pAF 5. I-CAT ;i-------? + @ r),
+-~-AC
Fig. 3. Effect of simvastatin on CAT expression jrom pAFS.l-CAT. A,fter transfection with pAFS. I-CATplusmid (A) or pSV2-CAT plasmid (B), HUH-~ cells were incubated in the absence or presence of IO ,amol/l simvastutin ulone or IO pmol/l simvastatin plus 300 pmol/l mevalonate. Two days later, CAT activities were determined as described in Materials and Methods. ~-AC, 3-acet_vlchloramphenicol; I-Ac, 3acetylchloramphenicol; Cm, chloramphenicol. Lane I, control; lane 2, 10 ,amol/l simvastatin; lune 3, 10 ,amol/l simvastatin plus 300 pmol/l mcvulonflte.
Effect of simvastatin on AFP expression
Fig. 4. Restoration of the rasva’-t2-mediated repression of pAF5.1-CAT activity by simvastatin. The HUH-~ cells were co-transfected with pAFS.I-CAT plasmid and pRSV Aneo plasmid as a control vector (lane I) or pAFS.l-CAT plasmid and pRS V-ras vat-‘2plasmid (lanes 24). To determine whether the rasva”‘2-mediated repression of pAF5. ICAT activity (lane 2) was restored by simvastatin, 10 pmol/l simvastatin (lane 3) or 10 pmol/l simvastatin plus 300 ,umol/l mevalonate which counteracts on the effect of simvastatin (lane 4) were added to the culture media. Two days later, CAT activities were determined as described in Materials and Methods. ~-AC, 3-acetylchloramphenicol; lAC, I-acetylchloramphenicol; Cm, chloramphenicol.
Suppression of the rasva”‘2-induced by simvastatin
activation
of
MAPK
By Western blot analysis, transfection of the pRSVplasmid into HUH-~ cells resulted in a large increase in expression of the active form of MAPK (Fig. 6A), while it did not affect pan-ERK expression used as a control (Fig. 6B). When the cells transfected with pRSV-ras”“‘-I2 plasmid were treated with 10 pmol/l simvastatin, rasval-12-induced activation of MAPK was apparently inhibited by this treatment.
Fig, 5. Effects of simvastatin or the transfection with rasvat-I2 gene expression vector on AFP-CAT expression. (A) After transfection with indicated CATplasmids (lanes 1 and 2) or co-transfection with indicated CAT plasmids and pRS V-ras vat-12plasmid (lane 3), HUH-~ cells were incubated in the absence (lanes 1 and 3) or presence of 10 pmol/l simvastatin (lane 2). Two days later, CAT activities were determined as described in Materials and Methods. 3AC, 3-acetylchloramphenicol; I-AC, 3-acetylchloramphenicol; Cm, chloramphenicol. (B) Quantitative analysis by densitometric scanning. CAT activities are expressed as acetylation (percentage). Each column shows the value of the control cells (open column), the cells treated with simvastatin (closed column) and the cells co-transfected with pRSV-ras va’-‘2 (shaded column).
rasval-12
Discussion The Ras-mediated signal transduction pathway is a key factor involved in regulation of cell growth and differentiation (9,lO). The activation of its pathway is unlikely to result from a signal-step event, but rather represents a complex process involving processing or
membrane localization of Ras and molecular switches mediated by GDP-bound inactive and GTP-bound active forms. Ras is farnesylated to the cystein residue of C-terminal CAAX motif (C, cystein; A, aliphatin amino acid; X, any other amino acid) and requires the motif for activation (14,29,30). HMG-CoA reductase inhibitors, including simvastatin, repress mevalonate synthesis (17). Since mevalonate is an essential substance for the farnesylation process (16), Ras function can be blocked by these agents through inhibition of this process. In this study, cell growth was inhibited by simvasta-
907
H. Mazume et al.
Fig. 6. Suppression of downstream activity of ras-mediated pathway by simvastatin. Cellular protein was extracted from the HUH-~ cells without any treatment (lane I), the cells transfected with pRSV Aneo plasmid (lane 2) or the cells transfected with pRSV-ras”“‘-‘2 plasmid (lanes 3 and 4), which were incubated in the absence (lane 3) or presence (lane 4) of 10 ymol/l simvastatin. Western blot analysis was performed using antibodies which recognized only active form of MAPK (A, active MAPK) and both active and inactive forms (B, pan ERK) as a control, respectively, as described in Materials and Methods.
tin in HUH-7 human hepatoma cells. This growth inhibition was restored by addition of mevalonate. Similar results were reported previously, where HMG-CoA reductase inhibitors such as lovastatin and simvastatin suppressed cell growth in many of transformed cell (27,28,31-33). This may be partly due to loss of Ras function caused by these agents. However, DeClue et al. reported that inhibition of cell growth by lovastatin did not exhibit the specificity for cell types and occurred irrespective of the Ras-induced transformation (33). In addition to farnesyl and cholesterol, other important substances for cell function are synthesized from mevalonate. Thus, depletion of these substances may contribute to the simvastatin-mediated growth inhibition in HUH-7 cells. Transcripts of the AFP gene but not the albumin gene were dose-dependently elevated in response to simvastatin. Transfection experiments with pAF5. lCAT which contains the full-length of the 5’-regulatory sequence of the AFP gene showed that AFP-CAT expression was stimulated by simvastatin and that this stimulatory effect was abolished by addition of mevalonate, suggesting a crucial role of mevalonate or its metabolites in regulation of the AFP gene as well as in cell growth. AFP-CAT expression was repressed by transfection with rasval-I2 expression vector. However, ras”“‘-‘*-mediated repression was restored by simvastatin. AFP promoter and enhancer activities were re908
duced by transfection with the rasva1-12expression vector, while both activities were enhanced by simvastatin. These results indicate that simvastatin up-regulates the AFP gene transcription through the inactivation of Ras signal transduction pathway, and that the Ras-mediated pathway functions to suppress AFP expression even in AFP-producing hepatoma cells. There is some controversy regarding the role of farnesylation in activation of Ras. However, when the downstream activity of the ras-mediated pathway was analyzed by Western blotting, the ras Va1-12-inducedactivation of MAPK was apparently reduced by 10 pmolil simvastatin. This would account for interaction of simvastatin with the ras-mediated pathway. Nakao et al. reported similar results: transfection with the c-Ha-ras expression vector into human hepatoma cells down-regulated the AFP gene but not the albumin gene in human hepatoma cells (34). Previous studies demonstrated that several growth factors repressed AFP gene expression in human hepatoma cells (68). Since these growth factors participate in the activation of the Ras-mediated pathway, our results, in part, account for mechanisms of growth factor-mediated repression of the AFP gene in these cells. In addition, Silberman et al. recently documented Ras response elements which were identical to or related to AP-1 sequence (35). This seems to agree with our results, because both AFP promoter and enhancer regions contain AP-l-like sequences and because AFP gene expression is down-regulated by cjun and c-fos in human hepatoma cells (36). Previous studies showed that the AFP silencer elements between its promoter and enhancer were relevant to the postneonatal repression of the AFP gene, and that the inactivation of the silencer activity would play a major role in the AFP gene re-expression in hepatoma cells (4,6). On the other hand, fluctuations in the serum level of AFP are often found in patients with cirrhosis. These fluctuations seem to result from cytokine- or growth factor-mediated stimuli to hepatocytes. Under these conditions, activation of the ras-mediated signal transduction pathway in hepatocytes is likely to repress the AFP gene transcription. Hepatocellular carcinoma (HCC) is one of the most common malignancies worldwide, especially in some areas of Asia and Africa. The development of HCC is closely associated with chronic liver disease, particularly cirrhosis (37-39). Several investigators have shown that, compared with other cancers, the ras gene mutation was less frequent in HCC (40-43). However, recent studies demonstrated that the activation of Ras expression is relevant to the development of cirrhotic nodules and HCC, but that its sustained elevation is no longer required for cell proliferation or progression
Effect of simvastatin on AFP expression
after tumor development (44,45). Consistent with this result is the report of Liu et al. that elevated levels of c-Ha-ras mRNA were constantly found in hepatocytes in patients with cirrhosis. Using immunohistochemical analysis (46), Takeuchi et al. reported that Ras overexpression in HCC was seen predominantly in patients with serum levels of AFP below 400 ng/ml (47). Thus, taking these findings together with our results, it is possible that AFP expression in cirrhotic nodules or HCC at an early stage is actively suppressed by the Ras-mediated signal transduction pathway.
References 1. Nahon JL. The regulation of albumin and alpha-fetoprotein gene expression in mammals [review]. Biochimie 1987; 69: 445-59. 2. Urano Y, Sakai M, Watanabe K, Tamaoki T. Tandem arrangement of the albumin and alpha-fetoprotein genes in the human genome. Gene 1984; 32: 255-61. 3. Alpert ME, Uriel J, Nechaud BD. Alpha-l fetoglobulin in the diagnosis of human hepatoma. N Engl J Med 1968; 278: 984-6. 4. Nakabayashi H, Hashimoto T, Miyao Y, Tjong KK, Chan J, Tamaoki T A position-dependent silencer plays a major role in repressing alpha-fetoprotein expression in human hepatoma. Mol Cell Biol 1991; 11: 5885-93. 5. Morinaga T, Yasuda H, Hashimoto T, Higashio K, Tamaoki T. A human alpha-fetoprotein enhancer-binding protein, ATBFl, contains four homeodomains and seventeen zinc lingers. Mol Cell Biol 1991; 11: 6041-9. 6. Nakata K, Motomura M, Nakabayashi H, Ido A, Tamaoki T. A possible mechanism of inverse developmental regulation of alpha-fetoprotein and albumin genes. Studies with epidermal growth factor and phorbol ester. J Biol Chem 1992; 267: 13314. 7. Nakao K, Nakata K, Mitsuoka S, Ohtsuru A, Ido A, Hatano M, et al. Transforming growth factor beta 1 differentially regulates alpha-fetoprotein and albumin in HUH-~ human hepatoma cells. Biochem Biophys Rcs Commun 1991; 174: 1294-9. 8. Hatano M, Nakata K, Nakao K, Tsutsumi T, Ohtsuru A, Nakamura T, et al. Hepatocyte growth factor down-regulates the alpha-fetoprotein gene expression in PLCIPRFIS human hepatoma cells. Biochem Biophys Rcs Commun 1992; 189: 385-91. 9. Egan SE, Weinberg RA. The pathway to signal achievement [news]. Nature 1993; 365: 781-3. 10. Khosravi FR, Der CJ. The Ras signal transduction pathway [review]. [221 refs]. Cancer Metast Rev 1994; 13: 67-89. 11. Bourne HR, Sanders DA, McCormick E The GTPase supcrfamily: conserved structure and molecular mechanism [review]. Nature 1991; 349: 117-27. 12. Bollag G, McCormick E Regulators and effecters of ras proteins [review]. Ann Rev Cell Biol 1991; 7: 601-32. 13. Quilliam LA, Khosravi FR, Huff SY, Der CJ. Guanine nucleotide exchange factors: activators of the Ras superfamily of proteins [review]. Bioessays 1995; 17: 395404. 14. Casey PJ, Solski PA, Der CJ, Buss JE. p2lras is modified by a farnesyl isoprenoid. Proc Nat1 Acad Sci USA 1989; 86: 8323-7. 15. Willumsen BM, Christensen A, Hubbert NL, Papageorge AG, Lowy DR. The p21 ras C-terminus is required for transformation and membrane association. Nature 1984; 310: 5836. 16. Goldstein JL, Brown MS, Regulation of the mevalonate pathway [review]. Nature 1990; 343: 425-30. 17. Alberts AW. Lovastatin and simvastatin - inhibitors of HMG CoA reductase and cholesterol biosynthesis. Cardiology 1990; 4: 14-21. 18. Nakabayashi H, Taketa K, Yamane T, Miyazaki M, Miyano K, Sato J. Phenotypical stability of a human hepatoma cell line, HUH-~, in long-term culture with chemically detined medium. Gann 1984; 75: 151-8.
19. Morinaga T, Sakai M, Wegmann TG, Tamaoki T. Primary structures of human alpha-fetoprotein and its mRNA. Proc Nat1 Acad Sci USA 1983; 80: 4604-8. 20. Watanabe K, Saito A, Tamaoki T. Cell-specific enhancer activity in a far upstream region of the human alpha-fetoprotein gene. J Biol Chem 1987; 262: 4812-8. 21. Nakabayashi H, Watanabe K, Saito A, Otsuru A, Sawadaishi K, Tamaoki T. Transcriptional regulation of alpha-fetoprotein expression by dexamethasone in human hepatoma cells. J Biol Chem 1989; 264: 266-71. 22. Miura K, Kamiya H, Tominaga M, Inoue Y, Ikehara M, Noguchi S, et al. Overproduction of cellular and activated Ha-ras proteins by mutating a synthetic gene. Chem Pharm Bull 1987; 35: 4878-82. 23. Gorman C, Padmanabhan R, Howard BH. High efficiency DNA-mediated transformation of primate cells. Science 1983; 221: 551-3. 24. Kamiya H, Miura K, Ohtomo N, Koda T, Kakinuma M, Nishimura S, et al. Transformation of NIH3T3 cells with synthetic cHa-ras genes. Jpn J Cancer Rcs 1989; 80: 200-3. 25. Tsutsumi T, Ido A, Nakao K, Hamasaki K, Kato Y, Ohtsuru A, et al. Reciprocal regulation of alpha-fetoprotein and albumin gene expression by butyrate in human hepatoma cells. Gastroenterology 1994; 107: 499-504. 26. Gorman CM, Moffat LF, Howard BH. Recombinant genomes which express chloramphenicol acetyltransferase in mammalian cells. Mol Cell Biol 1982; 2: 1044-51. 27. Morris TJ, Palm SL, Pena J, Furcht LT, Buchwald H. Inhibition of hepatoma tissue culture-4 cell growth by lovastatin: a novel chemotherapeuic strategy. Surg Forum 1991; 42: 452-4. 28. Morris TJ, Palm SL, Furcht LL, Buchwald H. Effect of lovastatin alone and as an adjuvant chemotherapeutic agent on hepatoma tissue culture-4 cell growth. Ann Surg Oncol 1995; 2: 26674. 29. Barbacid M. ras genes [review]. Ann Rev Biochem 1987; 56: 779827. 30. Lowy DR, Willumsen BM. Function and regulation of ras [rcview]. Ann Rev Biochem 1993; 62: 851-91. 3 1. Sumi S, Beauchamp RD, Townsend CJ, Uchida T, Murakami M, Rajaraman S, et al. Inhibition of pancreatic adenocarcinoma cell growth by lovastatin. Gastroenterology 1992; 103: 982-9. 32. Ura H, Obara T, Nishino N, Tanno S, Okamura K, Namiki M. Cytotoxicity of simvastatin to pancreatic adenocarcinoma cells containing mutant ras gene. Jpn J Cancer Res 1994; 85: 633-8. 33. DeClue JE, Vass WC, Papageorge AG, Lowy DR, Willumsen BM. Inhibition of cell growth by lovastatin is independent of ras function. Cancer Res 1991; 51: 712-7. 34. Nakao K, Lawless D, Ohe Y, Miyao Y, Nakabayashi H, Kamiya H, et al. c-Ha-ras down-regulates the alpha-fetoprotein gene but not the albumin gene in human hepatoma cells. Mol Cell Biol 1990; 10: 1461-9. 35. Silberman S, Janulis M, Schultz RM. Characterization of downstream Ras signals that induce alternative protease-dependent invasive phenotypes. J Biol Chem 1997; 272: 5927-35. 36. Saegusa M, Ido A, Nakabayashi H, Saki M, Tamaoki T. Downregulation of human a-fetoprotein promoter by c-jun and c-fos. J Tumor Marker Oncol 1994; 272: 29-34. 37. Kubo Y, Okuda K, Musha H, Nakashima T Detection of hepatocellular carcinoma during a clinical follow-up of chronic liver disease: observations in 31 patients. Gastroenterology 1978; 74: 578-82. 38. Oka H, Kurioka N, Kim K, Kanno T, Kuroki T, Mizoguchi Y, et al. Prospective study of early detection of hepatocellular carcinoma in patients with cirrhosis. Hepatology 1990; 12: 680-7. 39. Shinagawa T Ohto M, Kimura K, Tsunetomi S, Morita M, Saisho H, et al. Diagnosis and clinical features of small hepatocellular carcinoma with emphasis on the utility of real-time ultrasonography. A study in 51 patients. Gastroenterology 1984; 86: 495502. 40. Bjersing L, Andersson C, Lithner E Hepatocellular carcinoma in
909
H. Mazume et al. patients from northern Sweden with acute intermittent porphyria: morphology and mutations. Cancer Epidemiol Biomarkers Prevention 1996; 5: 393-7. 41. Cui Q, Wang Z, Chen J, Liu T. Detection of c-Ki-ras oncogene codon 12 point mutations in lung carcinoma, gastric carcinoma and hepatic carcinoma [in Chinese]. Chung Kuo i Hsueh Ko Hsueh Yuan Hsueh Pao Acta Academiae Medicinae Sinicae 1996; 18: 95-8. 42. Leon M, Kew MC. Analysis of ras gene mutations in hepatocellular carcinoma in southern African blacks. Anticancer Res 1995; 15: 859-61. 43. Tada M, Omata M, Ohto M. Analysis of ras gene mutations in human hepatic malignant tumors by polymerase chain reaction and direct sequencing. Cancer Res 1990; 50: 11214.
910
44. Nonomura A, Ohta G, Hayashi M, Izumi R, Watanabe K, Takayanagi N, et al. Immunohistochemical detection of ras oncogene p21 product in liver cirrhosis and hepatocellular carcinoma. Am J Gastroenterol 1987; 82: 512-8. 45. Radosevich JA, Gould KA, Koukoulis GK, Haines GK, Rosen ST, Lee I, et al. Immunolocalization of ras oncogene ~21 in human liver diseases. Ultrastruct Path01 1993; 17: l-8. 46. Liu P, Flejou JF, Feldmann G. Bernuau D. Activation of ras oncogene in livers with cirrhosis. J Hepdtol 1994; 21: 1103-g. 47. Takeuchi T, Kagawa K, Hikitd H, Tada H, Deguchi T, Okanoue T, et al. The expression of ras ~21 product in hepdtocellular carcinoma [in Japanese]. Nippon Shokakibyo Gakkai Zasshi [Jpn J Gastroenterol] 1990; 87: 1686-91.
Copyright 8 European Association for the Studv of the Liver 1999
1999; 30: 904-9 10
Printed in Denmark All rights reserved Munksgaord
Copenhagen
Journalof Hepatology ISSN 0168-8278
Effect of simvastatin, a 3-hydroxy-3-methylglutarylcoenzyme A reductase inhibitor,on a-fetoprotein gene expression through interaction with the ras-mediated pathway Hiroyuki
Mazume ‘, Keisuke Nakata2, Daisaku Hida’, Keisuke Hamasaki’, Kazuhiko Nakao’, Yuji Kato’ and Katsumi Eguchi’
‘The First Department ~f‘hternal
Shotaro
Tsuruta’,
Medicine, Nagasaki University School of‘A4edicine und ‘Health Research Center, Nagasaki University, Nugasuki, Jnpan
Background/Aims: The ras proto-oncogene encodes a small GTP-binding protein (Ras) which regulates cell growth and differentiation by relaying signals from the cell surface to the nucleus. In the present study, the role of Ras signal transduction pathway in a-fetoprotein (AFP) gene expression was evaluated in HuH7 human hepatoma cells using simvastatin, a 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitor, which blocks Ras function through inhibition of farnesylation, and the rasva’-*’expression vector. Methods: The HUH-~ cells were treated with simvastatin (10 -01/l), or both simvastatin and mevalonate (300 mol/l), and numbers of viable cells were counted after treatment. To elucidate the effects of simvastatin on AFP gene expression and the interactive effect of simvastatin on Ras signal transduction pathway, Northern blotting and transient chloramphenicol
acetyltransferase plasmid transfection assays were performed. Results: Cell growth was inhibited by simvastatin, and this growth inhibition was restored by addition of mevalonate. Levels of AFP mRNA but not albumin mRNA were elevated by simvastatin in a dose-dependent manner (l-10 pal/l). AFP promoter and enhancer activities were stimulated by simvastatin. In contrast, both activities were repressed by transfection with the rasvB’-‘2expression vector. The rasval-l2-mediated repression was restored by simvastatin and returned to the repressed level by simvastatin plus mevalonate. Conclusions: These results indicate that the Ras signal transduction pathway functions to down-regulate the AFP gene transcription in human hepatoma cells.
CL-FETOPROTEIN (AFP) and albumin genes are similar in structure and believed to be derived from a common ancestral gene (1). The two genes are arranged in tandem on human chromosome 4 and expressed at high levels during the fetal stage; after birth, AFP expression decreases rapidly to an almost undetectable level (2). However, AFP gene transcription is reactivated in hepatoma cells (3). Recent studies have led to much progress in characterization of cis- and trans-acting elements regulating AFP gene transcription (46). Many growth factors are now known to
regulate AFP gene expression. Epidermal growth factor suppresses AFP enhancer activity, resulting in the down-regulation of the AFP gene (6). Transforming growth factor-/31 and hepatocyte growth factor repress AFP gene transcription through the reduction of its promoter activity (7,8). Although these growth factors exert their effects through the activation of the specific cell surface receptors, the relationship between the receptor-mediated intracellular signal transduction pathway and AFP gene expression is not fully understood. The ras-mediated pathway plays a key role in relaying signals from growth factor receptor-mediated stimuli. Thus, alterations in its activity should affect AFP gene expression at the transcriptional level. The ras proto-oncogene encodes a small GTP-binding protein (Ras) which relays the receptor-mediated stimuli from the cell surface to the nucleus (9,lO). The
?“”
Received 6 May; revised 27 October; accepted 3 November 1998
Correspondence: Katsumi Eguchi, The First Department of Internal Medicine, Nagasaki University School of Medicine, 1-7-1 Sakamoto, Nagasaki 852, Japan. Tel: 81 95 849 7260. Fax: 81 95 849 7270.
904
Key words: a-Fetoprotein; Ras; Simvastatin.
Effect of simvastatin on AFP expression
activation of Ras is mediated by molecular switches, the GDP-bound inactive and the GTP-bound active forms regulated by guanine nucleotide exchange factors (11-13). Oncogenic Ras mutants are defective in responsiveness to the GTPase activating protein and remain in the GTP-bound active form. This results in the constitutive activation of the Ras-mediated signal transduction pathway, which promotes aberrant cell growth in a variety of tumor cells. Several lines of evidence indicate that farnesyl isoprenoid linkage of Ras to the cell membrane (14) is essential to Ras function (15). Since simvastatin, a 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitor, suppresses the formation of vital cholesterol precursors, including mevalonate (16), in addition to reduction of cholesterol synthesis (17), it can block processing or membrane localization of Ras through inhibition of farnesylation (16). In the present study, the effects of simvastatin not only on cell growth but also on AFP and albumin gene expression in HUH-~ human hepatoma cells were investigated by Northern blotting and transient chloramphenicol acetyltransferase (CAT) plasmid transfection assay.
Materials and Methods Chemicals Simvastatin was a generous gift from Banyu Co., Ltd. (Tokyo, Japan). To convert the inactive lactone form to the active form, the drug was dissolved in ethanol, heated at 50°C in 0.1 N NaOH, neutralized with HCl, and stored at -20°C. Mevalonate was purchased from Sigma Chem., Co. (St. Louis, MO, USA). [a-32P] deoxycytidine triphosphate and D-threo-[dichloroacetyl-1-i4C] chloramphenicol were purchased from Amersham Japan (Tokyo, Japan). Lipofectin reagents were purchased from Gibco (Gaithersburg, MD, USA). Antibodies to the mitogen-activated protein kinase (MAPK) superfamily of enzymes, which react only with active form (active MAPK) and both active and inactive forms (pan ERK) were purchased from Promega Co. (Madison, WI, USA). Cell culture HUH-7 human hepatoma cells were maintained in a chemically defined medium, IS-RPM1 (18). Cell growth was analyzed using 25 cm2 flasks (Falcon Plastics, Los Angeles, CA, USA), and l.5X106 cells were placed in each flask and incubated at 37°C in 5% CO*. One day later, the medium was replaced with fresh medium or fresh medium containing 10 mol/l simvastatin alone or simvastatin plus 300 mol/ 1 mevalonate. The cells were further incubated at 37°C in 5% COz, and numbers of viable cells were counted 24 and 48 h after incubation using the trypan blue dye exclusion method. Northern blot analysis Total cellular RNA was isolated by the acid guanidinium-phenolchloroform method. The extracted RNA (15 pg) was fractionated on a 1% formaldehyde agarose gel, transferred to a nylon membrane, and hybridized with a 32P-labeled cDNA probe. AFP cDNA (pHAF2) (19) and albumin cDNA (palb-7) (2) were used as probes. CATplasmids and ra.@” expression vector The CAT plasmids used in this study were described previously (20,21). The pBR-CAT plasmid contains the CAT coding sequence and the simian virus 40 polyadenylation signal but no upstream regu-
latory sequences. pAFS.l-CAT and pAF0.3-CAT contain 5.1~kilobase pairs (kb) and 0.3-kb of the AFP 5’-flanking sequence, respectively, linked to the CAT gene in pBR-CAT. pSVAF2.4CAT contains 2.4 kb full AFP enhancer region (-5.3 to -2.9 kb) inserted at the 5’ end of the CAT gene in pSVl’-CAT which contains the simian virus 40 TATA box but lacks most of the 72 base pairs (bp) repeat sequence. pAF5.1 [d 2.7]-CAT contains the 2.4 kb full AFP enhancer and promoter regions but lacks the 2.7 kb (-2.9 kb to -169 bp) fragment of pAFS.l-CAT, where the AFP “silencer” region is identified. pSVZCAT (6) plasmid was used as a control plasmid in this study. Based on the coding sequence of c-H-ras, the mutant ra@-I2 sequence which has a G-to-T substitution in codon 12 to encode valine instead of glycine, was synthesized as described previously (22). The fragment was inserted into pRSV dneo (23), by removing the neomycin resistance gene to yield pRSV-rasva’-i2 (24). Cell transfection and CAT assay Transfection was performed using 3 peg @SVAFZ.CCAT and pSV2CAT), 5 ,ug (pAFS.l-CAT and pAFS.l[d 2.71~CAT), 10 ,ug @AF0.3CAT), 2.5 peg (pRSVrasVa’-‘2 and pRSV dneo) of the plasmid DNA per dish (60 cm2) by the lipofection method (25). After transfection, HUH-7 cells were incubated with the fresh medium in the absence or presence of 10 eel/l simvastatin alone, or simvastatin plus 300 GoY 1 mevalonate. Two days later, cells were harvested and lysed by five cycles of freezing and thawing. The lysate was heated at 65°C for 10 min and centrifuged at 10 000 rpm for 5 min, and the supernatant was used for determination of CAT activity as described previously (26). Zmmunoblotting To investigate the alteration in the ras-mediated activation pathway by simvastatin, expression of its downstream activity known as ERK (extracellular signal regulated kinase) or MAPK (mitogen-activated protein kinase) was analyzed by Western blotting. The HUH-7 cells plasmid or pRSV dneo plasmid were transfected with pRSVras va1-12 as a control vector and incubated in the absence or presence of 10 eolil simvastatin. Twelve hours later, cellular protein was extracted with the lysis buffer containing 25 mM Tris, 25 mM NaCl, 10 mM sodium fluoride, 10 mM sodium pyrophosphate, 0.2 mM sodium molybate, 1 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride and 0.5 mM EGTA. Ten micrograms of cellular protein was electrophoresed in a 12% SDSpolyacrylamide gel and electroblotted to a nitrocellulose membrane. Nonspecific binding of the membrane was blocked with 1% bovine serum albumin for 1 h at room temperature. The membrane was then probed with anti-pan ERK as a control or anti-active MAPK for 2 h at room temperature. The signals were then visualized by the enhanced chemiluminescence system (Amersham Life Science).
Results Inhibition of cell growth by simvastatin Numbers of viable cells were counted 24 and 48 h after incubation in the absence or presence of 10 pmol/l simvastatin alone or simvastatin plus 300 pmol/l mevalonate. Cell growth was inhibited by simvastatin (Fig. 1). The simvastatin-treated cells had round shapes on the contrast microscopy as reported previously (27,28). The simvastatin-mediated growth inhibition was almost completely restored by addition of mevalonate. Increase in the level of AFP mRNA but not albumin mRNA by simvastatin Total cellular RNA was extracted from the cells incubated for 48 h with fresh medium or fresh medium containing varying concentrations (1 .O, 5.0, 10 PmoYl)
905
H. Mazume et al.
of simvastatin. Levels of AFP mRNA were dose dependently elevated by simvastatin (Fig. 2A). In contrast, levels of albumin mRNA were not significantly changed by this treatment (Fig. 2B).
0’
0
24
48 Ms.)
Time Fig. I. Ejfect of simvastatin on cell were incubated in the absence ( l) or simvastatin alone (A) or 10 pmol/l ,amol/l mevalonate(A), and numbers counted 24 and 48 h after treatment. meanIflSD (n=4). “p
growth. HUH-~ cells presence of 10 pmol/l simvastatin plus 300 of viable cells were Results represent the
Stimulation of AFP enhuncer and promoter activities by simvastatin through inhibition of Ras function
In an attempt to clarify the regulatory mechanism of AFP gene expression by simvastatin, transient CAT plasmid transfection experiments were performed. CAT expression from pAFS.l-CAT that has the fulllength of the 5’-regulatory sequence of the AFP gene was elevated by simvastatin, and this stimulatory effect was blocked by addition of mevalonate (Fig. 3). CAT expression from pSV2-CAT used as a control plasmid was not affected by these treatments. Co-transfection with pAFS.I-CAT and pRSV-ras”“‘-‘z, a ras”“‘-‘* expression vector, caused reduction of CAT activity, compared with co-transfection with pAFS.l-CAT and pRSV dneo used as a control vector (Fig. 4). However, when the cells co-transfected with pAFS.l-CAT and pRSV-ras”“‘-” were treated with 10 /tmol/l simvastatin, CAT expression from pAF5.1~C4T returned to the control level. In addition, treatment with both 10 pmolll simvastatin and 300 pmolil mevalonate resulted in restoration of the rash’-“-mediated repression. CAT expression from pAFS.l[d2.7]-CAT which contains both the AFP promoter and enhancer elements but lacks “silencer elements” was elevated by simvastatin and suppressed by transfection with pRSV-‘as”“‘-” (Fig. 5). Similarly, CAT activities from pAF0.3-CAT and pSVAF2.4-CAT, each of which contains only the AFP promoter and enhancer region, respectively, were stimulated by simvastatin and repressed by transfection with pRSV-ras”“‘-‘2.
A B
A AFP mRNA
AlbuminmRNA
Fig. 2. Efjects of simvastatin on the levels of AFP and albumin mRNA. HUH-~ cells were treated with vurying concentrations of simvustatin (O-10 ,amol/l). Two days later, total RNA was isolated and analyzedjor AFP mRNA (A) and albumin mRNA (B) by Northern blotting. Lane I, control; lane 2, I pmol/l simvastatin; lane 3, 5 pmol/l simvastatin; lane 4, 10 ,umol/l simvastatin.
906
B pAF 5. I-CAT ;i-------? + @ r),
+-~-AC
Fig. 3. Effect of simvastatin on CAT expression jrom pAFS.l-CAT. A,fter transfection with pAFS. I-CATplusmid (A) or pSV2-CAT plasmid (B), HUH-~ cells were incubated in the absence or presence of IO ,amol/l simvastutin ulone or IO pmol/l simvastatin plus 300 pmol/l mevalonate. Two days later, CAT activities were determined as described in Materials and Methods. ~-AC, 3-acet_vlchloramphenicol; I-Ac, 3acetylchloramphenicol; Cm, chloramphenicol. Lane I, control; lane 2, 10 ,amol/l simvastatin; lune 3, 10 ,amol/l simvastatin plus 300 pmol/l mcvulonflte.
Effect of simvastatin on AFP expression
Fig. 4. Restoration of the rasva’-t2-mediated repression of pAF5.1-CAT activity by simvastatin. The HUH-~ cells were co-transfected with pAFS.I-CAT plasmid and pRSV Aneo plasmid as a control vector (lane I) or pAFS.l-CAT plasmid and pRS V-ras vat-‘2plasmid (lanes 24). To determine whether the rasva”‘2-mediated repression of pAF5. ICAT activity (lane 2) was restored by simvastatin, 10 pmol/l simvastatin (lane 3) or 10 pmol/l simvastatin plus 300 ,umol/l mevalonate which counteracts on the effect of simvastatin (lane 4) were added to the culture media. Two days later, CAT activities were determined as described in Materials and Methods. ~-AC, 3-acetylchloramphenicol; lAC, I-acetylchloramphenicol; Cm, chloramphenicol.
Suppression of the rasva”‘2-induced by simvastatin
activation
of
MAPK
By Western blot analysis, transfection of the pRSVplasmid into HUH-~ cells resulted in a large increase in expression of the active form of MAPK (Fig. 6A), while it did not affect pan-ERK expression used as a control (Fig. 6B). When the cells transfected with pRSV-ras”“‘-I2 plasmid were treated with 10 pmol/l simvastatin, rasval-12-induced activation of MAPK was apparently inhibited by this treatment.
Fig, 5. Effects of simvastatin or the transfection with rasvat-I2 gene expression vector on AFP-CAT expression. (A) After transfection with indicated CATplasmids (lanes 1 and 2) or co-transfection with indicated CAT plasmids and pRS V-ras vat-12plasmid (lane 3), HUH-~ cells were incubated in the absence (lanes 1 and 3) or presence of 10 pmol/l simvastatin (lane 2). Two days later, CAT activities were determined as described in Materials and Methods. 3AC, 3-acetylchloramphenicol; I-AC, 3-acetylchloramphenicol; Cm, chloramphenicol. (B) Quantitative analysis by densitometric scanning. CAT activities are expressed as acetylation (percentage). Each column shows the value of the control cells (open column), the cells treated with simvastatin (closed column) and the cells co-transfected with pRSV-ras va’-‘2 (shaded column).
rasval-12
Discussion The Ras-mediated signal transduction pathway is a key factor involved in regulation of cell growth and differentiation (9,lO). The activation of its pathway is unlikely to result from a signal-step event, but rather represents a complex process involving processing or
membrane localization of Ras and molecular switches mediated by GDP-bound inactive and GTP-bound active forms. Ras is farnesylated to the cystein residue of C-terminal CAAX motif (C, cystein; A, aliphatin amino acid; X, any other amino acid) and requires the motif for activation (14,29,30). HMG-CoA reductase inhibitors, including simvastatin, repress mevalonate synthesis (17). Since mevalonate is an essential substance for the farnesylation process (16), Ras function can be blocked by these agents through inhibition of this process. In this study, cell growth was inhibited by simvasta-
907
H. Mazume et al.
Fig. 6. Suppression of downstream activity of ras-mediated pathway by simvastatin. Cellular protein was extracted from the HUH-~ cells without any treatment (lane I), the cells transfected with pRSV Aneo plasmid (lane 2) or the cells transfected with pRSV-ras”“‘-‘2 plasmid (lanes 3 and 4), which were incubated in the absence (lane 3) or presence (lane 4) of 10 ymol/l simvastatin. Western blot analysis was performed using antibodies which recognized only active form of MAPK (A, active MAPK) and both active and inactive forms (B, pan ERK) as a control, respectively, as described in Materials and Methods.
tin in HUH-7 human hepatoma cells. This growth inhibition was restored by addition of mevalonate. Similar results were reported previously, where HMG-CoA reductase inhibitors such as lovastatin and simvastatin suppressed cell growth in many of transformed cell (27,28,31-33). This may be partly due to loss of Ras function caused by these agents. However, DeClue et al. reported that inhibition of cell growth by lovastatin did not exhibit the specificity for cell types and occurred irrespective of the Ras-induced transformation (33). In addition to farnesyl and cholesterol, other important substances for cell function are synthesized from mevalonate. Thus, depletion of these substances may contribute to the simvastatin-mediated growth inhibition in HUH-7 cells. Transcripts of the AFP gene but not the albumin gene were dose-dependently elevated in response to simvastatin. Transfection experiments with pAF5. lCAT which contains the full-length of the 5’-regulatory sequence of the AFP gene showed that AFP-CAT expression was stimulated by simvastatin and that this stimulatory effect was abolished by addition of mevalonate, suggesting a crucial role of mevalonate or its metabolites in regulation of the AFP gene as well as in cell growth. AFP-CAT expression was repressed by transfection with rasval-I2 expression vector. However, ras”“‘-‘*-mediated repression was restored by simvastatin. AFP promoter and enhancer activities were re908
duced by transfection with the rasva1-12expression vector, while both activities were enhanced by simvastatin. These results indicate that simvastatin up-regulates the AFP gene transcription through the inactivation of Ras signal transduction pathway, and that the Ras-mediated pathway functions to suppress AFP expression even in AFP-producing hepatoma cells. There is some controversy regarding the role of farnesylation in activation of Ras. However, when the downstream activity of the ras-mediated pathway was analyzed by Western blotting, the ras Va1-12-inducedactivation of MAPK was apparently reduced by 10 pmolil simvastatin. This would account for interaction of simvastatin with the ras-mediated pathway. Nakao et al. reported similar results: transfection with the c-Ha-ras expression vector into human hepatoma cells down-regulated the AFP gene but not the albumin gene in human hepatoma cells (34). Previous studies demonstrated that several growth factors repressed AFP gene expression in human hepatoma cells (68). Since these growth factors participate in the activation of the Ras-mediated pathway, our results, in part, account for mechanisms of growth factor-mediated repression of the AFP gene in these cells. In addition, Silberman et al. recently documented Ras response elements which were identical to or related to AP-1 sequence (35). This seems to agree with our results, because both AFP promoter and enhancer regions contain AP-l-like sequences and because AFP gene expression is down-regulated by cjun and c-fos in human hepatoma cells (36). Previous studies showed that the AFP silencer elements between its promoter and enhancer were relevant to the postneonatal repression of the AFP gene, and that the inactivation of the silencer activity would play a major role in the AFP gene re-expression in hepatoma cells (4,6). On the other hand, fluctuations in the serum level of AFP are often found in patients with cirrhosis. These fluctuations seem to result from cytokine- or growth factor-mediated stimuli to hepatocytes. Under these conditions, activation of the ras-mediated signal transduction pathway in hepatocytes is likely to repress the AFP gene transcription. Hepatocellular carcinoma (HCC) is one of the most common malignancies worldwide, especially in some areas of Asia and Africa. The development of HCC is closely associated with chronic liver disease, particularly cirrhosis (37-39). Several investigators have shown that, compared with other cancers, the ras gene mutation was less frequent in HCC (40-43). However, recent studies demonstrated that the activation of Ras expression is relevant to the development of cirrhotic nodules and HCC, but that its sustained elevation is no longer required for cell proliferation or progression
Effect of simvastatin on AFP expression
after tumor development (44,45). Consistent with this result is the report of Liu et al. that elevated levels of c-Ha-ras mRNA were constantly found in hepatocytes in patients with cirrhosis. Using immunohistochemical analysis (46), Takeuchi et al. reported that Ras overexpression in HCC was seen predominantly in patients with serum levels of AFP below 400 ng/ml (47). Thus, taking these findings together with our results, it is possible that AFP expression in cirrhotic nodules or HCC at an early stage is actively suppressed by the Ras-mediated signal transduction pathway.
References 1. Nahon JL. The regulation of albumin and alpha-fetoprotein gene expression in mammals [review]. Biochimie 1987; 69: 445-59. 2. Urano Y, Sakai M, Watanabe K, Tamaoki T. Tandem arrangement of the albumin and alpha-fetoprotein genes in the human genome. Gene 1984; 32: 255-61. 3. Alpert ME, Uriel J, Nechaud BD. Alpha-l fetoglobulin in the diagnosis of human hepatoma. N Engl J Med 1968; 278: 984-6. 4. Nakabayashi H, Hashimoto T, Miyao Y, Tjong KK, Chan J, Tamaoki T A position-dependent silencer plays a major role in repressing alpha-fetoprotein expression in human hepatoma. Mol Cell Biol 1991; 11: 5885-93. 5. Morinaga T, Yasuda H, Hashimoto T, Higashio K, Tamaoki T. A human alpha-fetoprotein enhancer-binding protein, ATBFl, contains four homeodomains and seventeen zinc lingers. Mol Cell Biol 1991; 11: 6041-9. 6. Nakata K, Motomura M, Nakabayashi H, Ido A, Tamaoki T. A possible mechanism of inverse developmental regulation of alpha-fetoprotein and albumin genes. Studies with epidermal growth factor and phorbol ester. J Biol Chem 1992; 267: 13314. 7. Nakao K, Nakata K, Mitsuoka S, Ohtsuru A, Ido A, Hatano M, et al. Transforming growth factor beta 1 differentially regulates alpha-fetoprotein and albumin in HUH-~ human hepatoma cells. Biochem Biophys Rcs Commun 1991; 174: 1294-9. 8. Hatano M, Nakata K, Nakao K, Tsutsumi T, Ohtsuru A, Nakamura T, et al. Hepatocyte growth factor down-regulates the alpha-fetoprotein gene expression in PLCIPRFIS human hepatoma cells. Biochem Biophys Rcs Commun 1992; 189: 385-91. 9. Egan SE, Weinberg RA. The pathway to signal achievement [news]. Nature 1993; 365: 781-3. 10. Khosravi FR, Der CJ. The Ras signal transduction pathway [review]. [221 refs]. Cancer Metast Rev 1994; 13: 67-89. 11. Bourne HR, Sanders DA, McCormick E The GTPase supcrfamily: conserved structure and molecular mechanism [review]. Nature 1991; 349: 117-27. 12. Bollag G, McCormick E Regulators and effecters of ras proteins [review]. Ann Rev Cell Biol 1991; 7: 601-32. 13. Quilliam LA, Khosravi FR, Huff SY, Der CJ. Guanine nucleotide exchange factors: activators of the Ras superfamily of proteins [review]. Bioessays 1995; 17: 395404. 14. Casey PJ, Solski PA, Der CJ, Buss JE. p2lras is modified by a farnesyl isoprenoid. Proc Nat1 Acad Sci USA 1989; 86: 8323-7. 15. Willumsen BM, Christensen A, Hubbert NL, Papageorge AG, Lowy DR. The p21 ras C-terminus is required for transformation and membrane association. Nature 1984; 310: 5836. 16. Goldstein JL, Brown MS, Regulation of the mevalonate pathway [review]. Nature 1990; 343: 425-30. 17. Alberts AW. Lovastatin and simvastatin - inhibitors of HMG CoA reductase and cholesterol biosynthesis. Cardiology 1990; 4: 14-21. 18. Nakabayashi H, Taketa K, Yamane T, Miyazaki M, Miyano K, Sato J. Phenotypical stability of a human hepatoma cell line, HUH-~, in long-term culture with chemically detined medium. Gann 1984; 75: 151-8.
19. Morinaga T, Sakai M, Wegmann TG, Tamaoki T. Primary structures of human alpha-fetoprotein and its mRNA. Proc Nat1 Acad Sci USA 1983; 80: 4604-8. 20. Watanabe K, Saito A, Tamaoki T. Cell-specific enhancer activity in a far upstream region of the human alpha-fetoprotein gene. J Biol Chem 1987; 262: 4812-8. 21. Nakabayashi H, Watanabe K, Saito A, Otsuru A, Sawadaishi K, Tamaoki T. Transcriptional regulation of alpha-fetoprotein expression by dexamethasone in human hepatoma cells. J Biol Chem 1989; 264: 266-71. 22. Miura K, Kamiya H, Tominaga M, Inoue Y, Ikehara M, Noguchi S, et al. Overproduction of cellular and activated Ha-ras proteins by mutating a synthetic gene. Chem Pharm Bull 1987; 35: 4878-82. 23. Gorman C, Padmanabhan R, Howard BH. High efficiency DNA-mediated transformation of primate cells. Science 1983; 221: 551-3. 24. Kamiya H, Miura K, Ohtomo N, Koda T, Kakinuma M, Nishimura S, et al. Transformation of NIH3T3 cells with synthetic cHa-ras genes. Jpn J Cancer Rcs 1989; 80: 200-3. 25. Tsutsumi T, Ido A, Nakao K, Hamasaki K, Kato Y, Ohtsuru A, et al. Reciprocal regulation of alpha-fetoprotein and albumin gene expression by butyrate in human hepatoma cells. Gastroenterology 1994; 107: 499-504. 26. Gorman CM, Moffat LF, Howard BH. Recombinant genomes which express chloramphenicol acetyltransferase in mammalian cells. Mol Cell Biol 1982; 2: 1044-51. 27. Morris TJ, Palm SL, Pena J, Furcht LT, Buchwald H. Inhibition of hepatoma tissue culture-4 cell growth by lovastatin: a novel chemotherapeuic strategy. Surg Forum 1991; 42: 452-4. 28. Morris TJ, Palm SL, Furcht LL, Buchwald H. Effect of lovastatin alone and as an adjuvant chemotherapeutic agent on hepatoma tissue culture-4 cell growth. Ann Surg Oncol 1995; 2: 26674. 29. Barbacid M. ras genes [review]. Ann Rev Biochem 1987; 56: 779827. 30. Lowy DR, Willumsen BM. Function and regulation of ras [rcview]. Ann Rev Biochem 1993; 62: 851-91. 3 1. Sumi S, Beauchamp RD, Townsend CJ, Uchida T, Murakami M, Rajaraman S, et al. Inhibition of pancreatic adenocarcinoma cell growth by lovastatin. Gastroenterology 1992; 103: 982-9. 32. Ura H, Obara T, Nishino N, Tanno S, Okamura K, Namiki M. Cytotoxicity of simvastatin to pancreatic adenocarcinoma cells containing mutant ras gene. Jpn J Cancer Res 1994; 85: 633-8. 33. DeClue JE, Vass WC, Papageorge AG, Lowy DR, Willumsen BM. Inhibition of cell growth by lovastatin is independent of ras function. Cancer Res 1991; 51: 712-7. 34. Nakao K, Lawless D, Ohe Y, Miyao Y, Nakabayashi H, Kamiya H, et al. c-Ha-ras down-regulates the alpha-fetoprotein gene but not the albumin gene in human hepatoma cells. Mol Cell Biol 1990; 10: 1461-9. 35. Silberman S, Janulis M, Schultz RM. Characterization of downstream Ras signals that induce alternative protease-dependent invasive phenotypes. J Biol Chem 1997; 272: 5927-35. 36. Saegusa M, Ido A, Nakabayashi H, Saki M, Tamaoki T. Downregulation of human a-fetoprotein promoter by c-jun and c-fos. J Tumor Marker Oncol 1994; 272: 29-34. 37. Kubo Y, Okuda K, Musha H, Nakashima T Detection of hepatocellular carcinoma during a clinical follow-up of chronic liver disease: observations in 31 patients. Gastroenterology 1978; 74: 578-82. 38. Oka H, Kurioka N, Kim K, Kanno T, Kuroki T, Mizoguchi Y, et al. Prospective study of early detection of hepatocellular carcinoma in patients with cirrhosis. Hepatology 1990; 12: 680-7. 39. Shinagawa T Ohto M, Kimura K, Tsunetomi S, Morita M, Saisho H, et al. Diagnosis and clinical features of small hepatocellular carcinoma with emphasis on the utility of real-time ultrasonography. A study in 51 patients. Gastroenterology 1984; 86: 495502. 40. Bjersing L, Andersson C, Lithner E Hepatocellular carcinoma in
909
H. Mazume et al. patients from northern Sweden with acute intermittent porphyria: morphology and mutations. Cancer Epidemiol Biomarkers Prevention 1996; 5: 393-7. 41. Cui Q, Wang Z, Chen J, Liu T. Detection of c-Ki-ras oncogene codon 12 point mutations in lung carcinoma, gastric carcinoma and hepatic carcinoma [in Chinese]. Chung Kuo i Hsueh Ko Hsueh Yuan Hsueh Pao Acta Academiae Medicinae Sinicae 1996; 18: 95-8. 42. Leon M, Kew MC. Analysis of ras gene mutations in hepatocellular carcinoma in southern African blacks. Anticancer Res 1995; 15: 859-61. 43. Tada M, Omata M, Ohto M. Analysis of ras gene mutations in human hepatic malignant tumors by polymerase chain reaction and direct sequencing. Cancer Res 1990; 50: 11214.
910
44. Nonomura A, Ohta G, Hayashi M, Izumi R, Watanabe K, Takayanagi N, et al. Immunohistochemical detection of ras oncogene p21 product in liver cirrhosis and hepatocellular carcinoma. Am J Gastroenterol 1987; 82: 512-8. 45. Radosevich JA, Gould KA, Koukoulis GK, Haines GK, Rosen ST, Lee I, et al. Immunolocalization of ras oncogene ~21 in human liver diseases. Ultrastruct Path01 1993; 17: l-8. 46. Liu P, Flejou JF, Feldmann G. Bernuau D. Activation of ras oncogene in livers with cirrhosis. J Hepdtol 1994; 21: 1103-g. 47. Takeuchi T, Kagawa K, Hikitd H, Tada H, Deguchi T, Okanoue T, et al. The expression of ras ~21 product in hepdtocellular carcinoma [in Japanese]. Nippon Shokakibyo Gakkai Zasshi [Jpn J Gastroenterol] 1990; 87: 1686-91.