- Email: [email protected]
Norepinephrine reverses the effects of activin A on DNA synthesis and apoptosis in cultured rat hepatocytes
Norepinephrine Reverses the Effects of Activin A on DNA Synthesis and Apoptosis in Cultured Rat Hepatocytes YOU-QING ZHANG,1 MAKOTO KANZAKI,1 HIROSATO MASHIMA,1 TETSUYA MINE,2
Activin A, an autocrine factor produced by hepatocytes, inhibits mitogen-stimulated DNA synthesis and induces apoptotic death of cultured rat hepatocytes. Several lines of evidence indicate that norepinephrine (NE), as a comitogenic growth factor, alters the balance between growth stimulation and inhibition and acts as a trigger for the initiation of hepatocyte proliferation. In the present study, we examined whether NE modulated the effects of activin A on rat hepatocytes in primary culture. Activin A, at a concentration of 1009 mol/L, blocked the effect of epidermal growth factor (EGF) on DNA synthesis, that was assessed by measuring [3H] thymidine incorporation and nuclear labeling, almost completely, and NE reversed the inhibitory effect of activin A on DNA synthesis. This effect of NE was dosedependent, being significant at concentrations of 1006 mol/L and above, but was overcome by higher concentrations of activin A, and was attenuated by prazosin, but not by yohimbine or propranolol. NE exerted its effect during the first 24 hours of culture, but was ineffective when added after 24 hours. EGF augmented the release of follistatin, an activin-binding protein known to block the action of activin A, by hepatocytes and NE did not affect the amount of follistatin they released. In addition to inhibiting DNA synthesis by hepatocytes cultured with EGF, activin A induced death of hepatocytes cultured in the absence of EGF. The nuclear morphology of cells cultured with activin A alone was strikingly changed compared with untreated control cells and marked identation of the nuclear membranes and moderate chromatin condensation were observed. Fragmentation of DNA was also observed, suggesting that activin A induced apoptosis, and activin-mediated cell death was prevented significantly by NE. These results indicate that NE, acting on a1-adrenergic receptors, attenuates the effects of activin A on DNA synthesis by and Abbreviations: EGF, epidermal growth factor; NE, norepinephrine; EDTA, ethylenediamine-N,N,N*,N*-tetraacetic acid; BSA, bovine serum albumin; SDS, sodium dodecyl sulfate. From the 1Department of Cell Biology, Institute for Molecular and Cellular Regulation, Gunma University, Maebashi, Japan; and 2Fourth Department of Internal Medicine, University of Tokyo School of Medicine, Tokyo, Japan. Received August 9, 1994; accepted July 17, 1995. This study was supported by Grant-in-Aid for General and Developmental Scientific Research from the Ministry of Education, Science and Culture of Japan, and grants from the Kato Memorial Trust for Nanbyo Research and the Viral Hepatitis Research Foundation. Address reprint requests to: Itaru Kojima, M.D., Institute for Molecular and Cellular Regulation, Gunma University, Maebashi 371, Japan. Copyright q 1996 by the American Association for the Study of Liver Diseases. 0270-9139/96/2301-0013$3.00/0
01-24-96 21:43:57
hepa
KOJIMA1
apoptosis of cultured rat hepatocytes. (HEPATOLOGY 1996;23:288-293.)
Cellular proliferation is controlled by positive and negative growth regulators. Hepatocytes provide an ideal system for the study of cell growth because of special characteristics: they exist in a quiescent state in the normal adult animal, yet they are capable of extensive, coordinated proliferation following partial hepatectomy.1,2 Control of cell growth during liver regeneration has been studied extensively both in vivo and in vitro and synthesis of various growth stimulators of hepatocyte growth is induced during liver regeneration. Transforming growth factor a that shares the same receptors as epidermal growth factor (EGF), is synthesized as an autocrine factor in parenchymal liver cells,3 while hepatocyte growth factor is delivered via the endocrine mechanism and is synthesized in nonparenchymal cells,4-7 that subsequently synthesize transforming growth factor b, a potent inhibitor of hepatocyte growth.8 Norepinephrine (NE), as a comitogenic growth factor, has the potential to enhance the mitogenic effects of various growth stimulators. Blockade of a1-adrenergic receptors by prazosin attenuated the DNA synthesis peak seen during liver regeneration,9 that shows clearly that a1-adrenergic agonists play a critical role during the early stage of liver regeneration. NE also reduced the inhibitory effects of growth inhibitors such as transforming growth factor b, the median infective dose of which changed dramatically in the presence of NE.10 Reversal of the inhibitory effect of transforming growth factor b is, therefore, a potential mode of NE action. Recent studies in our laboratory showed that activin A is an autocrine negative regulator of DNA synthesis by rat hepatocytes and it is expressed in regenerating liver.11 Therefore, it may play a significant role in liver regeneration. Furthermore, Schwall et al. showed that activin A induced cell death both in vivo and in vitro,12 and apoptosis of parenchymal hepatocytes.13 It is, therefore, of interest to examine whether NE modulates these effects of activin A on hepatocytes. In the present study, we examined the effects of NE on activin-mediated cell death and inhibition of DNA synthesis by cultured rat hepatocytes. The results showed that NE, acting on a1-adrenergic receptors, reversed the effects of activin A on DNA synthesis and apoptosis.
288
m4730$0002
AND ITARU
WBS: Hepatology
HEPATOLOGY Vol. 23, No. 2, 1996
ZHANG ET AL.
MATERIALS AND METHODS Culture of Hepatocytes. Six-week-old male Wistar rats (about 170 to 200 g) were used for all the experiments. Parenchymal liver cells were isolated using the methods described by Berry and Friend,14 and plated on 24-well collagen-coated dishes at a density of 5 1 104/well in William’s E medium containing 10% fetal bovine serum, 1 nmol/L insulin, 10 nmol/ L dexamethasone, streptomycin, and penicillin, respectively. The cells were allowed to attach for 3 hours, after which the medium was changed to serum-free William’s medium containing 1 nmol/L dexamethasone, 0.1 nmol/L insulin, 0.05% bovine serum albumin, streptomycin, penicillin, and the required test agent(s). All the culture incubations were carried out at 377C under humidified conditions of 95% air and 5% CO2 . The experimental protocols were approved by the Institutional Animal Care and Use Committee, Institute for Molecular and Cellular Regulation, Gunma University, Maebashi, Japan. Measurement of DNA Synthesis. DNA synthesis was assessed in two ways. For measurement of the labeling index, hepatocytes were incubated, as described above, for a total of 72 hours. Bromodeoxyuridine (Amersham Corp., Arlington Heights, IL) was included from 48 to 72 hours and the labeled nuclei were stained with anti-Bromodeoxyuridine monoclonal antibody, according to the manufacturer’s instructions. Then the cells were stained with rabbit anti-mouse immunoglobulin G and the nuclear labeling was measured. To measure the [3H]thymidine incorporation into trichloroacetic acid-precipitable material, 0.5 mCi/mL [3H]thymidine was included from 48 to 72 hours and the [3H]thymidine incorporation was measured as described previously.11 The results were expressed as means { SE and statistical analysis was carried out using Student’s t test for paired data. Differences at P õ .05 were considered to be significant. Measurement of Follistatin Production. The follistatin content of the conditioned medium was measured using the method described by Saito et al.15 Briefly, samples or standard solutions of follistatin (200 mL) were added to glass tubes, to which 50 mL sulfate-cellulofine gel was added. After shaking at room temperature overnight, each suspension was centrifuged at 3000 rpm for 3 minutes, the supernatant was discarded and the gel was washed three times with 1 mL follistatin assay buffer comprising 50 mmol/L Tris/HCl (pH 7.4), 0.3 mol/L NaCl, 2 mmol/L ethylenediamine-N,N,N*,N*tetraacetic acid (EDTA), and 1% bovine serum albumin (BSA). Then, 125I–activin A solution was added to the tubes with or without unlabeled activin, the solutions were incubated overnight at 47C, the gels were washed with the assay buffer and the radioactivity was counted in a gamma counter.
289
were recovered by centrifugation at 13,0001 g for 20 minutes, resuspended in TE buffer (10 mmol/L Tris/HCl [pH 8.0] and 1 mmol/L EDTA) for 1 hour at 657C and then incubated with 200 mg/mL RNase A (Boehringer Mannheim, Mannheim, Germany) at 377C for 1 hour. The samples were electrophoresed in a 1.8% agarose gel, stained with ethidium bromide, visualized by exposure to ultraviolet light and photographed with an Amcel CRT camera (Tokyo, Japan). As a positive control for DNA fragmentation, HL-60 cells were treated with A23187 to induce apoptosis, the DNA was extracted and subjected to agarose gel electrophoresis, as described above. Electron Microscopy. For electron microscopy, the cells were fixed for 30 minutes at 47C with a solution comprising 2% paraformaldehyde and 2.5% glutaraldehyde in 0.1 mol/L phosphate buffer (pH 7.4), postfixed with 1% osmium tetroxide for 30 minutes, packed in 1% agar, dehydrated with an ascending ethanol series using routine procedures, and embedded in an Epon/araldite mixture. Ultrathin sections were cut, stained with uranyl acetate and lead citrate and observed with a JEM-1200 electron microscope (JEOL, Tokyo, Japan) at 80 keV. Materials. Recombinant human activin A and follistatin were provided by Dr Y. Eto, Ajinomoto, Inc. (Kawasaki, Japan). Collagenase (type XI), trypsin inhibitor, norepinephrine, prazosin hydrochloride, yohimbine hydrochloride, and propranolol were purchased from Sigma Chemical Co. (St. Louis, MO), EGF was from Collaborative Research (Lexington, MA), BSA (fraction V) was purchased from Seikagaku Kogyo (Tokyo, Japan), and [3H]thymidine was purchased from Dupont–New England Nuclear (Boston, MA). RESULTS Effects of Norepinephrine on Those of Activin A on DNA Synthesis and Cell Viability. Activin A acting on
its receptor system was found to be a potent inhibitor of EGF-induced DNA synthesis by cultured hepato-
Determination of Cell Viability and DNA Fragmentation. The viabilities of the cultured hepatocyte preparations
were assessed by staining cells with acridine orange/ethidium bromide, as described by Park et al.16 For measurement of DNA fragmentation, cells were collected by centrifugation and lysed by incubation for 3 hours at 507C in 0.7 mL lysis buffer comprising 10 mmol/L Tris/ HCl (pH 8.0), 10 mmol/L EDTA, 75 mmol/L NaCl, 0.5% sodium dodecyl sulfate (SDS), and 0.15 mg/ml proteinase K. The lysate was centrifuged for 20 minutes at 10,0001 g, the resulting supernatant was collected and the DNA preparations were extracted with phenol/chloroform followed by chloroform/isoamyl alcohol (48:1) to remove any protein and residual traces of phenol. The DNA was precipitated from the supernatant by adding 2 volumes of ethanol and maintaining the temperature at 0707C for 3 hours. The DNA precipitates
m4730$0002
01-24-96 21:43:57
hepa
FIG. 1. Dose-response relationship for the effect of norepinephrine on nuclear labeling. Cells were incubated with 1 nmol/L EGF and 1 nmol/L activin A in the presence of various doses of NE. Data represent the means { SE for three independent experiments each done in triplicate.
WBS: Hepatology
290 ZHANG ET AL.
HEPATOLOGY February 1996
cytes.11 Adding NE to the hepatocyte cultures reversed the inhibitory effect of activin A on nuclear labeling. Figure 1 shows the effects of various concentrations of NE on the labeling index of hepatocyte cultures treated with EGF and activin A. The labeling index of EGFtreated cells was 50.3 { 3.6% (mean { SE, n Å 3). When 1 nmol/L activin A was added together with EGF, the labeling index was 13.8 { 2.14% (n Å 3). NE reversed the inhibitory effect of activin A: in the presence of EGF, activin A and 1005 mol/L NE, the labeling index was 41.2 { 3.8% (n Å 3). This effect of NE was dosedependent and significant at concentrations of 1006 mol/L and above. Similar results were obtained when DNA synthesis was assessed by measuring [3H] thymidine incorporation (Table 1). In contrast, when activin A at concentrations of 10 nmol/L and above was added, NE did not reverse the inhibitory effect of activin A (data not shown). The acridine orange/ethidium bromide staining test showed that these cells were not dead. Schwall et al.12 reported recently that activin A induced hepatocyte death. They cultured hepatocytes in the absence of growth factor and administered activin A. In agreement with their results, activin A induced cell death of cultured hepatocytes in the absence of growth factor (Fig. 2). The dose-response relationship for activin A–induced cell death correlated with that for activin A–induced inhibition of DNA synthesis.11 Under our experimental conditions, however, approximately 50% of the hepatocytes were still alive in the presence of high concentrations of activin A. When the DNA extracted from activin-treated cells was subjected to agarose gel electrophoresis, a ladder pattern was observed, indicating that DNA fragmentation had occurred (Fig. 3). In contrast, virtually no DNA fragmentation was observed in hepatocytes incubated with EGF and activin A. Electron microscopic analysis revealed that the nuclear morphology of the activin A– treated cells had changed strikingly compared with control cells. The nuclear margins showed irregular indentations and the chromatin was moderately condensed (data not shown). As shown in Fig. 4, 1005 mol/L
TABLE 1. Effect of NE on Activin-Mediated Inhibition of [3H]Thymidine Incorporation [3H]Thymidine Incorporation (cpm 1 1003)
Addition
None EGF EGF / Activin Activin Activin Activin
activin A A / NE (1007 A / NE (1006 A / NE (1005 A / NE (1004
mol/L) mol/L) mol/L) mol/L)
5.2 36.0 11.6 12.9 21.6 26.4 27.9
{ { { { { { {
01-24-96 21:43:57
NE prevented activin-induced cell death but the prevention was not complete. The DNA fragmentation induced by activin A was reduced considerably by NE but slight fragmentation observed in cells incubated with activin A and NE, indicating that DNA fragmentation had occurred in at least some hepatocytes (Fig. 5). Time-Dependent Action of Norepinephrine. NE, as a comitogenic growth factor, was shown to enhance the mitogenic effect of EGF and, furthermore, it had to be present for a substantial portion of the first 24 hours of incubation to enhance DNA synthesis.17,18 We, therefore, measured the time at which NE exerted its action. As shown in Fig. 6, when NE was present from the beginning of the culture period, it reversed the inhibitory effect of activin A on DNA synthesis. However, when it was added after culture for 24 hours, it did not enhance DNA synthesis significantly. Note that NE was fully effective when added from 0 to 24 hours. Effect of Inhibitors of Adrenergic Receptor Antagonist. It is known that catecholamines are capable of
0.9 0.4 0.9 0.2* 3.9* 0.8* 3.5*
NOTE. Hepatocytes were cultured with 1 nmol/L EGF, 1 nmol/L activin A and various concentration of norepinephrine (NE). [3H]Thymidine incorporation was measured as described in Materials and Methods. Values are the means { SE for four determinations and the representative of three experiments. * P õ .01 vs. EGF / activin A.
m4730$0002
FIG. 2. Effect of activin A on viability of hepatocytes. Cells were cultured with various concentrations of activin A for 24 hours and viability of cells was determined by staining the cells with acridine orange/ethidium bromide. Values are the means { SE for three experiments.
hepa
stimulating DNA synthesis by cultured hepatocytes via interaction with a1-adrenergic receptors.17-19 Recent studies showed that b-adrenergic receptors also mediated such an effect, at least under some conditions.20 In our experiment, we investigated which types of adrenergic receptor mediated the activity of NE. Cultured hepatocytes were treated with 1 nmol/L EGF, 1 nmol/L activin A, 10 mmol/L NE, and various adrenergic receptor antagonists and the labeling index was measured using Bromodeoxyuridine. As shown in Table 2, NEinduced enhancement of DNA synthesis was abolished
WBS: Hepatology
HEPATOLOGY Vol. 23, No. 2, 1996
ZHANG ET AL.
291
by the a1-adrenergic receptor blocker prazosin (1006 mol/L), whereas either the a2-adrenergic receptor blocker yohimbine (1006 mol/L) nor the b-receptor blocker propranolol (1005 mol/L) had any effect. Likewise, NE-induced inhibition of cell death was prevented by prazosin (Table 3). Effect of Norepinephrine on Follistatin Production. Follistatin, an activin-binding protein, is known
to inhibit the effects of activin A.21 As follistatin is expressed in rat liver,22 it is possible that NE could reverse the effect of activin A by increasing follistatin production. Therefore, we investigated whether NE stimulated follistatin production. A significant amount of follistatin was released into the conditioned medium of hepatocytes in culture and EGF augmented its release (Table 4). Next, hepatocytes were cultured with NE in the presence or absence of EGF, the conditioned media were collected after 72 hours and the amount of follistatin released was measured. As shown in Table 4, EGF stimulated follistatin production significantly, whereas NE alone did not affect its output, neither did it enhance EGF-induced follistatin production. DISCUSSION
NE, as a comitogenic growth factor, is thought to play an important role in the initiation of DNA synthesis and modulation of various cellular functions during liver regeneration.9,10,17,18 Our present results showed that, in the presence of NE, the inhibitory effect of activin A on
DNA synthesis was reversed. The effect of NE was dosedependent and significant at concentrations of 1006 mol/L and above. A specific a1-adrenergic antagonist, prazosin, blocked this effect completely, whereas the a2-adrenergic antagonist yohimbine and b-receptor antagonist propranolol had no effect. Therefore, it is clear that the NE effect was mediated via a1-adrenergic receptors. It should be emphasized that the effect of NE on cultures treated with EGF was exerted during the first 24 hours in the presence and absence of activin A, that was because of at least partly the fact that a1-adrenergic receptor expression decreased after culture for 24 hours.23,24 Previous studies showed that the low levels of messenger RNA for the bA subunit of activin was expressed in intact liver, disappeared at 12 hours, and then increased markedly 24 hours after hepatectomy.11 Furthermore, blockade of the action of endogenous activin A accelerated liver regeneration and resulted in the appearance of nuclear labeling several hours earlier than in control rats.25 Therefore, activin A expressed in the remnant liver after partial hepatectomy would appear to exert tonic inhibition on the initiation of DNA synthesis. Given that NE is released from sympathetic nerve terminals shortly after partial hepatectomy, reversal of this inhibitory action of activin A may be an important mode of action of NE. Recent studies in our laboratory showed that EGF stimulated the production of follistatin, an activinbinding protein, by cultured hepatocytes.22 As follistatin enhanced EGF-mediated DNA synthesis and bound to activin A in a stoichiometric manner to neutralize the effect of activin significantly,21 we investi-
FIG. 3. Fragmentation of DNA in activin-treated cells. Cells were incubated for 24 hours with 1 nmol/L activin A. DNA was extracted, separated by agarose gel and visualized by ethidium bromide. DNA fragmentation was also examined in HL-60 cells treated with A23187 as a positive control. Lane 1, EGF / activin A; lane 2, activin A; lane 3, HL-60 cells treated with 1 mmol/L A23187.
m4730$0002
FIG. 4. Effect of norepinephrine on activin-induced cell death. Cells were incubated for 24 hours with 1 nmol/L EGF or 10 mmol/L NE in the presence and absence of 1 nmol/L activin A. Viability of cells was analyzed by using acridine orange/ethidium bromide. Values are the means { SE for four experiments. *P õ .01 vs. EGF, **P õ .01 vs. activin A.
01-24-96 21:43:57
hepa
WBS: Hepatology
292 ZHANG ET AL.
HEPATOLOGY February 1996 TABLE 2. Effects of Adrenergic Antagonists on the Effect of Norepinephrine on DNA Synthesis Addition
Control EGF EGF / Activin A / NE / NE / Prazosin / NE / Yohimbine / NE / Propranolol
Labeled Nuclei (%)
8.9 52.5 12.9 42.0 10.7 41.5 42.8
{ { { { { { {
1.8 3.7 0.9 3.3 0.8 3.4 2.5
NOTE. Hepatocytes were incubated with 1 nmol/L EGF, 1 nmol/L activin A, and 1005 mol/L NE in the presence of prazosin (1006 mol/L), yohimbine (1006 mol/L) or propranolol (1005 mol/L). Labeling index was measured by using BrdU. Values are the means { SE for three different experiments each done in triplicate.
FIG. 5. Effect of norepinephrine on activin-induced DNA fragmentation. Cells were incubated for 24 hours with 1 nmol/L activin A in the presence and absence of 10 mmol/L norepinephrine. Then DNA was extracted, separated by agarose gel and visualized by ethidium bromide. Lane 1, activin A; lane 2, activin / norepinephrine.
gated whether NE increased follistatin production. The results refuted the hypothesis that NE acted by stimulating follistatin synthesis. Schwall et al. reported that activin A, in the absence of growth factors, induced death of cultured hepatocytes.12 More recently, investigators reported that activin A induced apoptosis of cultured hepatocytes, that was adjudged by the presence of DNA fragmentation and typical morphological alterations.13 Our present results confirmed their data, showing that activin A induced DNA fragmentation and drastic changes in nu-
clear morphology, including moderate chromatin condensation. Taken together, these results suggest that activin A attenuates cell-cycle progression in the presence of appropriate growth stimulation, but, in the absence of growth factors, it induces apoptosis of cultured hepatocytes. Consistent with this notion, activin A administered intravenously 11 hours after partial hepatectomy delayed the initiation of DNA synthesis without causing apoptosis in the remnant liver.25 By acting on a1-adrenergic receptors, NE attenuated activin A– induced apoptosis, and although NE alone did not stimulate hepatocyte proliferation, it attenuated apoptosis induced by activin A. It has been established that activation of protein kinase C, by adding phorbol ester or an agonist that causes phosphoinositide breakdown, protects thymocytes against apoptosis.26,27 As activation of a1-adrenergic receptors results in hydrolysis of phosphoinositides leading to protein kinase C activation,28 it is possible that NE protects cells against apoptosis by a protein kinase C–dependent mechanism. In a previous study, we showed that activin A augmented glycogenolysis by causing breakdown of phosphoinosides in isolated rat hepatocytes.29 Despite the fact that activin A activated the calcium messenger system in freshly isolated hepatocytes, it lost its ability to cause hydrolysis of phosphoinositides in hepatocytes cultured for longer than 3 hours (unpublished observation). Although the precise mechanism responsible for
TABLE 3. Effects of Adrenergic Antagonists on the Effect of Norepinephrine on Cell Viability Addition
Activin Activin Activin Activin
FIG. 6. Effect of norepinephrine added at different time points. Hepatocytes were incubated with 1 nmol/L EGF with or without and 1 nmol/L activin A. NE was added at either 0 or 24 hours. Data represent the means { SE of experiments each done in triplicate.
m4730$0002
01-24-96 21:43:57
hepa
A A / NE A / NE / Prazosin A / NE / Yohimbine
Cell Viability (%)
61.0 81.0 63.3 82.0
{ { { {
2.8 2.1 3.3 3.2
NOTE. Hepatocytes were incubated with 1 nmol/L activin A and 1005 mol/L NE in the presence and absence of prazosin (1006 mol/L) or yohimbine (1006 mol/L). Cell viability was determined as described in Materials and Methods. Values are the means { SE for four experiments.
WBS: Hepatology
HEPATOLOGY Vol. 23, No. 2, 1996
ZHANG ET AL.
TABLE 4. Follistatin Released to the Medium of Cultured Hepatocytes Additions
Follistatin (ng/mL)
None NE EGF EGF / NE
7.0 6.8 30.9 26.2
{ { { {
0.9 0.6 4.5 2.8
NOTE. Hepatocytes were cultured with or without 1 nmol/L EGF in the presence of 1005 mol/L NE. Conditioned medium was collected at 72 hours. Values are the means { SE for 12 determinations from two experiments.
this refractoriness is uncertain at present, uncoupling of the activin receptor system to phospholipase C does appear to occur in cultured hepatocytes. Therefore, it seems unlikely that activin A activates protein kinase C in cultured hepatocytes. In conclusion, NE, acting on a1-adrenergic receptors, modulated the inhibitory actions of activin A on DNA synthesis by cell viability of rat cultured hepatocytes. These data, together with those of Michalopoulos and colleagues,9,10,17,18,30 suggest that NE plays a very important role in the initiation of the early DNA synthesis response during liver regeneration. Acknowledgment: The authors are grateful to Romi Nobusawa and Kiyomi Ohgi for secretarial assistance during the preparation of the report. REFERENCES 1. Michalopoulos GK. Liver regeneration: molecular mechanisms of growth control. FASEB J 1990;4:176-187. 2. Rabes HM, Wirshing R, Tuczek HV, Iseler, G. Analysis of cell cycle compartments of hepatocytes after partial hepatectomy. Cell Tissue Res 1976;6:517-532. 3. Mead JE, Fausto N. Transforming growth factor a may be a physiological regulator of liver regeneration by means of an autocrine mechanism. Proc Natl Acad Sci U S A 1989;86:1555815562. 4. Nakamura T, Nishizawa T, Hagiya M, Seki T, Shimonishi M, Sugiyama A, Tashiro K, et al. Molecular cloning and expression of human hepatocyte growth factor. Nature 1989;342:440-442. 5. Kinoshita T, Hirao S, Matsumoto K, Nakamura T. Possible endocrine control by hepatocyte growth factor of Liver Regeneration after partial hepatectomy. Biochem Biophys Res Commun 1991;177:330-335. 6. Noji S, Tashiro K, Koyama E, Nohno T, Ohyama K, Taniguchi S, Nakamura T. Expression of hepatocyte growth factor gene in endothelial and Kupffer cells of damaged rat liver, as revealed by in situ hybridization. Biochem Biophys Res Commun 1990;173:42-47. 7. Hamanoue M, Kawaida K, Takao S, Shimazu H, Noji S, Matsumoto K, Nakamura T. Rapid and marked induction of hepatocyte growth factor during liver regeneration after ischemic or crush injury. HEPATOLOGY 1992;16:1485-1492. 8. Braun L, Mead JE, Panzica M, Mikumo R, Bell GI, Fausto N. Transforming growth factor b mRNA increases during liver regeneration: a possible paracrine mechanism of growth regulation. Proc Natl Acad Sci U S A 1988;85:1539-1543. 9. Curise JL, Knechtle SJ, Bollinger RR, Kuhn C, Michalopoulos GK. a1-adrenergic effects and liver regeneration. HEPATOLOGY 1971;7:1189-1194.
m4730$0002
01-24-96 21:43:57
hepa
293
10. Houck KA, Cruise JL, Michalopoulos G. Norepinephrine modulates the growth-inhibitory effect of transforming growth factorbeta in primary rat hepatocyte cultures. J Cell Physiol 1988;135: 551-555. 11. Yasuda H, Mine T, Shibata H, Eto Y, Hasegawa Y, Takeuchi T, Asano S, et al. Activin A: an autocrine inhibitor of initiation of DNA synthesis in rat hepatocytes. J Clin Invest 1993;92:14911496. 12. Schwall RH, Robbins K, Jaedieu P, Chang L, Lai G, Terrell TG. Activin induces cell death in hepatocytes in vivo and in vitro. HEPATOLOGY 1993;18:347-356. 13. Hully JR, Chang L, Schwall RH, Widmer HR, Terrell TG, Gillett NA. Induction of apoptosis in the murine liver with recombinant human activin A. HEPATOLOGY 1994;20:854-861. 14. Berry MN, Friend DS. High-yield preparation of isolated rat liver parenchymal cells. J Cell Biol 1969;43:506-520. 15. Saito S, Nakamura T, Titani K, Sugino H. Production of activinbinding protein by rat granulosa cells in vitro. Biochem Biophys Res Commun 1991;176:413-422. 16. Park DR, Bryan VM, Oi VT, Herzeuberg LA. Antigen specific identification and cloning of hybridomas with a fluorescenceactivated cell sorter. Proc Natl Acad Sci U S A 1979;76:19621966. 17. Cruise JL, Michalopoulos G. Norepinephrine and epidermal growth factor: dynamics of their interaction in stimulation of hepatocyte DNA synthesis. J Cell Physiol 1985;125:45-50. 18. Cruise JL, Houck KA, Michalopoulos GK. Induction of DNA synthesis in cultured rat hepatocytes through stimulation of a1adrenoreceptor by norepinephrine. Science 1985;227:749-751. 19. Takai S, Nakamura T, Komi N, Ichihara A, Corrigan A, Woo WK. Mechanisms of stimulation of DNA synthesis induced by epinephrine in primary cultures of adult rat hepatocytes. J Biochem 1988;103:848-852. 20. Refsnes M, Thoresen GH, Sandnes D, Dajani DF, Dajani L, Christoffersen T. Stimulatory and inhibitory effect of catecholamines on DNA synthesis in primary rat hepatocyte cultures: role of a1- and b-adrenergic mechanisms. J Cell Physiol 1992;151: 164-171. 21. Nakamura T, Takio K, Eto Y, Shibai H, Titani K, Sugino H. Activin-binding protein in rat ovary is follistatin. Science 1990;247:836-838. 22. Kanzaki M, Zhang YQ, Yasuda H, Mine T, Kojima I. Production of follistatin in cultured rat hepatocytes. Biochem Biophys Res Commun 1994;202:422-428. 23. Noda C, Nakamura T, Ichihara A. a-Adrenergic regulation of enzymes of amino acid metabolism in primary cultures of adult rat hepatocytes. J Biol Chem 1983;258:1520-1525. 24. Nakamura T, Tomomura A, Kato S, Noda C, Ichihara A. Reciprocal expression of a,- and b-adrenergic receptors, but constant expression of glucagon receptor by rat hepatocytes during development and primary culture. J Biochem 1984;96:127-136. 25. Kogure K, Omata W, Kanzaki M, Zhang YQ, Yasuda H, Mine T, Kojima I. A single intraportal administration of follistatin accelerates liver regeneration in partially hepatectomized rats. Gastroenterology 1995;108:1136-1142. 26. McConkey DJ, Hartzell P, Amador-Perez JF, Orrenius S, Jondal M. Calcium-dependent killing of immature thymocytes by stimulation via the CD3/T cell receptor complex. J Immunol 1989;143: 1801-1806. 27. McConkey DJ, Hartzell P, Jondal M, Orrenius S. Inhibition of DNA fragmentation in thymocytes and isolated nuclei by agents that stimulate protein kinase C. J Biol Chem 1989;264:1339913402. 28. Williamson JR, Cooper RH, Joseph SK, Thomas AP. Inositol triphosphate and diacylglycerol as intracellular second messengers in liver. Am J Physiol 1985;248:C203-C216. 29. Mine T, Kojima I, Ogata E. Stimulation of glucose production by activin A in isolated rat hepatocytes. Endocrinology 1989;125: 586-591. 30. Houck KA, Michalopoulos GK. Altered response of regenerating hepatocytes to norepinephrine and transforming growth factor type b. J Cell Physiol 1989;141:503-509.
WBS: Hepatology
Activin A, an autocrine factor produced by hepatocytes, inhibits mitogen-stimulated DNA synthesis and induces apoptotic death of cultured rat hepatocytes. Several lines of evidence indicate that norepinephrine (NE), as a comitogenic growth factor, alters the balance between growth stimulation and inhibition and acts as a trigger for the initiation of hepatocyte proliferation. In the present study, we examined whether NE modulated the effects of activin A on rat hepatocytes in primary culture. Activin A, at a concentration of 1009 mol/L, blocked the effect of epidermal growth factor (EGF) on DNA synthesis, that was assessed by measuring [3H] thymidine incorporation and nuclear labeling, almost completely, and NE reversed the inhibitory effect of activin A on DNA synthesis. This effect of NE was dosedependent, being significant at concentrations of 1006 mol/L and above, but was overcome by higher concentrations of activin A, and was attenuated by prazosin, but not by yohimbine or propranolol. NE exerted its effect during the first 24 hours of culture, but was ineffective when added after 24 hours. EGF augmented the release of follistatin, an activin-binding protein known to block the action of activin A, by hepatocytes and NE did not affect the amount of follistatin they released. In addition to inhibiting DNA synthesis by hepatocytes cultured with EGF, activin A induced death of hepatocytes cultured in the absence of EGF. The nuclear morphology of cells cultured with activin A alone was strikingly changed compared with untreated control cells and marked identation of the nuclear membranes and moderate chromatin condensation were observed. Fragmentation of DNA was also observed, suggesting that activin A induced apoptosis, and activin-mediated cell death was prevented significantly by NE. These results indicate that NE, acting on a1-adrenergic receptors, attenuates the effects of activin A on DNA synthesis by and Abbreviations: EGF, epidermal growth factor; NE, norepinephrine; EDTA, ethylenediamine-N,N,N*,N*-tetraacetic acid; BSA, bovine serum albumin; SDS, sodium dodecyl sulfate. From the 1Department of Cell Biology, Institute for Molecular and Cellular Regulation, Gunma University, Maebashi, Japan; and 2Fourth Department of Internal Medicine, University of Tokyo School of Medicine, Tokyo, Japan. Received August 9, 1994; accepted July 17, 1995. This study was supported by Grant-in-Aid for General and Developmental Scientific Research from the Ministry of Education, Science and Culture of Japan, and grants from the Kato Memorial Trust for Nanbyo Research and the Viral Hepatitis Research Foundation. Address reprint requests to: Itaru Kojima, M.D., Institute for Molecular and Cellular Regulation, Gunma University, Maebashi 371, Japan. Copyright q 1996 by the American Association for the Study of Liver Diseases. 0270-9139/96/2301-0013$3.00/0
01-24-96 21:43:57
hepa
KOJIMA1
apoptosis of cultured rat hepatocytes. (HEPATOLOGY 1996;23:288-293.)
Cellular proliferation is controlled by positive and negative growth regulators. Hepatocytes provide an ideal system for the study of cell growth because of special characteristics: they exist in a quiescent state in the normal adult animal, yet they are capable of extensive, coordinated proliferation following partial hepatectomy.1,2 Control of cell growth during liver regeneration has been studied extensively both in vivo and in vitro and synthesis of various growth stimulators of hepatocyte growth is induced during liver regeneration. Transforming growth factor a that shares the same receptors as epidermal growth factor (EGF), is synthesized as an autocrine factor in parenchymal liver cells,3 while hepatocyte growth factor is delivered via the endocrine mechanism and is synthesized in nonparenchymal cells,4-7 that subsequently synthesize transforming growth factor b, a potent inhibitor of hepatocyte growth.8 Norepinephrine (NE), as a comitogenic growth factor, has the potential to enhance the mitogenic effects of various growth stimulators. Blockade of a1-adrenergic receptors by prazosin attenuated the DNA synthesis peak seen during liver regeneration,9 that shows clearly that a1-adrenergic agonists play a critical role during the early stage of liver regeneration. NE also reduced the inhibitory effects of growth inhibitors such as transforming growth factor b, the median infective dose of which changed dramatically in the presence of NE.10 Reversal of the inhibitory effect of transforming growth factor b is, therefore, a potential mode of NE action. Recent studies in our laboratory showed that activin A is an autocrine negative regulator of DNA synthesis by rat hepatocytes and it is expressed in regenerating liver.11 Therefore, it may play a significant role in liver regeneration. Furthermore, Schwall et al. showed that activin A induced cell death both in vivo and in vitro,12 and apoptosis of parenchymal hepatocytes.13 It is, therefore, of interest to examine whether NE modulates these effects of activin A on hepatocytes. In the present study, we examined the effects of NE on activin-mediated cell death and inhibition of DNA synthesis by cultured rat hepatocytes. The results showed that NE, acting on a1-adrenergic receptors, reversed the effects of activin A on DNA synthesis and apoptosis.
288
m4730$0002
AND ITARU
WBS: Hepatology
HEPATOLOGY Vol. 23, No. 2, 1996
ZHANG ET AL.
MATERIALS AND METHODS Culture of Hepatocytes. Six-week-old male Wistar rats (about 170 to 200 g) were used for all the experiments. Parenchymal liver cells were isolated using the methods described by Berry and Friend,14 and plated on 24-well collagen-coated dishes at a density of 5 1 104/well in William’s E medium containing 10% fetal bovine serum, 1 nmol/L insulin, 10 nmol/ L dexamethasone, streptomycin, and penicillin, respectively. The cells were allowed to attach for 3 hours, after which the medium was changed to serum-free William’s medium containing 1 nmol/L dexamethasone, 0.1 nmol/L insulin, 0.05% bovine serum albumin, streptomycin, penicillin, and the required test agent(s). All the culture incubations were carried out at 377C under humidified conditions of 95% air and 5% CO2 . The experimental protocols were approved by the Institutional Animal Care and Use Committee, Institute for Molecular and Cellular Regulation, Gunma University, Maebashi, Japan. Measurement of DNA Synthesis. DNA synthesis was assessed in two ways. For measurement of the labeling index, hepatocytes were incubated, as described above, for a total of 72 hours. Bromodeoxyuridine (Amersham Corp., Arlington Heights, IL) was included from 48 to 72 hours and the labeled nuclei were stained with anti-Bromodeoxyuridine monoclonal antibody, according to the manufacturer’s instructions. Then the cells were stained with rabbit anti-mouse immunoglobulin G and the nuclear labeling was measured. To measure the [3H]thymidine incorporation into trichloroacetic acid-precipitable material, 0.5 mCi/mL [3H]thymidine was included from 48 to 72 hours and the [3H]thymidine incorporation was measured as described previously.11 The results were expressed as means { SE and statistical analysis was carried out using Student’s t test for paired data. Differences at P õ .05 were considered to be significant. Measurement of Follistatin Production. The follistatin content of the conditioned medium was measured using the method described by Saito et al.15 Briefly, samples or standard solutions of follistatin (200 mL) were added to glass tubes, to which 50 mL sulfate-cellulofine gel was added. After shaking at room temperature overnight, each suspension was centrifuged at 3000 rpm for 3 minutes, the supernatant was discarded and the gel was washed three times with 1 mL follistatin assay buffer comprising 50 mmol/L Tris/HCl (pH 7.4), 0.3 mol/L NaCl, 2 mmol/L ethylenediamine-N,N,N*,N*tetraacetic acid (EDTA), and 1% bovine serum albumin (BSA). Then, 125I–activin A solution was added to the tubes with or without unlabeled activin, the solutions were incubated overnight at 47C, the gels were washed with the assay buffer and the radioactivity was counted in a gamma counter.
289
were recovered by centrifugation at 13,0001 g for 20 minutes, resuspended in TE buffer (10 mmol/L Tris/HCl [pH 8.0] and 1 mmol/L EDTA) for 1 hour at 657C and then incubated with 200 mg/mL RNase A (Boehringer Mannheim, Mannheim, Germany) at 377C for 1 hour. The samples were electrophoresed in a 1.8% agarose gel, stained with ethidium bromide, visualized by exposure to ultraviolet light and photographed with an Amcel CRT camera (Tokyo, Japan). As a positive control for DNA fragmentation, HL-60 cells were treated with A23187 to induce apoptosis, the DNA was extracted and subjected to agarose gel electrophoresis, as described above. Electron Microscopy. For electron microscopy, the cells were fixed for 30 minutes at 47C with a solution comprising 2% paraformaldehyde and 2.5% glutaraldehyde in 0.1 mol/L phosphate buffer (pH 7.4), postfixed with 1% osmium tetroxide for 30 minutes, packed in 1% agar, dehydrated with an ascending ethanol series using routine procedures, and embedded in an Epon/araldite mixture. Ultrathin sections were cut, stained with uranyl acetate and lead citrate and observed with a JEM-1200 electron microscope (JEOL, Tokyo, Japan) at 80 keV. Materials. Recombinant human activin A and follistatin were provided by Dr Y. Eto, Ajinomoto, Inc. (Kawasaki, Japan). Collagenase (type XI), trypsin inhibitor, norepinephrine, prazosin hydrochloride, yohimbine hydrochloride, and propranolol were purchased from Sigma Chemical Co. (St. Louis, MO), EGF was from Collaborative Research (Lexington, MA), BSA (fraction V) was purchased from Seikagaku Kogyo (Tokyo, Japan), and [3H]thymidine was purchased from Dupont–New England Nuclear (Boston, MA). RESULTS Effects of Norepinephrine on Those of Activin A on DNA Synthesis and Cell Viability. Activin A acting on
its receptor system was found to be a potent inhibitor of EGF-induced DNA synthesis by cultured hepato-
Determination of Cell Viability and DNA Fragmentation. The viabilities of the cultured hepatocyte preparations
were assessed by staining cells with acridine orange/ethidium bromide, as described by Park et al.16 For measurement of DNA fragmentation, cells were collected by centrifugation and lysed by incubation for 3 hours at 507C in 0.7 mL lysis buffer comprising 10 mmol/L Tris/ HCl (pH 8.0), 10 mmol/L EDTA, 75 mmol/L NaCl, 0.5% sodium dodecyl sulfate (SDS), and 0.15 mg/ml proteinase K. The lysate was centrifuged for 20 minutes at 10,0001 g, the resulting supernatant was collected and the DNA preparations were extracted with phenol/chloroform followed by chloroform/isoamyl alcohol (48:1) to remove any protein and residual traces of phenol. The DNA was precipitated from the supernatant by adding 2 volumes of ethanol and maintaining the temperature at 0707C for 3 hours. The DNA precipitates
m4730$0002
01-24-96 21:43:57
hepa
FIG. 1. Dose-response relationship for the effect of norepinephrine on nuclear labeling. Cells were incubated with 1 nmol/L EGF and 1 nmol/L activin A in the presence of various doses of NE. Data represent the means { SE for three independent experiments each done in triplicate.
WBS: Hepatology
290 ZHANG ET AL.
HEPATOLOGY February 1996
cytes.11 Adding NE to the hepatocyte cultures reversed the inhibitory effect of activin A on nuclear labeling. Figure 1 shows the effects of various concentrations of NE on the labeling index of hepatocyte cultures treated with EGF and activin A. The labeling index of EGFtreated cells was 50.3 { 3.6% (mean { SE, n Å 3). When 1 nmol/L activin A was added together with EGF, the labeling index was 13.8 { 2.14% (n Å 3). NE reversed the inhibitory effect of activin A: in the presence of EGF, activin A and 1005 mol/L NE, the labeling index was 41.2 { 3.8% (n Å 3). This effect of NE was dosedependent and significant at concentrations of 1006 mol/L and above. Similar results were obtained when DNA synthesis was assessed by measuring [3H] thymidine incorporation (Table 1). In contrast, when activin A at concentrations of 10 nmol/L and above was added, NE did not reverse the inhibitory effect of activin A (data not shown). The acridine orange/ethidium bromide staining test showed that these cells were not dead. Schwall et al.12 reported recently that activin A induced hepatocyte death. They cultured hepatocytes in the absence of growth factor and administered activin A. In agreement with their results, activin A induced cell death of cultured hepatocytes in the absence of growth factor (Fig. 2). The dose-response relationship for activin A–induced cell death correlated with that for activin A–induced inhibition of DNA synthesis.11 Under our experimental conditions, however, approximately 50% of the hepatocytes were still alive in the presence of high concentrations of activin A. When the DNA extracted from activin-treated cells was subjected to agarose gel electrophoresis, a ladder pattern was observed, indicating that DNA fragmentation had occurred (Fig. 3). In contrast, virtually no DNA fragmentation was observed in hepatocytes incubated with EGF and activin A. Electron microscopic analysis revealed that the nuclear morphology of the activin A– treated cells had changed strikingly compared with control cells. The nuclear margins showed irregular indentations and the chromatin was moderately condensed (data not shown). As shown in Fig. 4, 1005 mol/L
TABLE 1. Effect of NE on Activin-Mediated Inhibition of [3H]Thymidine Incorporation [3H]Thymidine Incorporation (cpm 1 1003)
Addition
None EGF EGF / Activin Activin Activin Activin
activin A A / NE (1007 A / NE (1006 A / NE (1005 A / NE (1004
mol/L) mol/L) mol/L) mol/L)
5.2 36.0 11.6 12.9 21.6 26.4 27.9
{ { { { { { {
01-24-96 21:43:57
NE prevented activin-induced cell death but the prevention was not complete. The DNA fragmentation induced by activin A was reduced considerably by NE but slight fragmentation observed in cells incubated with activin A and NE, indicating that DNA fragmentation had occurred in at least some hepatocytes (Fig. 5). Time-Dependent Action of Norepinephrine. NE, as a comitogenic growth factor, was shown to enhance the mitogenic effect of EGF and, furthermore, it had to be present for a substantial portion of the first 24 hours of incubation to enhance DNA synthesis.17,18 We, therefore, measured the time at which NE exerted its action. As shown in Fig. 6, when NE was present from the beginning of the culture period, it reversed the inhibitory effect of activin A on DNA synthesis. However, when it was added after culture for 24 hours, it did not enhance DNA synthesis significantly. Note that NE was fully effective when added from 0 to 24 hours. Effect of Inhibitors of Adrenergic Receptor Antagonist. It is known that catecholamines are capable of
0.9 0.4 0.9 0.2* 3.9* 0.8* 3.5*
NOTE. Hepatocytes were cultured with 1 nmol/L EGF, 1 nmol/L activin A and various concentration of norepinephrine (NE). [3H]Thymidine incorporation was measured as described in Materials and Methods. Values are the means { SE for four determinations and the representative of three experiments. * P õ .01 vs. EGF / activin A.
m4730$0002
FIG. 2. Effect of activin A on viability of hepatocytes. Cells were cultured with various concentrations of activin A for 24 hours and viability of cells was determined by staining the cells with acridine orange/ethidium bromide. Values are the means { SE for three experiments.
hepa
stimulating DNA synthesis by cultured hepatocytes via interaction with a1-adrenergic receptors.17-19 Recent studies showed that b-adrenergic receptors also mediated such an effect, at least under some conditions.20 In our experiment, we investigated which types of adrenergic receptor mediated the activity of NE. Cultured hepatocytes were treated with 1 nmol/L EGF, 1 nmol/L activin A, 10 mmol/L NE, and various adrenergic receptor antagonists and the labeling index was measured using Bromodeoxyuridine. As shown in Table 2, NEinduced enhancement of DNA synthesis was abolished
WBS: Hepatology
HEPATOLOGY Vol. 23, No. 2, 1996
ZHANG ET AL.
291
by the a1-adrenergic receptor blocker prazosin (1006 mol/L), whereas either the a2-adrenergic receptor blocker yohimbine (1006 mol/L) nor the b-receptor blocker propranolol (1005 mol/L) had any effect. Likewise, NE-induced inhibition of cell death was prevented by prazosin (Table 3). Effect of Norepinephrine on Follistatin Production. Follistatin, an activin-binding protein, is known
to inhibit the effects of activin A.21 As follistatin is expressed in rat liver,22 it is possible that NE could reverse the effect of activin A by increasing follistatin production. Therefore, we investigated whether NE stimulated follistatin production. A significant amount of follistatin was released into the conditioned medium of hepatocytes in culture and EGF augmented its release (Table 4). Next, hepatocytes were cultured with NE in the presence or absence of EGF, the conditioned media were collected after 72 hours and the amount of follistatin released was measured. As shown in Table 4, EGF stimulated follistatin production significantly, whereas NE alone did not affect its output, neither did it enhance EGF-induced follistatin production. DISCUSSION
NE, as a comitogenic growth factor, is thought to play an important role in the initiation of DNA synthesis and modulation of various cellular functions during liver regeneration.9,10,17,18 Our present results showed that, in the presence of NE, the inhibitory effect of activin A on
DNA synthesis was reversed. The effect of NE was dosedependent and significant at concentrations of 1006 mol/L and above. A specific a1-adrenergic antagonist, prazosin, blocked this effect completely, whereas the a2-adrenergic antagonist yohimbine and b-receptor antagonist propranolol had no effect. Therefore, it is clear that the NE effect was mediated via a1-adrenergic receptors. It should be emphasized that the effect of NE on cultures treated with EGF was exerted during the first 24 hours in the presence and absence of activin A, that was because of at least partly the fact that a1-adrenergic receptor expression decreased after culture for 24 hours.23,24 Previous studies showed that the low levels of messenger RNA for the bA subunit of activin was expressed in intact liver, disappeared at 12 hours, and then increased markedly 24 hours after hepatectomy.11 Furthermore, blockade of the action of endogenous activin A accelerated liver regeneration and resulted in the appearance of nuclear labeling several hours earlier than in control rats.25 Therefore, activin A expressed in the remnant liver after partial hepatectomy would appear to exert tonic inhibition on the initiation of DNA synthesis. Given that NE is released from sympathetic nerve terminals shortly after partial hepatectomy, reversal of this inhibitory action of activin A may be an important mode of action of NE. Recent studies in our laboratory showed that EGF stimulated the production of follistatin, an activinbinding protein, by cultured hepatocytes.22 As follistatin enhanced EGF-mediated DNA synthesis and bound to activin A in a stoichiometric manner to neutralize the effect of activin significantly,21 we investi-
FIG. 3. Fragmentation of DNA in activin-treated cells. Cells were incubated for 24 hours with 1 nmol/L activin A. DNA was extracted, separated by agarose gel and visualized by ethidium bromide. DNA fragmentation was also examined in HL-60 cells treated with A23187 as a positive control. Lane 1, EGF / activin A; lane 2, activin A; lane 3, HL-60 cells treated with 1 mmol/L A23187.
m4730$0002
FIG. 4. Effect of norepinephrine on activin-induced cell death. Cells were incubated for 24 hours with 1 nmol/L EGF or 10 mmol/L NE in the presence and absence of 1 nmol/L activin A. Viability of cells was analyzed by using acridine orange/ethidium bromide. Values are the means { SE for four experiments. *P õ .01 vs. EGF, **P õ .01 vs. activin A.
01-24-96 21:43:57
hepa
WBS: Hepatology
292 ZHANG ET AL.
HEPATOLOGY February 1996 TABLE 2. Effects of Adrenergic Antagonists on the Effect of Norepinephrine on DNA Synthesis Addition
Control EGF EGF / Activin A / NE / NE / Prazosin / NE / Yohimbine / NE / Propranolol
Labeled Nuclei (%)
8.9 52.5 12.9 42.0 10.7 41.5 42.8
{ { { { { { {
1.8 3.7 0.9 3.3 0.8 3.4 2.5
NOTE. Hepatocytes were incubated with 1 nmol/L EGF, 1 nmol/L activin A, and 1005 mol/L NE in the presence of prazosin (1006 mol/L), yohimbine (1006 mol/L) or propranolol (1005 mol/L). Labeling index was measured by using BrdU. Values are the means { SE for three different experiments each done in triplicate.
FIG. 5. Effect of norepinephrine on activin-induced DNA fragmentation. Cells were incubated for 24 hours with 1 nmol/L activin A in the presence and absence of 10 mmol/L norepinephrine. Then DNA was extracted, separated by agarose gel and visualized by ethidium bromide. Lane 1, activin A; lane 2, activin / norepinephrine.
gated whether NE increased follistatin production. The results refuted the hypothesis that NE acted by stimulating follistatin synthesis. Schwall et al. reported that activin A, in the absence of growth factors, induced death of cultured hepatocytes.12 More recently, investigators reported that activin A induced apoptosis of cultured hepatocytes, that was adjudged by the presence of DNA fragmentation and typical morphological alterations.13 Our present results confirmed their data, showing that activin A induced DNA fragmentation and drastic changes in nu-
clear morphology, including moderate chromatin condensation. Taken together, these results suggest that activin A attenuates cell-cycle progression in the presence of appropriate growth stimulation, but, in the absence of growth factors, it induces apoptosis of cultured hepatocytes. Consistent with this notion, activin A administered intravenously 11 hours after partial hepatectomy delayed the initiation of DNA synthesis without causing apoptosis in the remnant liver.25 By acting on a1-adrenergic receptors, NE attenuated activin A– induced apoptosis, and although NE alone did not stimulate hepatocyte proliferation, it attenuated apoptosis induced by activin A. It has been established that activation of protein kinase C, by adding phorbol ester or an agonist that causes phosphoinositide breakdown, protects thymocytes against apoptosis.26,27 As activation of a1-adrenergic receptors results in hydrolysis of phosphoinositides leading to protein kinase C activation,28 it is possible that NE protects cells against apoptosis by a protein kinase C–dependent mechanism. In a previous study, we showed that activin A augmented glycogenolysis by causing breakdown of phosphoinosides in isolated rat hepatocytes.29 Despite the fact that activin A activated the calcium messenger system in freshly isolated hepatocytes, it lost its ability to cause hydrolysis of phosphoinositides in hepatocytes cultured for longer than 3 hours (unpublished observation). Although the precise mechanism responsible for
TABLE 3. Effects of Adrenergic Antagonists on the Effect of Norepinephrine on Cell Viability Addition
Activin Activin Activin Activin
FIG. 6. Effect of norepinephrine added at different time points. Hepatocytes were incubated with 1 nmol/L EGF with or without and 1 nmol/L activin A. NE was added at either 0 or 24 hours. Data represent the means { SE of experiments each done in triplicate.
m4730$0002
01-24-96 21:43:57
hepa
A A / NE A / NE / Prazosin A / NE / Yohimbine
Cell Viability (%)
61.0 81.0 63.3 82.0
{ { { {
2.8 2.1 3.3 3.2
NOTE. Hepatocytes were incubated with 1 nmol/L activin A and 1005 mol/L NE in the presence and absence of prazosin (1006 mol/L) or yohimbine (1006 mol/L). Cell viability was determined as described in Materials and Methods. Values are the means { SE for four experiments.
WBS: Hepatology
HEPATOLOGY Vol. 23, No. 2, 1996
ZHANG ET AL.
TABLE 4. Follistatin Released to the Medium of Cultured Hepatocytes Additions
Follistatin (ng/mL)
None NE EGF EGF / NE
7.0 6.8 30.9 26.2
{ { { {
0.9 0.6 4.5 2.8
NOTE. Hepatocytes were cultured with or without 1 nmol/L EGF in the presence of 1005 mol/L NE. Conditioned medium was collected at 72 hours. Values are the means { SE for 12 determinations from two experiments.
this refractoriness is uncertain at present, uncoupling of the activin receptor system to phospholipase C does appear to occur in cultured hepatocytes. Therefore, it seems unlikely that activin A activates protein kinase C in cultured hepatocytes. In conclusion, NE, acting on a1-adrenergic receptors, modulated the inhibitory actions of activin A on DNA synthesis by cell viability of rat cultured hepatocytes. These data, together with those of Michalopoulos and colleagues,9,10,17,18,30 suggest that NE plays a very important role in the initiation of the early DNA synthesis response during liver regeneration. Acknowledgment: The authors are grateful to Romi Nobusawa and Kiyomi Ohgi for secretarial assistance during the preparation of the report. REFERENCES 1. Michalopoulos GK. Liver regeneration: molecular mechanisms of growth control. FASEB J 1990;4:176-187. 2. Rabes HM, Wirshing R, Tuczek HV, Iseler, G. Analysis of cell cycle compartments of hepatocytes after partial hepatectomy. Cell Tissue Res 1976;6:517-532. 3. Mead JE, Fausto N. Transforming growth factor a may be a physiological regulator of liver regeneration by means of an autocrine mechanism. Proc Natl Acad Sci U S A 1989;86:1555815562. 4. Nakamura T, Nishizawa T, Hagiya M, Seki T, Shimonishi M, Sugiyama A, Tashiro K, et al. Molecular cloning and expression of human hepatocyte growth factor. Nature 1989;342:440-442. 5. Kinoshita T, Hirao S, Matsumoto K, Nakamura T. Possible endocrine control by hepatocyte growth factor of Liver Regeneration after partial hepatectomy. Biochem Biophys Res Commun 1991;177:330-335. 6. Noji S, Tashiro K, Koyama E, Nohno T, Ohyama K, Taniguchi S, Nakamura T. Expression of hepatocyte growth factor gene in endothelial and Kupffer cells of damaged rat liver, as revealed by in situ hybridization. Biochem Biophys Res Commun 1990;173:42-47. 7. Hamanoue M, Kawaida K, Takao S, Shimazu H, Noji S, Matsumoto K, Nakamura T. Rapid and marked induction of hepatocyte growth factor during liver regeneration after ischemic or crush injury. HEPATOLOGY 1992;16:1485-1492. 8. Braun L, Mead JE, Panzica M, Mikumo R, Bell GI, Fausto N. Transforming growth factor b mRNA increases during liver regeneration: a possible paracrine mechanism of growth regulation. Proc Natl Acad Sci U S A 1988;85:1539-1543. 9. Curise JL, Knechtle SJ, Bollinger RR, Kuhn C, Michalopoulos GK. a1-adrenergic effects and liver regeneration. HEPATOLOGY 1971;7:1189-1194.
m4730$0002
01-24-96 21:43:57
hepa
293
10. Houck KA, Cruise JL, Michalopoulos G. Norepinephrine modulates the growth-inhibitory effect of transforming growth factorbeta in primary rat hepatocyte cultures. J Cell Physiol 1988;135: 551-555. 11. Yasuda H, Mine T, Shibata H, Eto Y, Hasegawa Y, Takeuchi T, Asano S, et al. Activin A: an autocrine inhibitor of initiation of DNA synthesis in rat hepatocytes. J Clin Invest 1993;92:14911496. 12. Schwall RH, Robbins K, Jaedieu P, Chang L, Lai G, Terrell TG. Activin induces cell death in hepatocytes in vivo and in vitro. HEPATOLOGY 1993;18:347-356. 13. Hully JR, Chang L, Schwall RH, Widmer HR, Terrell TG, Gillett NA. Induction of apoptosis in the murine liver with recombinant human activin A. HEPATOLOGY 1994;20:854-861. 14. Berry MN, Friend DS. High-yield preparation of isolated rat liver parenchymal cells. J Cell Biol 1969;43:506-520. 15. Saito S, Nakamura T, Titani K, Sugino H. Production of activinbinding protein by rat granulosa cells in vitro. Biochem Biophys Res Commun 1991;176:413-422. 16. Park DR, Bryan VM, Oi VT, Herzeuberg LA. Antigen specific identification and cloning of hybridomas with a fluorescenceactivated cell sorter. Proc Natl Acad Sci U S A 1979;76:19621966. 17. Cruise JL, Michalopoulos G. Norepinephrine and epidermal growth factor: dynamics of their interaction in stimulation of hepatocyte DNA synthesis. J Cell Physiol 1985;125:45-50. 18. Cruise JL, Houck KA, Michalopoulos GK. Induction of DNA synthesis in cultured rat hepatocytes through stimulation of a1adrenoreceptor by norepinephrine. Science 1985;227:749-751. 19. Takai S, Nakamura T, Komi N, Ichihara A, Corrigan A, Woo WK. Mechanisms of stimulation of DNA synthesis induced by epinephrine in primary cultures of adult rat hepatocytes. J Biochem 1988;103:848-852. 20. Refsnes M, Thoresen GH, Sandnes D, Dajani DF, Dajani L, Christoffersen T. Stimulatory and inhibitory effect of catecholamines on DNA synthesis in primary rat hepatocyte cultures: role of a1- and b-adrenergic mechanisms. J Cell Physiol 1992;151: 164-171. 21. Nakamura T, Takio K, Eto Y, Shibai H, Titani K, Sugino H. Activin-binding protein in rat ovary is follistatin. Science 1990;247:836-838. 22. Kanzaki M, Zhang YQ, Yasuda H, Mine T, Kojima I. Production of follistatin in cultured rat hepatocytes. Biochem Biophys Res Commun 1994;202:422-428. 23. Noda C, Nakamura T, Ichihara A. a-Adrenergic regulation of enzymes of amino acid metabolism in primary cultures of adult rat hepatocytes. J Biol Chem 1983;258:1520-1525. 24. Nakamura T, Tomomura A, Kato S, Noda C, Ichihara A. Reciprocal expression of a,- and b-adrenergic receptors, but constant expression of glucagon receptor by rat hepatocytes during development and primary culture. J Biochem 1984;96:127-136. 25. Kogure K, Omata W, Kanzaki M, Zhang YQ, Yasuda H, Mine T, Kojima I. A single intraportal administration of follistatin accelerates liver regeneration in partially hepatectomized rats. Gastroenterology 1995;108:1136-1142. 26. McConkey DJ, Hartzell P, Amador-Perez JF, Orrenius S, Jondal M. Calcium-dependent killing of immature thymocytes by stimulation via the CD3/T cell receptor complex. J Immunol 1989;143: 1801-1806. 27. McConkey DJ, Hartzell P, Jondal M, Orrenius S. Inhibition of DNA fragmentation in thymocytes and isolated nuclei by agents that stimulate protein kinase C. J Biol Chem 1989;264:1339913402. 28. Williamson JR, Cooper RH, Joseph SK, Thomas AP. Inositol triphosphate and diacylglycerol as intracellular second messengers in liver. Am J Physiol 1985;248:C203-C216. 29. Mine T, Kojima I, Ogata E. Stimulation of glucose production by activin A in isolated rat hepatocytes. Endocrinology 1989;125: 586-591. 30. Houck KA, Michalopoulos GK. Altered response of regenerating hepatocytes to norepinephrine and transforming growth factor type b. J Cell Physiol 1989;141:503-509.
WBS: Hepatology