help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Takabe, K.
Right arrow Articles by Vale, W. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Takabe, K.
Right arrow Articles by Vale, W. W.
Endocrinology Vol. 140, No. 7 3125-3132
Copyright © 1999 by The Endocrine Society

ARTICLES

Interruption of Activin A Autocrine Regulation by Antisense Oligodeoxynucleotides Accelerates Liver Tumor Cell Proliferation1

Kazuaki Takabe2, Jean-Jacques Lebrun3, Yoji Nagashima, Yasushi Ichikawa, Masato Mitsuhashi, Nobuyoshi Momiyama, Takashi Ishikawa, Hiroshi Shimada and Wylie W. Vale4

The Clayton Foundation Laboratories for Peptide Biology (K.T., J.-J.L., W.V.), The Salk Institute, La Jolla, California 92037; Second Department of Surgery (K.T., Y.I., N.M., T.I.) and Department of Pathology (Y.N.), Yokohama City University School of Medicine, 3–9 Fukuura Kanazawa-ku Yokohama 236-0004, Japan; and Medical Science Division (M.M.), Hitachi Chemical Research Center, Irvine, California 92715

Address all correspondence and requests for reprints to: Wylie Vale, The Salk Institute, The Clayton Foundation for Peptide Biology, 10010 North Torrey Pines Road, La Jolla, California 92037.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Administration of activin A, a member of the transforming growth factor-ß superfamily inhibits hepatocyte proliferation in vitro and reduces liver mass in vivo. However, a role of endogenous activin A in local growth modulation has not been established in any system. The aim of this study was to examine the production of activin A in the human hepatoma cell line HLF and to explore a possible autocrine role of activin as a cell growth inhibitor by blocking production of endogenous activin using antisense oligodeoxynucleotides. Administration of exogenous activin A suppressed HLF cell growth, and immunoreactive activin A was shown to be produced in the cells at confluency by Western blotting analysis. Cells were exposed to phosphorothioate-modified oligodeoxynucleotides, synthesized with antisense or randomly shuffled base sequences of activin ßA subunit messenger RNA, under serum-free conditions. Uptake of the oligodeoxynucleotides into the cells was confirmed by use of fluorescein isothiocyanate-labeled oligodeoxynucleotides. Administration of antisense oligodeoxynucleotides reduced activin A production as confirmed by both competitive PCR and Western blotting. Activin ßA antisense oligodeoxynucleotides significantly increased cell proliferation compared with controls. These findings are consistent with the existence of an autocrine role of activin A as an inhibitor of hepatocyte proliferation.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
ACTIVINS, members of the transforming growth factor-ß (TGFß) superfamily are homodimeric proteins composed of two inhibin/activin ß-subunits linked by a disulfide bridge. Four ß-subunits have been identified in mammals so far: ßA, ßB, ßC, and ßE. Of the possible dimers, only ßAßA (activin A), ßAßB (activin AB), and ßBßB (activin B) have been isolated. The roles of ßC-subunit cloned from a human liver complementary DNA (cDNA) library (1) have yet to be determined. ßD- and ßE-subunits have so far been identified only in Xenopus and mouse, respectively (2, 3). Inhibins are heterodimeric molecules structurally related to activins, but oppose most of its activities (4). To date, {alpha}ßA (inhibin A) and {alpha}ßB (inhibin B) have been isolated as dimeric proteins.

Activin A acts as a multipotential growth and differentiation factor in many systems and is expressed in various organs during development and in acquired conditions (5). Activin effects on the liver have been well characterized. Administration of an excessive amount of exogenous activin A to rats in vivo markedly reduced liver mass, and the effect was reversible (6). Mice in which the inhibin {alpha}-subunit gene has been deleted exhibit severe liver dysplasia, presumably as a consequence of the effects of unrestrained and excessive production of activin A (7, 8). Finally, exogenous administration of activin A inhibits DNA synthesis (9) and cell proliferation (10) in a human hepatoma cell line, HepG2. The presence of messenger RNA (mRNA) for inhibin/activin ßA-subunit in hepatocytes suggests that either inhibin A and/or activin A is produced in the liver (11). Such locally produced activin A could function in the liver to inhibit cell proliferation in an autocrine manner.

TGFßs are also potent inhibitors of hepatocyte proliferation (12). TGFßs and activins act through similar signaling pathways involving heteromerization of two classes of receptor serine kinases and phosphorylation of common Smad proteins. Both normal hepatocytes (13) and hepatoma cells (HepG2) (14) produce TGFß; however, most TGFß in the liver appears to be produced by nonparenchymal cells (15).

Follistatin, an activin binding protein (16), has been shown to inhibit activin effects in the liver (11). Interpretation of results using follistatin are mitigated by questions regarding the specificity of follistatin as well as the observation that at least one isoform of this protein binds to cells (17) and may interact with activin and its target cells in a complex manner.

Antisense oligodeoxynucleotides (ODNs) can selectively and specifically inhibit target gene expression at the level of transcription. This strategy has been used extensively as a sensitive and specific tool to identify putative functions of endogenous proteins (18). We describe here the use of antisense oligodeoxynucleotides to explore the role of endogenous activin as an autocrine cell growth inhibitor of liver cells. We report that the activin A protein is produced endogenously in HLF cells as well as in the normal liver tissue. By using specific inhibin/activin ßA phosphorothioate antisense ODNs, our results indicate that blocking activin ßA mRNA and endogenous activin A protein expression increases the rate of hepatocyte proliferation. These data suggest that locally produced activin regulates hepatocyte growth in these tumor cells and raise the possibility that an autocrine loop may occur in normal liver.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
ODN sequences
Sequence and targeting sites on activin ßA mRNA of each ODN were designed on the basis of the DNA sequence of the human inhibin/activin ßA-subunit (19) using a computer program, HYBsimulator (Advanced Gene Computing Technologies, Irvine, CA) (20, 21) according to our previously published design strategy (22). Two candidate sequences were chosen from two different sites in activin ßA mRNA: Antisense 1(AS1); 5'-GCTGTAGGTTTGTGG-3' (position 14–29), and Antisense 2(AS2); 5'-GACAACTCTTGCTCC-3' (position 1541–1556). As controls, two 15-mer ODNs (Ctr1, 5'-ACTGATCCCGACTGA-3'; and Ctr2, 5'-ACTGATCCCGTGAGA-3') were synthesized in which A (adenine) and T (thymine) or G (guanine) and C (cytosine) were exchanged at random. Thus, the control of AS1 had the same length and GC content, but would not be expected to hybridize with the target gene. Phosphorothioate ODNs, which have a sulfur substituted for one of the nonbridging oxygens in the phosphate backbone, were chosen because of their enhanced stability in the presence of cellular and serum nucleases (23).

Cell Culture and human sample
A human hepatocellular carcinoma cell line, HLF, was obtained from the Japanese Cancer Resources Bank (Tokyo, Japan). Cells were cultured in DMEM supplemented with 10% FBS, 100 IU/ml penicillin, 100 µg/ml streptomycin, and 20 mM L-glutamine, under humidified conditions with 95% air/5% CO2 at 37 C. Cells were split every 3–4 days. Human liver samples were obtained from a 65-yr-old male undergoing hepatectomy at Yokohama City University Hospital after informed consent. The patient was free from cirrhosis.

RT-Competitive PCR (c-PCR)
Quantification of activin ßA-subunit mRNA was analyzed by RT-cPCR following a previously reported method (24). Total RNA was extracted by a modified acid guanidine method using TRIzol (Gibco BRL, Gaithersburg, MD). To eliminate residual genomic DNA from the RNA sample, RNA was treated with 0.1 µg/µl of deoxyribonuclease 1 (Gibco BRL) with 20 mM Tris-HCl (pH 8.4), 50 mM KCl, and 2 mM MgCl2, for 10 min at 37 C, and then inactivated by 2.5 mM EDTA. Randomly primed cDNA was reverse-transcribed from the isolated RNA by M-MLV reverse transcriptase (Gibco BRL) with 1x RT buffer, 0.25 mM deoxynucleoside triphosphate, 0.01 M dithiothreitol containing ribonuclease inhibitor, followed by PCR amplification in a thermal cycler (Perkin Elmer Corp., Norwalk, CT). The sense and antisense primers were 5'-CTTTGGCTGAGAGGATTTCT-3' and 5'-GGTGACATCGGGTCTCTTCT-3', respectively, which were designed following our previously published criteria (25). The activin A competitor cDNA template was synthesized to be identical to the amplified PCR except for the length. Serial dilution of the competitor was added to each tube together with 30 µg of total RNA. After 10 min of denaturation at 94 C, 30 cycles of PCR were started, 55 C for 1.5 min, 72 C for 4 min, and 94 C for 1.5 min. The PCR products were electrophoretically separated on 8% polyacrylamide gel followed by staining with ethidium bromide. The linearity of the assay was confirmed by 10-, 100-, and 1000-fold dilutions of a sample, respectively (data not shown).

Western blot analysis after immunoprecipitation
Cells were incubated in six-well plates for 24 h in serum-free DMEM before treatment with oligonucleotides as described. Cells or human liver sample were then lysed in 1 ml of lysis buffer (50 mM Tris-HCl, pH 7.5; 150 mM NaCl; 10% glycerol; 0.5% NP-40; 0.5% sodium deoxycholate) containing protease inhibitors (1 mM phenylmethylsulfonyl floride; 1 µg/ml pepstatin A; 2 µg/ml leupeptin; 5 µg/ml aprotinin) for 15 min at 4 C. The lysate was centrifuged at 12,000 rpm for 15 min, and the supernatant was transferred to a new tube. Cell lysates were immunoprecipitated overnight at 4 C with an antiactivin ßA antibody and 30 µl of protein A agarose beads. The affinity-purified rabbit antibody to amino acids 81–113 of porcine activin ßA-subunit has been previously characterized (26). Samples were then washed three times with 1 ml of washing buffer (50 mM Tris-HCl, pH 7.5; 2 mM EDTA; 150 mM NaCl; 10% glycerol) and eluted in 20 µl of SDS loading buffer (20% glycerol; 10% ß-mercaptoethanol; 4.6% SDS; 0.125 M Tris-HCl, pH 6.8). Proteins were electrophoretically separated on a 15% polyacrylamide gel, transferred onto nitrocellulose, and incubated with an anti-activin ßA antibody (at 0.5 µg/ml), overnight at 4 C. An antihuman TGFß1 antibody (Promega Corp., Madison, WI), antihuman TGFß2, and antihuman TGFß3 antibody (R&D Systems, Minneapolis, MN) were used for Western blotting of TGFß. For reference purposes, either 10 ng human TGFß1, TGFß2, or TGFß3 were coelectrophoresed in the respective gels. After incubation, the membranes were washed twice for 15 min in washing buffer (50 mM Tris-HCl, pH 7.6; 200 mM NaCl; 0.05% Tween 20) and incubated with a secondary antirabbit antibody conjugated with horseradish peroxidase (Amersham, Arlington Heights, IL; antirabbit Ig-HRP, NA 934) at a 4000-fold dilution for 1 h at room temperature. Finally, the membranes were washed four times for 15 min in the washing buffer, and immunoreactivity was detected by chemiluminescence (Amersham, ECL kit, RPN 2106) according to the manufacturer’s instructions.

Antisense or control ODNs treatment
Administration of the ODNs were performed according to our previous reports (27, 28). Briefly, the cells were seeded into 96-well plates at a density of 1 x 104 cells/100 µl/well for 24 h. The medium was replaced by serum-free DMEM for 24 h, and then antisense ODNs or control ODNs were added to the medium to reach a final concentration of 10 µM. Media with fresh ODNs were replaced daily.

Uptake of antisense ODNs into the cells
To examine whether ODNs will enter the cells, fluorescein isothiocyanate (FITC)-labeled phosphorothioate activin A antisense ODNs was exposed to the cells as described above. FITC-labeled ODNs were obtained from Nisshin Seifun, Inc. (Kanagawa, Japan). Twenty-four hours after the administration of FITC-labeled ODNs, the cells were washed thoroughly with PBS to eliminate extracellular ODNs and subjected to both phase-contrast and fluorescence microscopic observation.

Proliferation of HLF cells
The effect of exogenous activin A administration on HLF cell proliferation was studied by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay (29). The effects on HLF cell proliferation after antisense or control ODNs, follistatin, TGFß1, or additional activin A administration were measured by cell counting. Follistatin, activin, and TGFß1 were added to reach a final concentration of 2 nM, 1 nM, and 0.1 nM, respectively, in the serum-free medium with 1% BSA. The cells were harvested every 24 h, and the numbers of viable cells were counted in triplicate. Cell viability was routinely assessed by trypan blue dye exclusion test, and it was constantly more than 90% in all the experiments.

Statistical analysis
All values are expressed as mean ± SD. Statistical significance (defined as P < 0.01) was evaluated using Duncan’s multiple comparisons test.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Effect of exogenous activin A on HLF cell proliferation
Primary cultures of rat hepatocytes express activin ßA mRNA and are growth inhibited by exogenous administration of activin A (6, 11). Because not all hepatocellular carcinoma cell lines are growth inhibited by activin A (10), we first examined whether HLF cells are responsive to activin A. The MTT assay was used to evaluate the effect of exogenous activin A administration on HLF cells growing in logarithmic phase. As shown in Fig. 1Go, activin clearly limits HLF cell proliferation in a dose-dependent manner. The maximal effect was obtained by 1 nM of activin A, which correspond to a 25% blockage of the rate of cell proliferation compared with untreated cells.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 1. Dose-response relationship for activin A-induced changes in HLF cell growth. HLF cells were incubated for 5 days in DMEM containing 10% FBS in the presence of various concentrations of activin A. Cell growth was measured by MTT assay and is shown as percentage of cells cultured without activin A. Values are expressed as the mean ± SD from quadricate determinations.

 
Expression of endogenous activin A by HLF cells
Activin ßA mRNA is expressed in normal rat liver tissue, suggesting a local production of activin A (11). However, a previous study of several hepatocellular carcinoma cell lines (HepG2, HLE, and PLC/PRF/5) indicated that no detectable activin A was produced by these cells (10). To investigate whether HLF cells express endogenous activin A, cells were grown to 70 or 100% confluency and incubated 24 h in serum-free DMEM before being harvested. Proteins were immunoprecipitated and analyzed by Western blotting using an antiactivin ßA antibody. As shown in Fig. 2AGo, the ßA monomer was clearly detectable in HLF cells, suggesting the presence of endogenously produced activin ßA protein in these cells. Activin A expression increased when cells were grown to confluency, consistent with a role of activin in cell proliferation inhibition. This result may explain a previous report that exogenous activin A did not show effects at high density in HepG2 cells (10). Additional proteins of higher molecular weights were also detected by antiactivin ßA antibody in 100% confluent HLF cells. These bands were removed by the immunogen peptide (lane 3) and probably correspond to unprocessed ßA precursor. Additionally, expression of activin ßA monomer was compared between noncirrhotic human liver and HLF cells. As shown in Fig. 2BGo, the signal of the ßA monomer was easily detectable in normal liver tissue and of similar size to that from HLF cells.



View larger version (45K):
[in this window]
[in a new window]
  		Figure 2. Expression of endogenous activin A by HLF cells and liver tissue. HLF cells were grown to 70% or 100% confluency. Cells were incubated 24 h in DMEM without serum, then collected, lysed, immunoprecipitated, and analyzed by Western blotting with antiactivin ßA antibody (panel A). Noncirrhotic human liver tissue was compared with 100% confluent HLF cells (panel B). All the samples were adjusted to constant amounts of protein by Bradford assay. Twenty micrograms of immunogen peptide were added to 100% confluent sample before immunoprecipitation as a control (panel A, lane 3). Apparent molecular weights are indicated in the figures.

 
Length of ODNs
The effects of five kinds of activin A antisense ODNs (15, 17, 20, 22 and 24-mers) and PBS-treated HLF cells (PBS) were assessed. The results obtained at 72 h after ODN treatment are shown in Fig. 3Go. The sequence of the ODNs used was that of a 5'-untranslated region of the activin ßA mRNA. Of these, the 15-mer ODNs showed the greatest effect (40.9 ± 2.38) and increased cell numbers (P < 0.01) more than did the other ODNs. This result is consistent with our previous study on matrilysin antisense ODNs (28). Therefore, the 15-mer ODN (AS1) was used for subsequent experiments. Similar results were also obtained with ODNs from the 3'-untranslated region of the activin ßA (AS2).



View larger version (17K):
[in this window]
[in a new window]
  		Figure 3. Determination of the appropriate length of activin A antisense ODNs. HLF cells were seeded in 96-well plates at a density of 1 x 104 cells per 100 µl/well in DMEM with 10% FBS to culture for 24 h and replaced with serum-free DMEM. After 24 h incubation, 10 µM of activin A antisense ODNs with 15, 17, 20, 22, and 24-base length were administered (PBS in the figure indicates PBS-treated wells). Media were changed daily together with the oligomers. Cells were counted 72 h after the first administration. Values are the mean ± SD from triplicate determinations. The asterisk indicates a statistically significant difference (P < 0.01) between 15-mer ODNs given cells and the other cells.

 
Uptake of antisense ODNs into the HLF cells
To ensure that ODN penetration into the cell was not an issue (30), we followed the cellular localization of FITC-labeled ODNs. We have previously observed that amounts of phosphorothioate ODNs incorporated into the cells increased linearly within 24 h (28). Therefore, FITC luminescence was examined 24 h after administration of the ODNs. FITC-labeled ODNs were successfully incorporated into more than 80% of HLF cells in all wells examined in three independent experiments. As shown in Fig. 4Go, FITC-labeled ODNs accumulated in the nucleus as well as the cytoplasm. Similar results were obtained with FITC-labeled control ODNs (data not shown).



View larger version (29K):
[in this window]
[in a new window]
  		Figure 4. Uptake of antisense ODNs into HLF cells. HLF cells were seeded in 96-well plates at density of 1 x 104 cells/100 µl/well in DMEM with 10% FBS to culture for 24 h and replaced with serum-free DMEM. After 24 h incubation, FITC-labeled AS1 was administered to a final concentration of 10 µM. Both the phase-contrast photomicrograph (A) and the fluorescence photomicrograph (B) of the same HLF cells demonstrate that FITC-labeled AS1 are accumulated in the nucleus as well as the cytoplasm.

 
Effect of antisense ODNs on activin A mRNA expression
Activin ßA mRNA expression in HLF cells and the effect of antisense ODNs on this expression were examined. Neither Northern blotting analysis nor ribonuclease protection assays were successful in detecting activin ßA mRNA expression. Therefore, we used RT-cPCR to evaluate mRNA expression. As shown in Fig. 5Go, activin ßA mRNA expression was suppressed to 1 x 10-7 to 10-6 M in both AS1- and AS2-treated cells, compared from 1 x 10-5 M in control ODN-treated cells, 1 h after administration of the ODNs. To confirm that the amplified DNA is derived from activin ßA mRNA, the nucleotide sequence of the PCR product was determined by direct sequencing. Furthermore, to exclude the possibility of genomic DNA contamination, RNA samples were also amplified by PCR without RT that showed no notable band (data not shown).



View larger version (51K):
[in this window]
[in a new window]
  		Figure 5. Effect of activin A antisense ODNs on activin A mRNA expression. Cells treated with 10 µM of either activin A antisense ODNs or control ODNs were exposed to serum-free DMEM. mRNA was extracted from 1 x 105 cells by the procedure described at 1 h after each stimulation. cPCR was performed using activin A-specific primer pairs. The PCR products were electrophoresed on 8% polyacrylamide gel followed by staining with ethidium bromide.

 
Effect of antisense ODNs on activin A protein expression
To investigate whether activin ßA antisense ODNs suppress activin A protein expression in HLF cells, cells were incubated in serum-free conditions and treated with either PBS, control, or antisense ODNs. Activin ßA protein expression was analyzed by Western blotting. As previously observed (Fig. 2Go, lane 2), the activin ßA monomer was detected in nontreated HLF cells (Fig. 6AGo, lane 1). It was also detected in PBS-treated cells as shown in lane 2. Incubation of the cells with two different control ODNs did not affect the level of ßA monomer protein (Fig. 6AGo, lanes 3 and 4). However, the levels of activin ßA monomer are remarkably low when the cells are incubated with the two specific antisense ODNs (AS1, lane 5; and AS2, lane 6). An additional level of specificity was examined by testing the effects of the activin antisense ODNs on the expression of the related proteins, TGFß1, TGFß2, and TGFß3, as monitored by Western blotting. As shown in Fig. 6BGo, no change was observed in any of the TGFßs in response to any of the ODN treatments. These results confirm the specificity of our antisense ODNs and reaffirm the significance of their ability to block activin ßA expression in HLF cells.



View larger version (55K):
[in this window]
[in a new window]
  		Figure 6. Effect of activin A antisense ODNs on activin A and TGFß1, -2, and -3 protein expression. HLF cells were seeded in six-well plates at a density of 2 x 106cells per well in DMEM with 10% FBS to culture for 24 h and then replaced with serum-free DMEM. After 24 h incubation, 10 µM of PBS, Ctr1, Ctr2, AS1, and AS2 were administered (lanes 2–6). Media were changed every 6 h together with ODNs, and cells were collected 24 h after the first administration. Amount of protein loaded was adjusted by Bradford assay. Western blotting was carried out using antiactivin ßA antibody (A) or anti-TGFß1, -2, or -3 antibody (B), respectively.

 
Effect of activin A antisense ODNs, follistatin, and TGFß1 on HLF cell proliferation
To determine whether endogenous activin A regulates cell growth in an autocrine manner, antisense ODNs were used to block activin A production by HLF cells. Cell proliferation was then assessed by cell number counting as shown in Fig. 7AGo. The number of cells after antisense ODN treatment (AS1, AS2) increased 2.1-fold in 24 h and 2.5-fold in 48 h. This increase was significantly greater (P < 0.01) than that of the control ODN-treated cells (Ctr1, Ctr2). These data suggest that local activin A inhibits the proliferation of HLF cells in an autocrine manner. Additionally, when HLF cells were treated with the activin binding protein, follistatin, a similar increase in cell proliferation was observed (Fig. 7AGo).



View larger version (23K):
[in this window]
[in a new window]
  		Figure 7. Effect of antisense ODNs, follistatin, TGFß, or exogenous activin A on HLF cell proliferation. HLF cells were seeded in 96-well plates at a density of 1 x 104 cells/100 µl/well in DMEM with 10% FBS to culture for 24 h and replaced with serum-free DMEM. After 24 h incubation, 10 µM of activin A antisense ODNs (open circle, AS1; and closed circle, AS2), or control ODNs (open triangle, Ctr1; and closed square, Ctr2) were administered (0 h). To confirm the specificity of the ODNs, 2 nM of follistatin, 1 nM of activin A, or 0.1 nM of TGFß1 was added to AS1 (panel B), or AS2 (panel C). Media were changed daily together with ODNs. Viable cell number was counted daily by trypan blue dye exclusion method. Values were the mean ± SD from triplicate determinations. Single administration of AS1, AS2, or follistatin demonstrated significant increase compared with controls.

 
To further test the specificity of the antisense ODNs, we examined whether exogenous activin or TGFß could reverse the effects of antisense treatment (Fig. 7Go, B and C). We demonstrated that 1 nM exogenous activin or 0.1 nM TGFß1 completely reversed the effects of the two antisense ODNs (AS1 and AS2) on cell proliferation. These results indicate that the activin and TGFß signaling pathways are not compromised by the treatment with antisense ODNs. The addition of 2 nM follistatin did not induce extra growth stimulation, probably a reflection of the complete suppression of the production of activin by the antisense ODNs.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Gene expression can be specifically suppressed by delivery of antisense ODNs (31). However, a high level of specificity is sometimes difficult to achieve for the ODNs (18). In this study, we designed two antisense ODNs that specifically hybridize to sequences in the 5'-untranslated region and in the 3'-untranslated region of the activin ßA mRNA. Each sequence was carefully analyzed for its sequence similarity/dissimilarity and hybridizability against other published gene sequences to maximize sequence specificity. Hybridizability ({Delta}G) is based on the nearest neighbor calculation of the free hybridization energy between the antisense sequences and all other published gene sequences in the GenBank database (22). Therefore, both antisense ODNs (AS1 and AS2) were highly specific to activin ßA mRNA, but not to all other published nucleotide sequences including the members of TGFß superfamily (as summarized in Table 1Go). We first selected different ODNs with a variety of lengths and showed that the 15-mer antisense ODNs were most potent (Fig. 3Go). We also confirmed the uptake of the ODNs into the cells (Fig. 4Go). RT-cPCR and Western blotting demonstrated that activin ßA expression was reduced significantly by ODNs at both the mRNA and protein levels (Figs. 5Go and 6Go) indicating the ability of this approach to reduce activin ßA expression in HLF cells.


View this table:
[in this window]
[in a new window]
  		Table 1. Specificity of oligodeoxynucleotide sequences

 
We demonstrated in this study that activin A is locally produced by HLF cells. In addition, we showed that this endogenous activin A is capable of inhibiting cell proliferation in these cells. Previous studies used follistatin to block the antiproliferative effect of activin on hepatocytes (11, 16). Indeed, intraportal or intravenous administration of follistatin has been shown to accelerate DNA synthesis of the rat liver after hepatectomy (32, 33). However, it is known that follistatin also binds to the cell surface of hepatocytes (10, 34) and could exert separate effects on its own. Therefore, it may be possible that the proliferative effect of follistatin observed is due not only to inactivation of activin but also to the effect of follistatin itself. Moreover, it remains unclear whether follistatin binds to other members of the TGFß superfamily (35). Therefore, follistatin might not be the best tool to study the antiproliferative effects of activin. As an alternative to follistatin, we developed a technique using antisense ODNs to investigate the effect of specific inhibition of activin A expression in HLF cells. Using this approach we found that HLF proliferation was significantly increased by blocking endogenous activin A production. Our data revealed the existence of an autocrine loop of activin in HLF cells and indicate that this autocrine activin A appears to be a major factor accounting for the growth inhibition of HLF cells. Further, by comparing the effects of follistatin and antisense ODNs on HLF cell proliferation, we were able to confirm that follistatin indeed worked as an inhibitor of the autocrine effects of activin on cell proliferation.

Recently, new members of activin ß-subunits were cloned from a liver cDNA library (1, 3). Although the roles of ßC- and ßE-subunits have yet to be determined, it has been shown that both subunits are expressed in the liver (36, 37, 38). Further, activin ßC mRNA expression was shown to decrease as early as 6 h after 70% hepatectomy in mice and rats, which is slower than the sharp drop observed in the ßA-subunit (39, 40). Interestingly, ßC mRNA expression was continuously suppressed until 24 h (39) or 48 h (40), whereas ßA mRNA expression begins to increase by 12 h. It is feasible that the rapid decrease of endogenous activin A after hepatectomy might contribute to the events supporting proliferation of hepatocytes, and activin ßC expression might play a role in terminating liver regeneration.

Given that normal human liver tissue produces activin A (Fig. 2BGo), our results raise an intriguing possibility that blockage of autocrine activin A in normal liver would effectively accelerate hepatocyte proliferation. Treatments that block activin A activity, such as activin A antisense ODNs or follistatin, have potential possibility to be used for liver failure caused by fulminant hepatitis or excessive hepatectomy by means of increasing the liver tissue. For instance, preoperative enlargement of the remnant liver may enable extensive hepatectomy to be safely performed. Antisense ODNs could be therapeutically effective because the agent is biologically labile and exerts only transient effects, which should not result in hepatomegaly. Our antisense technology provides a potential possibility both as an investigative tool and as a therapeutic agent.


   Acknowledgments
 
The authors are thankful to Joan Vaughan for the preparation and purification of the antiactivin ßA antibody and to D. Hyndman (Advanced Gene Computing Technologies, Irvine, CA) for technical support. We also thank E. Wiater for critical reading of this manuscript.


   Footnotes
 
1 This work was supported by NIH Grant HD-13527, the Kleberg Foundation, the Adler Foundation, and the Foundation for Medical Research, Inc. Back

2 Supported by the Yoshida Scholarship Foundation. Back

3 Former Fellow of the Adler Foundation and currently a Medical Research Council Scholar. Back

4 Foundation for Medical Research, Inc., Senior Investigator. Back

Received September 10, 1998.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Hotten G, Neidhardt H, Schneider C, Pohl J 1995 Cloning of a new member of the TGF-ß family: a putative new activin ßC chain. Biochem Biophys Res Commun 206:608–613[CrossRef][Medline]
  2. Oda S, Nishimatsu S, Murakami K, Ueno N 1995 Molecular cloning and functional analysis of a new activin ß subunit: a dorsal mesoderm-inducing activity in Xenopus. Biochem Biophys Res Commun 210:581–588[CrossRef][Medline]
  3. Fang J, Yin W, Smiley E, Wang SQ, Bonadio J 1996 Molecular cloning of the mouse activin ßE subunit gene. Biochem Biophys Res Commun 228:669–674[CrossRef][Medline]
  4. Vale W, Hsueh A, Rivier C, Yu J 1990 The inhibin/activin family of hormones and growth factors. In: Sporn M, Roberts A (eds) Peptide Growth Factors and Their Receptors. Springer-Verlag, Berlin, pp 211–248
  5. Lebrun JJ, Chen Y, Vale WW 1997 Receptor serine kinases and signaling by activins and inhibins. In: Aono T, Sugino H, Vale WW (eds) Inhibin, Activin and Follistatin, Regulatory Functions in System and Cell Biology. Springer-Verlag, New York, pp 1–20
  6. Schwall RH, Robbins K, Jardieu P, Chang L, Lai C, Terrell TG 1993 Activin induces cell death in hepatocytes in vivo and in vitro. Hepatology 18:347–356[CrossRef][Medline]
  7. Matzuk MM, Finegold MJ, Mather JP, Krummen L, Lu H, Bradley A 1994 Development of cancer cachexia-like syndrome and adrenal tumors in inhibin-deficient mice. Proc Natl Acad Sci USA 91:8817–8821[Abstract/Free Full Text]
  8. Coerver KA, Woodruff TK, Finegold MJ, Mather J, Bradley A, Matzuk MM 1996 Activin signaling through activin receptor type II causes the cachexia-like symptoms in inhibin-deficient mice. Mol Endocrinol 10:534–543[Abstract]
  9. Xu J, McKeehan K, Matsuzaki K, McKeehan WL 1995 Inhibin antagonizes inhibition of liver cell growth by activin by a dominant-negative mechanism. J Biol Chem 270:6308–6313[Abstract/Free Full Text]
  10. Mashima H, Kanzaki M, Nobusawa R, Zhang YQ, Suzuki M, Mine T, Kojima I 1995 Derangements in the activin-follistatin system in hepatoma cells. Gastroenterology 108:834–840
  11. Yasuda H, Mine T, Shibata H, Eto Y, Hasegawa Y, Takeuchi T, Asano S, Kojima I 1993 Activin A: an autocrine inhibitor of initiation of DNA synthesis in rat hepatocytes. J Clin Invest 92:1491–1496
  12. Fausto N, Laird AD, Webber EM 1995 Role of growth factors and cytokines in hepatic regeneration. FASEB J 9:1527–1536[Abstract]
  13. Gao C, Gressner G, Zoremba M, Gressner A 1996 Transforming growth factor ß (TGF-ß) expression in isolated and cultured rat hepatocytes. J Cell Physiol 167:394–405[CrossRef][Medline]
  14. Derynck R, Goeddel D, Ullrich A, Gutterman J, Williams R, Bringman T, Berger W 1987 Synthesis of messenger RNAs for transforming growth factors {alpha} and ß and the epidermal growth factor receptor by human tumors. Cancer Res 47:707–712[Abstract/Free Full Text]
  15. Michalopoulos GK, DeFrances MC 1997 Liver regeneration. Science 276:60–66[Abstract/Free Full Text]
  16. Nakamura T, Takio K, Eto Y, Shibai H, Titani K, Sugino H 1990 Activin-binding protein from rat ovary is follistatin. Science 247:836–838[Abstract/Free Full Text]
  17. Hashimoto O, Nakamura T, Shoji H, Shimasaki S, Hayashi Y, Sugino H 1997 A novel role of follistatin, an activin-binding protein, in the inhibition of activin action in rat pituitary cells. Endocytotic degradation of activin and its acceleration by follistatin associated with cell-surface heparan sulfate. J Biol Chem 272:13835–13842[Abstract/Free Full Text]
  18. Wagner RW 1994 Gene inhibition using antisense oligodeoxynucleotides. Nature 372:333–335[CrossRef][Medline]
  19. Mason A, Niall HD, Seeburg PH 1986 Structure of two human ovarian inhibins. Biochem Biophys Res Commun 135:957–964[CrossRef][Medline]
  20. Mitsuhashi M, Cooper A, Ogura M, Shinagawa T, Yano K, Hosokawa T 1994 Oligonucleotide probe design–a new approach. Nature 367:759–761[CrossRef][Medline]
  21. Hyndman D, Cooper A, Pruzinsky S, Coad D, Mitsuhashi M 1996 Software to determine optimal oligonucleotide sequences based on hybridization simulation data. Biotechniques 20:1090–1094, 1096–1097[Medline]
  22. Mitsuhashi M 1997 Strategy for designing specific antisense oligonucleotide sequences. J Gastroenterol 32:282–287[Medline]
  23. Agrawal S, Akhtar S 1995 Advances in antisense efficacy and delivery. Trends Biotechnol 13:197–199[CrossRef][Medline]
  24. Gilliland G, Perrin S, Blanchard K, Bunn H 1990 Analysis of cytokine mRNA and DNA: detection and quantitation by competitive polymerase chain reaction. Proc Natl Acad Sci USA 87:2725–2729[Abstract/Free Full Text]
  25. Mitsuhashi M 1996 Technical report: Part 2. Basic requirements for designing optimal PCR primers. J Clin Lab Anal 10:285–293[CrossRef][Medline]
  26. Vaughan J, Rivier J, Corrigan A, McClintock R, Campen C, Jolley D, Voglmayr J, Bardin C, Rivier C, Vale W 1989 Detection and purification of inhibin using antisera generated against synthetic peptide fragments. In: Conn PM (ed) Methods in Enzymology. Academic Press, Orlando, FL, vol 168:588–617
  27. Momiyama N, Shimada H, Mitsuhashi M 1996 Suppression of c-jun by antisense oligonucleotides inhibits cell adhesion but not respiratory burst during phorbol ester-induced differentiation of U937 human monoblastic cells. Cell Growth Differ 7:1005–1012[Abstract]
  28. Momiyama N, Koshikawa N, Ishikawa T, Ichikawa Y, Hasegawa S, Nagashima Y, Mitsuhashi M, Miyazaki K, Shimada H 1998 Inhibitory effect of matrilysin antisense oligonucleotides on human colon cancer cell invasion in vitro. Mol Carcinog 22:57–63[CrossRef][Medline]
  29. Hansen MB, Nielsen SE, Berg K 1989 Re-examination and further development of a precise and rapid dye method for measuring cell growth/cell kill. J Immunol Methods 119:203–210[CrossRef][Medline]
  30. Lebleu B, Robbins I, Bastide L, Vives E, Gee J 1997 Pharmacokinetics of oligonucleotides in cell culture. In: Chadwick D, Cardew G (eds) Symposium on Oligonucleotides as Therapeutic Agents. John Wiley & Sons, London, vol 209:47–59
  31. Stein CA, Cheng YC 1993 Antisense oligonucleotides as therapeutic agents–is the bullet really magical? Science 261:1004–1012[Abstract/Free Full Text]
  32. Kogure K, Omata W, Kanzaki M, Zhang YQ, Yasuda H, Mine T, Kojima I 1995 A single intraportal administration of follistatin accelerates liver regeneration in partially hepatectomized rats. Gastroenterology 108:1136–1142[CrossRef][Medline]
  33. Kogure K, Zhang YQ, Kanzaki M, Omata W, Mine T, Kojima I 1996 Intravenous administration of follistatin: delivery to the liver and effect on liver regeneration after partial hepatectomy. Hepatology 24:361–366[CrossRef][Medline]
  34. Zhang YQ, Kanzaki M, Mashima H, Mine T, Kojima I 1996 Characterization of the activin receptor in cultured rat hepatocytes. Hepatology 24:446–450[CrossRef][Medline]
  35. Iemura S, Yamamoto T, Takagi C, Uchiyama H, Natsume T, Shimasaki S, Sugino II, Ueno N 1998 Direct binding of follistatin to a complex of bone-morphogenetic protein and its receptor inhibits ventral and epidermal cell fates in early Xenopus embryo. Proc Natl Acad Sci USA 95:9337–9342[Abstract/Free Full Text]
  36. Schmitt J, Hotten G, Jenkins NA, Gilbert DJ, Copeland NG, Pohl J, Schrewe H 1996 Structure, chromosomal localization, and expression analysis of the mouse inhibin/activin ßC (Inhbc) gene. Genomics 32:358–366[CrossRef][Medline]
  37. Lau AL, Nishimori K, Matzuk MM 1996 Structural analysis of the mouse activin ßC gene. Biochim Biophys Acta 1307:145–148[Medline]
  38. Fang J, Wang SQ, Smiley E, Bonadio J 1997 Genes coding for mouse activin ßC and ßE are closely linked and exhibit a liver-specific expression pattern in adult tissues. Biochem Biophys Res Commun 231:655–661[CrossRef][Medline]
  39. Esquela A, Zimmers T, Koniaris L, Sitzmann J, Lee S 1997 Transient down-regulation of inhibin-ßC expression following partial hepatectomy. Biochem Biophys Res Commun 235:553–556[CrossRef][Medline]
  40. Zhang YQ, Shibata H, Schrewe H, Kojima I 1997 Reciprocal expression of mRNA for inhibin ßC and ßA subunits in hepatocytes. Endocr J 44:759–764[Medline]



This article has been cited by other articles:


Home page
 Mol. Endocrinol.Home page
A. Lacerte, E.-H. Lee, R. Reynaud, L. Canaff, C. de Guise, D. Devost, S. Ali, G. N. Hendy, and J.-J. Lebrun
Activin Inhibits Pituitary Prolactin Expression and Cell Growth through Smads, Pit-1 and Menin
Mol. Endocrinol., June 1, 2004; 18(6): 1558 - 1569.
[Abstract] [Full Text] [PDF]

Home page
 CarcinogenesisHome page
S. Vejda, N. Erlach, B. Peter, C. Drucker, W. Rossmanith, J. Pohl, R. Schulte-Hermann, and M. Grusch
Expression of activins C and E induces apoptosis in human and rat hepatoma cells
Carcinogenesis, November 1, 2003; 24(11): 1801 - 1809.
[Abstract] [Full Text] [PDF]

Home page
 Endocr. Rev.Home page
G. P. Risbridger, J. F. Schmitt, and D. M. Robertson
Activins and Inhibins in Endocrine and Other Tumors
Endocr. Rev., December 1, 2001; 22(6): 836 - 858.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Takabe, K.
Right arrow Articles by Vale, W. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Takabe, K.
Right arrow Articles by Vale, W. W.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals