- Email: [email protected]
ERK supports both proliferation and survival
PI3K-FRAP/mTOR Pathway Is Critical for Hepatocyte Proliferation Whereas MEK/ERK Supports Both Proliferation and Survival Alexandre Coutant, Claude Rescan, David Gilot, Pascal Loyer, Christiane Guguen-Guillouzo, and Georges Baffet Growth factors are known to favor both proliferation and survival of hepatocytes. In this work, we investigated the role of 2 main signaling pathways, phosphoinositide 3-kinase (PI3K) and mitogen-activated protein kinase (MEK)/extracellular signal–regulated kinase (ERK), in these processes. First, evidence was provided that the PI3K cascade as well as the MEK/ERK cascade is a key transduction pathway controlling hepatocyte proliferation, as ascertained by arrest of DNA synthesis in the presence of LY294002, a specific PI3K inhibitor. Inhibition of FRAP/mTOR by rapamycin also abrogated DNA replication and protein synthesis induced by growth factor. We showed that expression of cyclin D1 at messenger RNA (mRNA) and protein levels was regulated by this pathway. We highlighted that 4E-BP1 phosphorylation was not activated by epidermal growth factor (EGF) but was under an insulin-regulation mechanism through a PI3K-FRAP/mTOR activation that could account for the permissive role of insulin on hepatocyte proliferation. No interference between the MEK/ERK pathway and 4E-BP1 phosphorylation was detected, whereas p70S6K phosphorylation induced by EGF was under a U0126-sensitive regulation. Last, we established that the antiapoptotic function of EGF was dependent on MEK, whereas LY294002 and rapamycin had no direct effect on cell survival. Taken together, these data highlight the regulation and the role of 2 pathways that mediate growth-related response by acting onto distinct steps. In conclusion, hepatocyte progression in late G1 phase induced by EGF generates survival signals depending on MEK activation, whereas PI3K and MEK/ERK cascades are both necessary for hepatocyte replication. (HEPATOLOGY 2002;36:1079-1088.)
C
ell cycle progression through G1 phase requires the integration of signals from the extracellular environment such as growth factors, cytokines, and extracellular matrix proteins mediated by different transduction cascades. The ability to induce cellular proliferation is often correlated with the ability to promote Abbreviations: MEK, mitogen-activated protein kinase; ERK, extracellular signal–regulated kinase; PI3K, phosphoinositide 3-kinase; mRNA, messenger RNA; EGF, epidermal growth factor. From INSERM U522, Unite´ de Recherches He´patologiques, IFR 97, Hoˆpital Pontchaillou, Rennes, France. Received February 15, 2002; accepted July 24, 2002. Supported by the Institut National de la Sante´ et de la Recherche Me´dicale and the Association pour la Recherche contre le Cancer. C.R. and D.G. are recipients of a fellowship from the Ligue d’Ille et Vilaine contre le Cancer. A.C. is a recipient of a fellowship from the Ministe`re de l’Education nationale, de la Recherche et de la Technologie. A.C. and C.R. contributed equally to this work. Address reprint requests to: Georges Baffet, INSERM U522, Unite´ de Recherches He´patologiques, IFR 97, Hoˆpital Pontchaillou, 35033 Rennes, France. E-mail: [email protected]; fax: (33) 2-99-54-01-37. Copyright © 2002 by the American Association for the Study of Liver Diseases. 0270-9139/02/3605-0010$35.00/0 doi:10.1053/jhep.2002.36160
cell survival. Two signaling cascades have emerged as major players in the mitogenic and antiapoptotic response in many cells: the mitogen-activated protein kinase (MEK)/ extracellular signal–regulated kinase (ERK) and the phosphoinositide 3-kinase (PI3K) pathways. Growth factors as survival factors bind to cell surface receptors and trigger the activation of several kinases, including PI3K, which seems to be implicated in the control of growth and apoptosis of a wide range of cell types.1 This kinase activates PDK1 that in turn leads to the activation of a serine/threonine kinase termed AKT or PKB, which plays a central role in promoting the survival of many cell types. PDK1 also phosphorylates and activates FRAP/mTOR (target of rapamycin). p70s6k, which regulates the ribosomal S6 subunit phosphorylation in response to mitogens, is controlled by FRAP/mTOR and plays an essential role in the translation machinery.2-4 FRAP/mTOR also controls the phosphorylation of 4EBP1, a key protein involved in the regulation of ribosome binding to the messenger RNA (mRNA) via eIF4E.5 Interferences between MEK/ERK and PI3K pathways have 1079
1080
COUTANT ET AL.
often been described; the sensibility of ERK activation to inhibition by PI3K inhibitors depends in many cases on cell types and on the differentiation state of the cells.6,7 Activation of ERK has also proved to be a major regulator of cell proliferation and survival in response to growth factor.8 Once Ras has been activated, GTP-bound ras interacts with the serine/threonine kinase raf and facilitates its activation. It also works the same way with other target enzymes, including PI3K.9 Activated raf phosphorylates and activates the downstream MEK1/2, which in turn phosphorylates ERK1/2. Growth factors play an important role in normal liver growth, in liver after partial hepatectomy, and in proliferative hepatocytes in vitro.10,11 For instance, epidermal growth factor (EGF) inhibits programmed cell death and induces proliferation in hepatocytes.12 To understand the potential importance of components of the cellular homeostasis mechanism, knowledge of G1 phase regulation is of prime importance. Hepatocytes in primary culture seem to be a powerful model to address this question. Following tissue disruption of cell-cell contact during cell isolation, the G0/G1 transition occurs.13 This mimics the entry into G1 of proliferating hepatocytes in vivo in the regenerating liver after partial hepatectomy.10,12 Then, in primary culture, hepatocytes spontaneously progress in G1 up to a restriction point located at the 2/3 of G1 phase.14 We recently showed that, in hepatocytes, the first third of G1 phase is exclusively devoted to growth factor– dependent morphogenic events, whereas its mitogenic signal occurs only around mid-late G1 phase at the restriction point.15,16 A relation between MEK/ERK activation in hepatocytes in vivo and the mitogen-dependent capacity of the cells in vitro has been established. In late G1, MEK/ERK activation is associated with accumulation of cyclin D1 and mitogen-dependent progression of hepatocytes to S phase. Concerning PI3K, few studies have been performed on hepatocytes, and those studies had conflicting results.17-20 In this report, we analyzed the possible role of the PI3K pathway in the regulation of EGF-induced DNA replication and survival in mid-late G1 phase. We examined the relation that could exist between this pathway and the expression of cyclin D1, a critical player in G1/S transition, and questioned the interference that could exist between the PI3K and MEK/ERK pathways.
Materials and Methods Cell Cultures. Hepatocytes were isolated from Sprague-Dawley male (150-200 g) rat livers by the 2-step perfusion procedure using 0.025% Liberase (BoehringerIngelheim, Indianapolis, IN) buffered with 0.1 mol/L
HEPATOLOGY, November 2002
HEPES (pH 7.4) as previously described.21 At indicated times, wortmannin, LY294002, rapamycin, cycloheximide, or U0126 dissolved in dimethyl sulfoxide was added at defined concentrations. All control features in dimethyl sulfoxide at a final concentration of 0.2% or 0.37% were changed at the same time as that of treated cells. Transfection Experiments. Hepatocytes were transfected using GB12 as previously described.22,23 Four micrograms of plasmid empty vector or pcDNA-HA-Akt (K179M) was mixed with cationic lipid (10 g/mL) in 800 L of OptiMEM at room temperature for 30 minutes. This medium was added drop-wise onto 24-hourold hepatocytes. Twelve hours later, hepatocytes were stimulated by EGF. [3H]Thymidine Incorporation and [14C]Leucine Incorporation. The rates of DNA and protein synthesis were measured in primary cultures by adding 2 Ci of [methyl-3H]thymidine or 1 Ci/mL of L-[U-14C]leucine for given periods of time before cell harvesting and precipitation with ice-cold trichloroacetic acid (5%). Chemicals. [␣-32P]deoxycytidine triphosphate (3,000 Ci/mmol), L-[U-14C]leucine (31.5 mCi/mmol), and [methyl-3H]thymidine (5 Ci/mmol) were from Amersham Corp. (Buckinghamshire, England); recombinant human EGF and MEK inhibitor U0126 were from Promega (Madison, WI); wortmannin was from Calbiochem (San Diego, CA); and insulin I-5500, LY294002, rapamycin, and cycloheximide were from Sigma (St. Louis, MO). Immunoblotting. After sodium dodecyl sulfate/polyacrylamide gel electrophoresis, proteins were transferred and detected according to the SuperSignal Ultra Chemiluminescent Substrate procedure (Pierce, Rockford, IL). Anti–phospho-GSK 3 ␣/, anti–phospho-ERK, and anti–phospho-AKT antibodies as well as a mixing of anti-phospho Thr 389 and anti-phospho Thr 421/424 antibodies for Phospho-p70S6Kinase detection were from New England Biolabs (Beverly, MA). Anti-AKT (sc-1618) and anti-ERK1/2, a mixing of ERK1 (sc-94) and ERK2 antibodies (sc-154), were from Santa Cruz Biotechnology (Santa Cruz, CA). Polyclonal antibody against cyclin D1 was obtained from Neomarkers (Union City, CA). Rabbit polyclonal anti– 4E-BP124 was a gift from Drs. Pyronnet and Sonenberg. Northern Blotting. Cells were lysed at different times of culture, RNA was extracted with the Qiagen RNeasy kit (Qiagen, Valencia, CA), and Northern blotting was performed as described previously.15 Caspase Activity Assay. Caspase-mediated cleavage of DEVD-AMC was measured by spectrofluorometry (Molecular Devices, Wokingham, England) at 380- and 440-nm wavelength (excitation and emission, respec-
HEPATOLOGY, Vol. 36, No. 5, 2002
COUTANT ET AL.
1081
tively).25 For each measure, the background of fluorescence was subtracted. The caspase activity was presented in arbitrary units of fluorescence (per 80 g of total proteins). Assessment of Mitochondrial ⌬m. 3,3⬘-Dihexyloxacarbocyanine iodide (DiOC6) was purchased from Molecular Probes, Inc. (Eugene, OR). Changes in the inner mitochondrial transmembrane potential ⌬m were determined by incubating 6 ⫻ 105 hepatocytes in 40 nmol/L DiOC626 in culture medium for 30 minutes at 37°C. Cells were analyzed using a FACScan flow cytometer (Becton Dickinson, Mountain View, CA). All the experiments described in this report were performed at least 3 times.
Results PI3K Inhibition Blocks DNA Synthesis in Hepatocytes. PI3K and MEK/ERK pathways have been involved in the regulation of cell cycle progression in many cell types. We already showed that the MEK/ERK cascade is a key signaling pathway involved in the regulation of G1 phase progression in proliferating hepatocytes.15,16 To improve our understanding of the role of PI3K cascade activation onto hepatocyte cell cycle progression, PI3K inhibition experiments were performed. First, hepatocyte primary cultures were maintained for 48 hours in basal conditions (48-hour-old hepatocytes) and then stimulated with EGF in the presence or absence of a specific PI3K inhibitor, LY294002 (Fig. 1A). This inhibitor at a concentration of 10 mol/L totally inhibited DNA replication all along the time course analyzed, showing that PI3K inhibition not only delayed but totally abolished DNA replication in hepatocytes. In parallel, we confirmed that the MEK cascade is an important player in DNA replication by showing that the MEK inhibitor U0126 as well as PD9805915,16 completely blocked DNA synthesis. We performed dose-response experiments in the presence of PI3K inhibitor (Fig. 1B). A dose response was obtained; 5 mol/L LY294002 inhibited DNA replication by 50%, whereas 10 mol/L decreased thymidine incorporation by 100%. To provide evidence that cells remained blocked at the restriction point or progressed in late G1 phase in the presence of U0126 or LY294002, reversion experiments were performed. Hepatocytes (48 hours old) were stimulated by EGF and treated for 24 hours with the 2 inhibitors. U0126 and LY294002 were then removed and replication analyzed in the presence of growth factor alone (Fig. 2A). As already detailed,15 hepatocytes from the reversion experiments after 24-hour MEK/ERK inhibition started to replicate DNA with an 18- to 24-hour delay,
Fig. 1. Regulation of DNA replication by the PI3K pathway. (A) Time course of [3H-methyl]thymidine incorporation into DNA; 48-hour-old hepatocytes were maintained in basal medium (basal) or stimulated with EGF (EGF) in the absence or presence of LY294002 (15 mol/L) or U0126 (50 mol/L) and analyzed at the indicated time after stimulation. (B) Dose-dependent inhibition of DNA replication by LY294002; 48-hour-old hepatocytes were stimulated with EGF in the presence of increasing concentrations of LY294002. DNA replication was measured at the indicated times after stimulation.
corresponding to the time needed to reach the G1/S transition, confirming that the cells were blocked at the restriction point in the presence of MEK inhibitor. Concerning the PI3K cascade, our data showed that, unlike the MEK/ERK inhibition, the cells were able to undergo DNA synthesis with only a 6- to 12-hour delay on removal of the LY inhibitor. In hepatocytes, as in many cells, the up-regulation of cyclin D1 is an important player in late G1 progression and drives the cells to mitogenic response.27 To confirm the effect of LY294002 and U0126 inhibitions on late cell cycle progression, we analyzed the induction of cyclin D1 on inhibitor removal after 24-hour inhibition. As shown in Fig. 2B, cyclin D1 protein could be detected 3 hours after removal of LY294002, whereas its expression was
1082
COUTANT ET AL.
HEPATOLOGY, November 2002
Fig. 2. DNA replication, cyclin D1 expression, and GSK3 phosphorylation after LY294002 or U0126 treatment and drug removal. (A, B, and D) Reversion experiments; 48-hourold hepatocytes were cultured in basal conditions or stimulated with EGF in the presence of LY294002 (15 mol/L) or U0126 (50 mol/ L). Then, 24 hours after treatment, inhibitors were removed and hepatocytes cultured in the presence of EGF alone. (A) Methyl thymidine incorporation into DNA, (B) cyclin D1 protein expression, and (D) GSK3 phosphorylation were analyzed at the indicated times after drug removal. Error bars indicate SD; *P ⬍ .001 (vs. reversion U0126) by Student’s t test. (C) Control of phospho-GSK3 inhibition; 48-hour-old hepatocytes were stimulated with EGF in the presence of LY294002 or U0126 and analyzed at the indicated times after stimulation.
only detected after 12 hours in the experiment with U0126. All these results strongly argue for a control of late G1 progression by a mechanism not involving PI3K but MEK activation. As control, we confirmed that the 2 pathways were activated by the same kinetic after drug removal. We chose to look at the phospho-GSK3, which was under both PI3K and MEK activating processes. We first showed that the 2 drugs partially inhibited the GSK-3 phosphorylation induced by EGF (Fig. 2C). Second, in reversion experiments, phosphorylation of GSK-3 was induced simultaneously after removal of LY294002 and U0126 (Fig. 2D), indicating that recovery of intracellular pathways was not delayed in U0126- compared with LY294002-treated hepatocytes. FRAP/mTOR Inhibition Abolishes DNA and Protein Synthesis. Two downstream signaling pathways have been described for PI3K (AKT and FRAP/mTOR) that can be distinguished by their sensibility to the drug rapamycin.28 As shown in Fig. 3A, rapamycin, which acts at the FRAP/mTOR level, was able to decrease [3H]thymidine incorporation in a dose-response– dependent manner at any of the times analyzed. A total of 50% and 80% of DNA synthesis was inhibited in the presence of 0.1 and 0.5 nmol/L of rapamycin, respectively, whereas 1 nmol/L totally abolished DNA replication. As LY294002, rapamycin did not only delay but totally inhibited DNA replication over a longer period of time, up to 72 hours (result not shown). Next, we investigated whether the effect of rapamycin on thymidine incorporation could correlate with inhibi-
tion of protein synthesis (Fig. 3B and C). As in vascular endothelial cells,29 rapamycin inhibited thymidine incorporation with the same dose dependence as protein synthesis. In both cases, the median inhibitory concentration for rapamycin was 0.05 nmol/L. Even at high concentrations, rapamycin was not able to decrease thymidine and leucine incorporations below the basal level of unstimulated control hepatocytes (Fig. 3B and data not shown). The correlation between the rate of DNA and protein synthesis was confirmed with the use of cycloheximide, a potent inhibitor of protein synthesis (Fig. 3C). The median inhibitory concentration for cycloheximide was identical for the inhibition of DNA replication and protein synthesis. Cyclin D1 Is Under a FRAP/mTOR Regulation Mechanism. Cyclin D1 has been shown to be regulated at 2 different levels in many cell types, including hepatocytes30; FRAP/mTOR, which plays a role in the regulation of the translation machinery, could be involved in this process. We therefore investigated the effects of LY294002 and rapamycin at the cyclin D1 mRNA and protein level. As expected, cyclin D1 mRNA and protein rapidly accumulated after EGF stimulation (Fig. 4A and B). In contrast, in cells exposed to medium with EGF plus LY294002 or rapamycin, cyclin D1 mRNA expression was diminished by approximately 90% and 80%, respectively. At the protein level, the 2 concentrations of rapamycin used (1 and 0.1 nmol/L) and LY294002 at 15 mol/L also inhibited cyclin D1 protein expression. Inhibition was less efficient by using LY294002 at 1
HEPATOLOGY, Vol. 36, No. 5, 2002
COUTANT ET AL.
1083
dicated that cyclin D1 protein was inhibited by 82% ⫾ 15%, 78% ⫾ 16%, and 75% ⫾ 20% in the presence of LY294002, rapamycin, and U0126, respectively. AKT Phosphorylation Is PI3K Dependent but Not Directly Related to the Replication Process. AKT that lies downstream of PI3K is a general mediator of cell survival and could be involved in proliferation of many cell types. Interferences between PI3K and MEK/ERK pathways have been described in many cell types. In this context, we first investigated the effect of PI3K and FRAP/mTOR inhibitors on AKT and ERK1/2 activations (Fig. 5). AKT was rapidly and transiently phosphorylated after EGF stimulation of 48-hour-old hepatocytes, reaching a maximum at 0.5 hours and decreasing thereafter at 1 hour to near zero 4 hours after stimulation (Fig. 5A). Second, in the presence of LY294002, this phosphorylation was totally abolished, confirming that AKT activation is downstream of PI3K in EGF-stimulated hepatocytes. Third, the FRAP/mTOR inhibitor, at a concentration that totally inhibited DNA replication, was unable to decrease AKT phosphorylation induced by the growth fac-
Fig. 3. Inhibition of DNA synthesis by rapamycin correlates with the inhibition of protein synthesis. (A) Hepatocytes (48 hours old) were stimulated with EGF in the presence or absence of increasing concentrations of rapamycin. DNA replication was measured at the indicated times after stimulation. (B and C) Hepatocytes (48 hours old) were stimulated with EGF in the presence of increasing concentrations of (B) rapamycin or (C) cycloheximide. Thymidine (■) and leucine (‚) incorporation was measured as described in Materials and Methods and is expressed as a percentage of the maximal incorporation obtained in the presence of EGF alone.
mol/L, in accordance with low inhibition of DNA replication (Fig. 1B). As control, the visualization of nonphosphorylated ERK1/2 protein showed similar protein concentrations loaded on polyacrylamide gel electrophoresis. Quantification from 4 separate experiments in-
Fig. 4. LY294002 and rapamycin block cyclin D1 mRNA and protein expression induced by EGF. (A) Cyclin D1 mRNA and (B) cyclin D1 protein expressions; 48-hour-old hepatocytes were stimulated or not with EGF in the presence or absence of LY294002 (15 mol/L) or rapamycin (1 nmol/L) and analyzed at the indicated times after stimulation.
1084
COUTANT ET AL.
Fig. 5. AKT protein phosphorylation in EGF-stimulated hepatocytes. (A) AKT and ERK1/2 phosphorylations were analyzed by Western blot; 48-hour-old hepatocytes were stimulated with EGF in the presence or absence of LY294002 (15 mol/L), rapamycin (1 nmol/L), or U0126 (50 mol/L) and analyzed at the indicated time after stimulation. (B and C) Hepatocytes (24 hours old) were transfected by dominant-negative AKT (DN) or empty vector (EV) and stimulated or not with EGF at 36 hours and analyzed at the indicated time after EGF stimulation for (B) DNA replication and (C) GSK3 phosphorylation and total AKT protein expression as control. Experiments were performed 4 times.
tor. We confirm that the phosphorylation of ERK1/2 was rapid and sustained on EGF stimulation. Interestingly, the 2 inhibitors of the PI3K pathway (LY294002 and rapamycin) had no effect on these phosphorylations.
HEPATOLOGY, November 2002
We then specified the effect of AKT inhibition on DNA replication. AKT dominant-negative vector was transfected 24 hours after seeding, and hepatocytes were stimulated by EGF 12 hours later (Fig. 5B). From the kinetics of 4 independent experiments, dominant-negative AKT, expressed at high levels in the cells (Fig. 5C), was not able to decrease DNA replication induced by EGF compared with control cells (EGF ⫹ empty vector). By this procedure, 35% to 40% of the hepatocytes appeared strongly positive for the GFP signal, confirming a high transfection efficiency.23 Furthermore, we showed that phosphorylation of GSK-3 that lies downstream of AKT was partially inhibited by the dominant-negative AKT compared with the empty vector. EGF-Independent 4E-BP1 Phosphorylation Is Under an Insulin-Activation Mechanism. The translational proteins p70S6K and 4E-BP1 are principally regulated by the PI3K-FRAP/mTOR pathway, and this regulation plays a key role in the translation machinery activation.5 Recently, studies indicated that the MEK/ ERK pathway could also play a major role in p70S6K activation. Therefore, impacts of LY294002, rapamycin, and U0126 on p70S6K and 4E-BP1 phosphorylations are important keys to understand the mechanism involved in hepatocyte proliferation. This study evidenced first that p70S6K is up-phosphorylated after EGF stimulation. We showed that phosphop70 basal level is under a PI3K/mTOR mechanism, whereas U0126 blocked the p70S6K phosphorylation induced by the growth factor (Fig. 6A). This result argues that p70S6K phosphorylation induced by EGF is MEK dependent, indicating that interferences between the MEK/ERK pathway and p70S6K could occur in hepatocytes. As control, we confirmed that LY294002 and rapamycin had no effect on ERK1/2 phosphorylation, whereas these phosphorylations were completely abrogated by U0126. Second, we looked at the role of PI3K pathway activation/inhibition on 4E-BP1 phosphorylation (Fig. 6B, C, and D). Surprisingly, in basal conditions, in the absence of EGF, 4E-BP1 appeared as a protein of low electrophoretic mobilities (/␥) corresponding to phosphorylated forms.24,31 Moreover, EGF stimulation did not influence 4E-BP1 mobility status. However, the 2 inhibitors, LY294002 and rapamycin, dephosphorylated the protein as visualized by a faster electrophoretic mobility (␣), showing that 4E-BP1 phosphorylation is under a PI3K-FRAP/mTOR mechanism that is independent of the presence of EGF (Fig. 6B and C). The MEK/ERK pathway did not interfere with this mechanism because the MEK inhibitor U0126 had no effect on 4E-BP1 electrophoretic mobility.
HEPATOLOGY, Vol. 36, No. 5, 2002
COUTANT ET AL.
1085
tion evidenced by the decrease of the mitochondrial transmembrane potential (⌬m), measured by flow cytometry, using the potential-sensitive dye DiOC6. In basal experiments, DEVD-AMC cleavage increased between 72 and 96 hours of culture and the growth factor protected hepatocytes from apoptosis during this period of stimulation (Fig. 7A). This protection by EGF was PI3K and FRAP/mTOR independent because inhibition by LY294002 as well as rapamycin did not induce DEVD-AMC cleavage. On the opposite, MEK/ERK inhibition by U0126 induced apoptosis as in basal conditions. Last, we confirmed that the MEK/ERK pathway is a key cascade in hepatocyte survival by showing that the mitochondrial potential is altered in U0126-treated cells (Fig. 7B). After 24 hours of exposure, the number of cells exhibiting a high ⌬m decreased, whereas that of cells with low ⌬m, corresponding to dead cells, increased by 21% ⫾ 3% in U0126-treated hepatocytes, showing that mitochondrial permeability transition occurred in this condition. In contrast, in LY294002- and rapamycintreated hepatocytes, the number of high-⌬m cells remained elevated, as in the presence of EGF alone.
Discussion
Fig. 6. Regulation of p70S6K and 4E-BP1 phosphorylation state. (A-C) Hepatocytes (48 hours old) in basal medium containing insulin were stimulated or not by EGF in the presence or absence of LY294002 (15 mol/L), rapamycin (1 nmol/L), or U0126 (50 mol/L) and analyzed at the indicated time after stimulation. (D) Freshly isolated hepatocytes were cultured in the absence or presence of insulin or EGF and analyzed 1 hour after stimulation.
Because insulin, used as a comitogen in the basal culture medium for hepatocytes, could contribute to the 4EBP1 activation, experiments were performed in the absence of the hormone (Fig. 6D). Under this condition, the ␣ form disappeared and only higher apparent molecular forms were detected in the presence of insulin (, ␥). In all cases, contrary to p70S6K, EGF did not influence the phosphorylation rate of the protein. Survival Is PI3K Independent but MEK Dependent in Mid-Late G1 Phase. Cell death was investigated by 2 distinct approaches: (1) the executioner caspases 3/7 activity measured through the proteolytic DEVD-AMC degradation, and (2) mitochondrial permeability transi-
In hepatocytes, growth factors have been shown to activate different signaling cascades, including MEK/ERK, PI3K, c-Jun-N-terminal kinase, and P38 pathways.32-35 This complex network participates in liver homeostasis by the regulation of hepatocyte proliferation and survival. Our study underlines the role of the 2 MEK/ERK and PI3K pathways in the regulation of hepatocyte proliferation and apoptosis. Hepatocytes in primary culture progress in G1 independently of growth factor stimulation up to the restriction point located in mid-late G1 phase, where they remain arrested in the absence of mitogenic signal. After mitogenic stimulation, hepatocytes progress in late G1 phase and undergo DNA synthesis. We and others previously showed that after growth factor stimulation for progression up to S phase, there was marked induction of cyclin D1 in late G1 phase.14,15,27,36 From previous data, the question arose as to whether PI3K cascade regulates DNA synthesis and expression of cyclin D1 or not. Although dominant-negative PI3K did not inhibit DNA synthesis in hepatocytes, other data suggested that the cascade inhibition could abolish or only delay entry to S phase.17,19,37 Our study clearly shows that the PI3K inhibitor, LY294002, and the FRAP/mTOR inhibitor, rapamycin, do not delay entry to S phase and expression of cyclin D1 but totally abolish DNA replication and the cyclin induction in growth factor–stimulated hepatocytes. Uncoupling of mRNA and protein levels has been
1086
COUTANT ET AL.
HEPATOLOGY, November 2002
Fig. 7. EGF protects hepatocytes from apoptosis via the MEK/ERK pathway. (A) Hepatocytes (48 hours old) were maintained in basal conditions (basal) or stimulated by EGF in the absence or presence of U0126 (50 mol/L), LY294002 (15 mol/L), or rapamycin (1 nmol/L) and analyzed for (A) DEVDAMC caspase activities at the indicated times and (B) mitochondrial transmembrane potential (⌬m) using the DiOC6 potential-sensitive dye 24 hours after stimulation. Experiments were performed on hepatocytes attached to the plate.
pointed out in hepatocytes, suggesting that cyclin D1 is subject to control on a translation level and/or protein stability in rat liver.30 In this context, we showed that LY294002 and rapamycin completely prevented cyclin D1 at the mRNA and protein levels. These results show that PI3K and FRAP/mTOR as well as the MEK/ERK cascade15 are linked to expression of cyclin D1, highlighting the pivotal role of this protein. We previously showed that hepatocytes treated with the MEK inhibitor, PD98059, behaved as if they remained blocked at the restriction point.15 We confirmed these results with U0126; in reversion experiments, hepatocytes were able to undergo DNA synthesis with a corresponding 24-hour delay like cells blocked at the restriction point. Whereas LY294002 completely blocked EGF-stimulated DNA synthesis, evidence was provided that hepatocytes progressed in late G1 phase in the presence of this inhibitor. Indeed, in reversion experiments, the time interval between removal of LY294002 and DNA synthesis was only 6 to 12 hours, suggesting that cells were not blocked at the restriction point. Moreover, cyclin D1 protein could be rapidly detected 3 hours after removal of LY294002, contrasting with cells treated with U0126. All these data indicate that hepatocytes could progress in late G1 phase independently of PI3K activation, whereas its activation controls the replication process. Inhibition of FRAP/mTOR was sufficient to inhibit DNA synthesis and expression of cyclin D1 without af-
fecting AKT phosphorylation. Moreover, AKT dominant-negative mutant did not interfere with DNA synthesis induced by EGF. Other data also suggest that inhibition of hepatocyte proliferation by transforming growth factor  did not involve AKT but might be partly mediated by the inhibition of ERK2 and p70S6K.18 Whereas previous studies have shown that Raf/MEK signaling is neither necessary nor sufficient for p70S6K activation by mitogen, recent data show that the expression of either c-Raf or MEK is sufficient for S6K1/2 activation in an ERK-dependent manner.38-40 Our results argue that p70S6K phosphorylation induced by EGF is under a U0126-sensitive regulation. On the opposite hand, in this report, we highlight that PI3K cascade activation at the 4E-BP1 level is independent of EGF stimulation but controlled by an insulin-regulation mechanism. From the literature, 4E-BP1 activation seems to be complex and the effect of insulin could involve multiple phosphorylation events, which have been described as rapamycin sensitive and insensitive.20,31,41 We show that a rapamycin-sensitive pathway is involved in 4E-BP1 phosphorylation in response to insulin. This regulation could account for the permissive role of insulin on hepatocyte proliferation. This finding, which highlights the key role of FRAP/ mTOR in hepatocyte replication, was also established for vascular endothelial cells and lymphocytes.29 However, Jiang et al.20 have recently reported that the effect of rapamycin on 4E-BP1 function in vivo can be significantly different from its effect in cultured cells. This discrepancy
HEPATOLOGY, Vol. 36, No. 5, 2002
could be in part due to the environmental conditions and via the release of humoral factors from other tissues in vivo, as reported by these investigators. Moreover, 4EBP1 underwent partial dephosphorylation after 4 days of treatment with rapamycin in vivo. We established that the antiapoptotic function of EGF in primary cultures of rat hepatocytes is dependent on the MEK/ERK pathway in mid-late G1 phase, whereas the inhibition of the PI3K cascade has no effect on hepatocyte survival. Moreover, our data highlight that hepatocyte progression in late G1 generates survival signals that counteract cell death induced in hepatocytes blocked at the restriction point in basal conditions or in the presence of MEK inhibitor. Cells blocked at the restriction point did not survive and rapidly progressed in apoptosis. Interestingly, the location of cells in G1, the state of differentiation, and the presence of specific apoptotic inducers in the medium could be of prime importance to transduce specific signals. In 20-day-old fetal hepatocytes, PI3K mediated the survival effect of EGF on transforming growth factor –induced death, whereas the MEK inhibitors did not block the survival effect of growth factor.42 In adult hepatocytes, EGF inhibition of transforming growth factor –induced apoptosis is dependent on the PI3K and ERK activities, whereas suppression of apoptosis induced by tumor necrosis factor ␣ does not require PI3K but is dependent on ERK and P38 mitogen-activated protein kinase activation.43 The survival and/or mitogenic orderly events induced by the growth factor seem to be dependent on a cellular clock in adult hepatocytes. In conclusion, the activation of different pathways might contribute to the complex regulation of hepatocyte DNA replication and survival. The data presented here emphasize the importance of PI3K-FRAP/mTOR in DNA replication, whereas MEK activation is fundamental for replication and survival. The 2 cascade activations are not redundant and are necessary for a full response to growth factor. Our results help to understand the intracellular signaling activation in G1 phase, which is of prime importance for the knowledge of hepatocyte growth and survival controls that regulate homeostasis during the regeneration process following liver injury and diseases. Acknowledgment: The authors thank Drs. S. Pyronnet and N. Sonenberg for giving us the anti– 4E-BP1 antibody and J. C. Andrieux for critical reading of the manuscript.
References 1. Datta SR, Brunet A, Greenberg ME. Cellular survival: a play in three Akts. Genes Dev 1999;13:2905-2927.
COUTANT ET AL.
1087
2. Chou MM, Blenis J. The 70 kDa S6 kinase: regulation of a kinase with multiple roles in mitogenic signalling. Curr Opin Cell Biol 1995;7:806814. 3. Chung J, Grammer TC, Lemon KP, Kazlauskas A, Blenis J. PDGF- and insulin-dependent pp70S6k activation mediated by phosphatidylinositol3-OH kinase. Nature 1994;370:71-75. 4. Conus NM, Hemmings BA, Pearson RB. Differential regulation by calcium reveals distinct signaling requirements for the activation of Akt and p70S6k. J Biol Chem 1998;273:4776-4782. 5. Gingras AC, Raught B, Sonenberg N. Regulation of translation initiation by FRAP/mTOR. Genes Dev 2001;15:807-826. 6. Berra E, Diaz-Meco MT, Moscat J. The activation of p38 and apoptosis by the inhibition of Erk is antagonized by the phosphoinositide 3-kinase/Akt pathway [published erratum appears in J Biol Chem 1998 Jun 26;273(26): 16630]. J Biol Chem 1998;273:10792-10797. 7. Rommel C, Clarke BA, Zimmermann S, Nunez L, Rossman R, Reid K, Moelling K, et al. Differentiation stage-specific inhibition of the RafMEK-ERK pathway by Akt. Science 1999;286:1738-1741. 8. Brunet A, Pouyssegur J. Mammalian MAP kinase modules: how to transduce specific signals. Essays Biochem 1997;32:1-16. 9. Kauffmann-Zeh A, Rodriguez-Viciana P, Ulrich E, Gilbert C, Coffer P, Downward J, Evan G. Suppression of c-Myc-induced apoptosis by Ras signalling through PI(3)K and PKB. Nature 1997;385:544-548. 10. Fausto N, Laird AD, Webber EM. Liver regeneration. 2. Role of growth factors and cytokines in hepatic regeneration. FASEB J 1995;9:15271536. 11. Fausto N. Liver regeneration. J Hepatol 2000;32:19-31. 12. Michalopoulos GK, DeFrances MC. Liver regeneration. Science 1997; 276:60-66. 13. Etienne PL, Baffet G, Desvergne B, Boisnard-Rissel M, Glaise D, GuguenGuillouzo C. Transient expression of c-fos and constant expression of c-myc in freshly isolated and cultured normal adult rat hepatocytes. Oncogene Res 1988;3:255-262. 14. Loyer P, Cariou S, Glaise D, Bilodeau M, Baffet G, Guguen-Guillouzo C. Growth factor dependence of progression through G1 and S phases of adult rat hepatocytes in vitro. Evidence of a mitogen restriction point in mid-late G1. J Biol Chem 1996;271:11484-11492. 15. Talarmin H, Rescan C, Cariou S, Glaise D, Zanninelli G, Bilodeau M, Loyer P, et al. The mitogen-activated protein kinase kinase/extracellular signal-regulated kinase cascade activation is a key signalling pathway involved in the regulation of G(1) phase progression in proliferating hepatocytes. Mol Cell Biol 1999;19:6003-6011. 16. Rescan C, Coutant A, Talarmin H, Theret N, Glaise D, Guguen-Guillouzo C, Baffet G. Mechanism in the sequential control of cell morphology and s phase entry by epidermal growth factor involves distinct mek/erk activations. Mol Biol Cell 2001;12:725-738. 17. Auer KL, Contessa J, Brenz-Verca S, Pirola L, Rusconi S, Cooper G, Abo A, et al. The Ras/Rac1/Cdc42/SEK/JNK/c-Jun cascade is a key pathway by which agonists stimulate DNA synthesis in primary cultures of rat hepatocytes. Mol Biol Cell 1998;9:561-573. 18. Dixon M, Agius L, Yeaman SJ, Day CP. Inhibition of rat hepatocyte proliferation by transforming growth factor beta and glucagon is associated with inhibition of ERK2 and p70 S6 kinase. HEPATOLOGY 1999;29:14181424. 19. Liu Y, Gorospe M, Kokkonen GC, Boluyt MO, Younes A, Mock YD, Wang X, et al. Impairments in both p70 S6 kinase and extracellular signalregulated kinase signaling pathways contribute to the decline in proliferative capacity of aged hepatocytes. Exp Cell Res 1998;240:40-48. 20. Jiang YP, Ballou LM, Lin RZ. Rapamycin-insensitive regulation of 4e-BP1 in regenerating rat liver. J Biol Chem 2001;276:10943-10951. 21. Guguen C, Guillouzo A, Boisnard M, Le Cam A, Bourel M. [Ultrastructural study of monolayer hepatocytes in adult rat cultures in the presence of hydrocortisone hemisuccinate]. Biol Gastroenterol (Paris) 1975;8: 223-231. 22. Floch V, Legros N, Loisel S, Guillaume C, Guilbot J, Benvegnu T, Ferrieres V, et al. New biocompatible cationic amphiphiles derivative from
1088
23.
24.
25. 26. 27.
28. 29.
30.
31.
32.
33.
34.
COUTANT ET AL.
glycine betaine: a novel family of efficient nonviral gene transfer agents. Biochem Biophys Res Commun 1998;251:360-365. Gilot D MM, Benvegnu T, Ferrieres V, Loreal O, Guguen-Guillouzo C, Plusquellec D, Loyer P. Cationic lipids derived from glycine betaine promote efficient and non-toxic gene transfection in cultured hepatocytes. J. Gene Med 2002;4:1-13. Gingras AC, Svitkin Y, Belsham GJ, Pause A, Sonenberg N. Activation of the translational suppressor 4E-BP1 following infection with encephalomyocarditis virus and poliovirus. Proc Natl Acad Sci U S A 1996;93:55785583. Stennicke HR, Salvesen GS. Biochemical characteristics of caspases-3, -6, -7, and -8. J Biol Chem 1997;272:25719-25723. Feldmann G. Liver apoptosis. J Hepatol 1997;26:1-11. Nelsen CJ, Rickheim DG, Timchenko NA, Stanley MW, Albrecht JH. Transient expression of cyclin D1 is sufficient to promote hepatocyte replication and liver growth in vivo. Cancer Res 2001;61:8564-8568. Toker A, Cantley LC. Signalling through the lipid products of phosphoinositide-3-OH kinase. Nature 1997;387:673-676. Vinals F, Chambard JC, Pouyssegur J. p70 S6 kinase-mediated protein synthesis is a critical step for vascular endothelial cell proliferation. J Biol Chem 1999;274:26776-26782. Albrecht JH, Hu MY, Cerra FB. Distinct patterns of cyclin D1 regulation in models of liver regeneration and human liver. Biochem Biophys Res Commun 1995;209:648-655. Khaleghpour K, Pyronnet S, Gingras AC, Sonenberg N. Translational homeostasis: eukaryotic translation initiation factor 4E control of 4Ebinding protein 1 and p70 S6 kinase activities. Mol Cell Biol 1999;19: 4302-4310. Francavilla A, Carr BI, Starzl TE, Azzarone A, Carrieri G, Zeng QH. Effects of rapamycin on cultured hepatocyte proliferation and gene expression. HEPATOLOGY 1992;15:871-877. Westwick JK, Fleckenstein J, Yin M, Yang SQ, Bradham CA, Brenner DA, Diehl AM. Differential regulation of hepatocyte DNA synthesis by cAMP in vitro in vivo. Am J Physiol 1996;271:G780-G790. Jarvis WD, Auer KL, Spector M, Kunos G, Grant S, Hylemon P, Mikkelsen R, et al. Positive and negative regulation of JNK1 by protein
HEPATOLOGY, November 2002
35.
36. 37.
38.
39.
40.
41.
42.
43.
kinase C and p42 (MAP kinase) in adult rat hepatocytes. FEBS Lett 1997; 412:9-14. Spector MS, Auer KL, Jarvis WD, Ishac EJ, Gao B, Kunos G, Dent P. Differential regulation of the mitogen-activated protein and stress-activated protein kinase cascades by adrenergic agonists in quiescent and regenerating adult rat hepatocytes. Mol Cell Biol 1997;17:3556-3565. Albrecht JH, Hansen LK. Cyclin D1 promotes mitogen-independent cell cycle progression in hepatocytes. Cell Growth Differ 1999;10:397-404. Band CJ, Mounier C, Posner BI. Epidermal growth factor and insulininduced deoxyribonucleic acid synthesis in primary rat hepatocytes is phosphatidylinositol 3-kinase dependent and dissociated from protooncogene induction. Endocrinology 1999;140:5626-5634. Iijima Y, Laser M, Shiraishi H, Willey CD, Sundaravadivel B, Xu L, McDermott PJ, et al. c-Raf/MEK/ERK pathway controls protein kinase C-mediated p70s6k activation in adult cardiac muscle cells. J Biol Chem 2002;277:23065-23075. Martin KA, Schalm SS, Romanelli A, Keon KL, Blenis J. Ribosomal S6 kinase 2 inhibition by a potent C-terminal repressor domain is relieved by mitogen-activated protein-extracellular signal-regulated kinase kinase-regulated phosphorylation. J Biol Chem 2001;276:7892-7898. Wang L, Gout I, Proud CG. Cross-talk between the ERK and p70 S6 kinase (S6K) signaling pathways. MEK-dependent activation of S6K2 in cardiomyocytes. J Biol Chem 2001;276:32670-32677. Diggle TA, Subkhankulova T, Lilley KS, Shikotra N, Willis AE, Redpath NT. Phosphorylation of elongation factor-2 kinase on serine 499 by cAMP-dependent protein kinase induces Ca2⫹/calmodulin-independent activity. Biochem J 2001;353:621-626. Fabregat I, Herrera B, Fernandez M, Alvarez AM, Sanchez A, Roncero C, Ventura JJ, et al. Epidermal growth factor impairs the cytochrome C/Caspase-3 apoptotic pathway induced by transforming growth factor beta in rat fetal hepatocytes via a phosphoinositide 3-kinase-dependent pathway. HEPATOLOGY 2000;32:528-535. Roberts RA, James NH, Cosulich SC. The role of protein kinase B and mitogen-activated protein kinase in epidermal growth factor and tumor necrosis factor alpha-mediated rat hepatocyte survival and apoptosis. HEPATOLOGY 2000;31:420-427.
C
ell cycle progression through G1 phase requires the integration of signals from the extracellular environment such as growth factors, cytokines, and extracellular matrix proteins mediated by different transduction cascades. The ability to induce cellular proliferation is often correlated with the ability to promote Abbreviations: MEK, mitogen-activated protein kinase; ERK, extracellular signal–regulated kinase; PI3K, phosphoinositide 3-kinase; mRNA, messenger RNA; EGF, epidermal growth factor. From INSERM U522, Unite´ de Recherches He´patologiques, IFR 97, Hoˆpital Pontchaillou, Rennes, France. Received February 15, 2002; accepted July 24, 2002. Supported by the Institut National de la Sante´ et de la Recherche Me´dicale and the Association pour la Recherche contre le Cancer. C.R. and D.G. are recipients of a fellowship from the Ligue d’Ille et Vilaine contre le Cancer. A.C. is a recipient of a fellowship from the Ministe`re de l’Education nationale, de la Recherche et de la Technologie. A.C. and C.R. contributed equally to this work. Address reprint requests to: Georges Baffet, INSERM U522, Unite´ de Recherches He´patologiques, IFR 97, Hoˆpital Pontchaillou, 35033 Rennes, France. E-mail: [email protected]; fax: (33) 2-99-54-01-37. Copyright © 2002 by the American Association for the Study of Liver Diseases. 0270-9139/02/3605-0010$35.00/0 doi:10.1053/jhep.2002.36160
cell survival. Two signaling cascades have emerged as major players in the mitogenic and antiapoptotic response in many cells: the mitogen-activated protein kinase (MEK)/ extracellular signal–regulated kinase (ERK) and the phosphoinositide 3-kinase (PI3K) pathways. Growth factors as survival factors bind to cell surface receptors and trigger the activation of several kinases, including PI3K, which seems to be implicated in the control of growth and apoptosis of a wide range of cell types.1 This kinase activates PDK1 that in turn leads to the activation of a serine/threonine kinase termed AKT or PKB, which plays a central role in promoting the survival of many cell types. PDK1 also phosphorylates and activates FRAP/mTOR (target of rapamycin). p70s6k, which regulates the ribosomal S6 subunit phosphorylation in response to mitogens, is controlled by FRAP/mTOR and plays an essential role in the translation machinery.2-4 FRAP/mTOR also controls the phosphorylation of 4EBP1, a key protein involved in the regulation of ribosome binding to the messenger RNA (mRNA) via eIF4E.5 Interferences between MEK/ERK and PI3K pathways have 1079
1080
COUTANT ET AL.
often been described; the sensibility of ERK activation to inhibition by PI3K inhibitors depends in many cases on cell types and on the differentiation state of the cells.6,7 Activation of ERK has also proved to be a major regulator of cell proliferation and survival in response to growth factor.8 Once Ras has been activated, GTP-bound ras interacts with the serine/threonine kinase raf and facilitates its activation. It also works the same way with other target enzymes, including PI3K.9 Activated raf phosphorylates and activates the downstream MEK1/2, which in turn phosphorylates ERK1/2. Growth factors play an important role in normal liver growth, in liver after partial hepatectomy, and in proliferative hepatocytes in vitro.10,11 For instance, epidermal growth factor (EGF) inhibits programmed cell death and induces proliferation in hepatocytes.12 To understand the potential importance of components of the cellular homeostasis mechanism, knowledge of G1 phase regulation is of prime importance. Hepatocytes in primary culture seem to be a powerful model to address this question. Following tissue disruption of cell-cell contact during cell isolation, the G0/G1 transition occurs.13 This mimics the entry into G1 of proliferating hepatocytes in vivo in the regenerating liver after partial hepatectomy.10,12 Then, in primary culture, hepatocytes spontaneously progress in G1 up to a restriction point located at the 2/3 of G1 phase.14 We recently showed that, in hepatocytes, the first third of G1 phase is exclusively devoted to growth factor– dependent morphogenic events, whereas its mitogenic signal occurs only around mid-late G1 phase at the restriction point.15,16 A relation between MEK/ERK activation in hepatocytes in vivo and the mitogen-dependent capacity of the cells in vitro has been established. In late G1, MEK/ERK activation is associated with accumulation of cyclin D1 and mitogen-dependent progression of hepatocytes to S phase. Concerning PI3K, few studies have been performed on hepatocytes, and those studies had conflicting results.17-20 In this report, we analyzed the possible role of the PI3K pathway in the regulation of EGF-induced DNA replication and survival in mid-late G1 phase. We examined the relation that could exist between this pathway and the expression of cyclin D1, a critical player in G1/S transition, and questioned the interference that could exist between the PI3K and MEK/ERK pathways.
Materials and Methods Cell Cultures. Hepatocytes were isolated from Sprague-Dawley male (150-200 g) rat livers by the 2-step perfusion procedure using 0.025% Liberase (BoehringerIngelheim, Indianapolis, IN) buffered with 0.1 mol/L
HEPATOLOGY, November 2002
HEPES (pH 7.4) as previously described.21 At indicated times, wortmannin, LY294002, rapamycin, cycloheximide, or U0126 dissolved in dimethyl sulfoxide was added at defined concentrations. All control features in dimethyl sulfoxide at a final concentration of 0.2% or 0.37% were changed at the same time as that of treated cells. Transfection Experiments. Hepatocytes were transfected using GB12 as previously described.22,23 Four micrograms of plasmid empty vector or pcDNA-HA-Akt (K179M) was mixed with cationic lipid (10 g/mL) in 800 L of OptiMEM at room temperature for 30 minutes. This medium was added drop-wise onto 24-hourold hepatocytes. Twelve hours later, hepatocytes were stimulated by EGF. [3H]Thymidine Incorporation and [14C]Leucine Incorporation. The rates of DNA and protein synthesis were measured in primary cultures by adding 2 Ci of [methyl-3H]thymidine or 1 Ci/mL of L-[U-14C]leucine for given periods of time before cell harvesting and precipitation with ice-cold trichloroacetic acid (5%). Chemicals. [␣-32P]deoxycytidine triphosphate (3,000 Ci/mmol), L-[U-14C]leucine (31.5 mCi/mmol), and [methyl-3H]thymidine (5 Ci/mmol) were from Amersham Corp. (Buckinghamshire, England); recombinant human EGF and MEK inhibitor U0126 were from Promega (Madison, WI); wortmannin was from Calbiochem (San Diego, CA); and insulin I-5500, LY294002, rapamycin, and cycloheximide were from Sigma (St. Louis, MO). Immunoblotting. After sodium dodecyl sulfate/polyacrylamide gel electrophoresis, proteins were transferred and detected according to the SuperSignal Ultra Chemiluminescent Substrate procedure (Pierce, Rockford, IL). Anti–phospho-GSK 3 ␣/, anti–phospho-ERK, and anti–phospho-AKT antibodies as well as a mixing of anti-phospho Thr 389 and anti-phospho Thr 421/424 antibodies for Phospho-p70S6Kinase detection were from New England Biolabs (Beverly, MA). Anti-AKT (sc-1618) and anti-ERK1/2, a mixing of ERK1 (sc-94) and ERK2 antibodies (sc-154), were from Santa Cruz Biotechnology (Santa Cruz, CA). Polyclonal antibody against cyclin D1 was obtained from Neomarkers (Union City, CA). Rabbit polyclonal anti– 4E-BP124 was a gift from Drs. Pyronnet and Sonenberg. Northern Blotting. Cells were lysed at different times of culture, RNA was extracted with the Qiagen RNeasy kit (Qiagen, Valencia, CA), and Northern blotting was performed as described previously.15 Caspase Activity Assay. Caspase-mediated cleavage of DEVD-AMC was measured by spectrofluorometry (Molecular Devices, Wokingham, England) at 380- and 440-nm wavelength (excitation and emission, respec-
HEPATOLOGY, Vol. 36, No. 5, 2002
COUTANT ET AL.
1081
tively).25 For each measure, the background of fluorescence was subtracted. The caspase activity was presented in arbitrary units of fluorescence (per 80 g of total proteins). Assessment of Mitochondrial ⌬m. 3,3⬘-Dihexyloxacarbocyanine iodide (DiOC6) was purchased from Molecular Probes, Inc. (Eugene, OR). Changes in the inner mitochondrial transmembrane potential ⌬m were determined by incubating 6 ⫻ 105 hepatocytes in 40 nmol/L DiOC626 in culture medium for 30 minutes at 37°C. Cells were analyzed using a FACScan flow cytometer (Becton Dickinson, Mountain View, CA). All the experiments described in this report were performed at least 3 times.
Results PI3K Inhibition Blocks DNA Synthesis in Hepatocytes. PI3K and MEK/ERK pathways have been involved in the regulation of cell cycle progression in many cell types. We already showed that the MEK/ERK cascade is a key signaling pathway involved in the regulation of G1 phase progression in proliferating hepatocytes.15,16 To improve our understanding of the role of PI3K cascade activation onto hepatocyte cell cycle progression, PI3K inhibition experiments were performed. First, hepatocyte primary cultures were maintained for 48 hours in basal conditions (48-hour-old hepatocytes) and then stimulated with EGF in the presence or absence of a specific PI3K inhibitor, LY294002 (Fig. 1A). This inhibitor at a concentration of 10 mol/L totally inhibited DNA replication all along the time course analyzed, showing that PI3K inhibition not only delayed but totally abolished DNA replication in hepatocytes. In parallel, we confirmed that the MEK cascade is an important player in DNA replication by showing that the MEK inhibitor U0126 as well as PD9805915,16 completely blocked DNA synthesis. We performed dose-response experiments in the presence of PI3K inhibitor (Fig. 1B). A dose response was obtained; 5 mol/L LY294002 inhibited DNA replication by 50%, whereas 10 mol/L decreased thymidine incorporation by 100%. To provide evidence that cells remained blocked at the restriction point or progressed in late G1 phase in the presence of U0126 or LY294002, reversion experiments were performed. Hepatocytes (48 hours old) were stimulated by EGF and treated for 24 hours with the 2 inhibitors. U0126 and LY294002 were then removed and replication analyzed in the presence of growth factor alone (Fig. 2A). As already detailed,15 hepatocytes from the reversion experiments after 24-hour MEK/ERK inhibition started to replicate DNA with an 18- to 24-hour delay,
Fig. 1. Regulation of DNA replication by the PI3K pathway. (A) Time course of [3H-methyl]thymidine incorporation into DNA; 48-hour-old hepatocytes were maintained in basal medium (basal) or stimulated with EGF (EGF) in the absence or presence of LY294002 (15 mol/L) or U0126 (50 mol/L) and analyzed at the indicated time after stimulation. (B) Dose-dependent inhibition of DNA replication by LY294002; 48-hour-old hepatocytes were stimulated with EGF in the presence of increasing concentrations of LY294002. DNA replication was measured at the indicated times after stimulation.
corresponding to the time needed to reach the G1/S transition, confirming that the cells were blocked at the restriction point in the presence of MEK inhibitor. Concerning the PI3K cascade, our data showed that, unlike the MEK/ERK inhibition, the cells were able to undergo DNA synthesis with only a 6- to 12-hour delay on removal of the LY inhibitor. In hepatocytes, as in many cells, the up-regulation of cyclin D1 is an important player in late G1 progression and drives the cells to mitogenic response.27 To confirm the effect of LY294002 and U0126 inhibitions on late cell cycle progression, we analyzed the induction of cyclin D1 on inhibitor removal after 24-hour inhibition. As shown in Fig. 2B, cyclin D1 protein could be detected 3 hours after removal of LY294002, whereas its expression was
1082
COUTANT ET AL.
HEPATOLOGY, November 2002
Fig. 2. DNA replication, cyclin D1 expression, and GSK3 phosphorylation after LY294002 or U0126 treatment and drug removal. (A, B, and D) Reversion experiments; 48-hourold hepatocytes were cultured in basal conditions or stimulated with EGF in the presence of LY294002 (15 mol/L) or U0126 (50 mol/ L). Then, 24 hours after treatment, inhibitors were removed and hepatocytes cultured in the presence of EGF alone. (A) Methyl thymidine incorporation into DNA, (B) cyclin D1 protein expression, and (D) GSK3 phosphorylation were analyzed at the indicated times after drug removal. Error bars indicate SD; *P ⬍ .001 (vs. reversion U0126) by Student’s t test. (C) Control of phospho-GSK3 inhibition; 48-hour-old hepatocytes were stimulated with EGF in the presence of LY294002 or U0126 and analyzed at the indicated times after stimulation.
only detected after 12 hours in the experiment with U0126. All these results strongly argue for a control of late G1 progression by a mechanism not involving PI3K but MEK activation. As control, we confirmed that the 2 pathways were activated by the same kinetic after drug removal. We chose to look at the phospho-GSK3, which was under both PI3K and MEK activating processes. We first showed that the 2 drugs partially inhibited the GSK-3 phosphorylation induced by EGF (Fig. 2C). Second, in reversion experiments, phosphorylation of GSK-3 was induced simultaneously after removal of LY294002 and U0126 (Fig. 2D), indicating that recovery of intracellular pathways was not delayed in U0126- compared with LY294002-treated hepatocytes. FRAP/mTOR Inhibition Abolishes DNA and Protein Synthesis. Two downstream signaling pathways have been described for PI3K (AKT and FRAP/mTOR) that can be distinguished by their sensibility to the drug rapamycin.28 As shown in Fig. 3A, rapamycin, which acts at the FRAP/mTOR level, was able to decrease [3H]thymidine incorporation in a dose-response– dependent manner at any of the times analyzed. A total of 50% and 80% of DNA synthesis was inhibited in the presence of 0.1 and 0.5 nmol/L of rapamycin, respectively, whereas 1 nmol/L totally abolished DNA replication. As LY294002, rapamycin did not only delay but totally inhibited DNA replication over a longer period of time, up to 72 hours (result not shown). Next, we investigated whether the effect of rapamycin on thymidine incorporation could correlate with inhibi-
tion of protein synthesis (Fig. 3B and C). As in vascular endothelial cells,29 rapamycin inhibited thymidine incorporation with the same dose dependence as protein synthesis. In both cases, the median inhibitory concentration for rapamycin was 0.05 nmol/L. Even at high concentrations, rapamycin was not able to decrease thymidine and leucine incorporations below the basal level of unstimulated control hepatocytes (Fig. 3B and data not shown). The correlation between the rate of DNA and protein synthesis was confirmed with the use of cycloheximide, a potent inhibitor of protein synthesis (Fig. 3C). The median inhibitory concentration for cycloheximide was identical for the inhibition of DNA replication and protein synthesis. Cyclin D1 Is Under a FRAP/mTOR Regulation Mechanism. Cyclin D1 has been shown to be regulated at 2 different levels in many cell types, including hepatocytes30; FRAP/mTOR, which plays a role in the regulation of the translation machinery, could be involved in this process. We therefore investigated the effects of LY294002 and rapamycin at the cyclin D1 mRNA and protein level. As expected, cyclin D1 mRNA and protein rapidly accumulated after EGF stimulation (Fig. 4A and B). In contrast, in cells exposed to medium with EGF plus LY294002 or rapamycin, cyclin D1 mRNA expression was diminished by approximately 90% and 80%, respectively. At the protein level, the 2 concentrations of rapamycin used (1 and 0.1 nmol/L) and LY294002 at 15 mol/L also inhibited cyclin D1 protein expression. Inhibition was less efficient by using LY294002 at 1
HEPATOLOGY, Vol. 36, No. 5, 2002
COUTANT ET AL.
1083
dicated that cyclin D1 protein was inhibited by 82% ⫾ 15%, 78% ⫾ 16%, and 75% ⫾ 20% in the presence of LY294002, rapamycin, and U0126, respectively. AKT Phosphorylation Is PI3K Dependent but Not Directly Related to the Replication Process. AKT that lies downstream of PI3K is a general mediator of cell survival and could be involved in proliferation of many cell types. Interferences between PI3K and MEK/ERK pathways have been described in many cell types. In this context, we first investigated the effect of PI3K and FRAP/mTOR inhibitors on AKT and ERK1/2 activations (Fig. 5). AKT was rapidly and transiently phosphorylated after EGF stimulation of 48-hour-old hepatocytes, reaching a maximum at 0.5 hours and decreasing thereafter at 1 hour to near zero 4 hours after stimulation (Fig. 5A). Second, in the presence of LY294002, this phosphorylation was totally abolished, confirming that AKT activation is downstream of PI3K in EGF-stimulated hepatocytes. Third, the FRAP/mTOR inhibitor, at a concentration that totally inhibited DNA replication, was unable to decrease AKT phosphorylation induced by the growth fac-
Fig. 3. Inhibition of DNA synthesis by rapamycin correlates with the inhibition of protein synthesis. (A) Hepatocytes (48 hours old) were stimulated with EGF in the presence or absence of increasing concentrations of rapamycin. DNA replication was measured at the indicated times after stimulation. (B and C) Hepatocytes (48 hours old) were stimulated with EGF in the presence of increasing concentrations of (B) rapamycin or (C) cycloheximide. Thymidine (■) and leucine (‚) incorporation was measured as described in Materials and Methods and is expressed as a percentage of the maximal incorporation obtained in the presence of EGF alone.
mol/L, in accordance with low inhibition of DNA replication (Fig. 1B). As control, the visualization of nonphosphorylated ERK1/2 protein showed similar protein concentrations loaded on polyacrylamide gel electrophoresis. Quantification from 4 separate experiments in-
Fig. 4. LY294002 and rapamycin block cyclin D1 mRNA and protein expression induced by EGF. (A) Cyclin D1 mRNA and (B) cyclin D1 protein expressions; 48-hour-old hepatocytes were stimulated or not with EGF in the presence or absence of LY294002 (15 mol/L) or rapamycin (1 nmol/L) and analyzed at the indicated times after stimulation.
1084
COUTANT ET AL.
Fig. 5. AKT protein phosphorylation in EGF-stimulated hepatocytes. (A) AKT and ERK1/2 phosphorylations were analyzed by Western blot; 48-hour-old hepatocytes were stimulated with EGF in the presence or absence of LY294002 (15 mol/L), rapamycin (1 nmol/L), or U0126 (50 mol/L) and analyzed at the indicated time after stimulation. (B and C) Hepatocytes (24 hours old) were transfected by dominant-negative AKT (DN) or empty vector (EV) and stimulated or not with EGF at 36 hours and analyzed at the indicated time after EGF stimulation for (B) DNA replication and (C) GSK3 phosphorylation and total AKT protein expression as control. Experiments were performed 4 times.
tor. We confirm that the phosphorylation of ERK1/2 was rapid and sustained on EGF stimulation. Interestingly, the 2 inhibitors of the PI3K pathway (LY294002 and rapamycin) had no effect on these phosphorylations.
HEPATOLOGY, November 2002
We then specified the effect of AKT inhibition on DNA replication. AKT dominant-negative vector was transfected 24 hours after seeding, and hepatocytes were stimulated by EGF 12 hours later (Fig. 5B). From the kinetics of 4 independent experiments, dominant-negative AKT, expressed at high levels in the cells (Fig. 5C), was not able to decrease DNA replication induced by EGF compared with control cells (EGF ⫹ empty vector). By this procedure, 35% to 40% of the hepatocytes appeared strongly positive for the GFP signal, confirming a high transfection efficiency.23 Furthermore, we showed that phosphorylation of GSK-3 that lies downstream of AKT was partially inhibited by the dominant-negative AKT compared with the empty vector. EGF-Independent 4E-BP1 Phosphorylation Is Under an Insulin-Activation Mechanism. The translational proteins p70S6K and 4E-BP1 are principally regulated by the PI3K-FRAP/mTOR pathway, and this regulation plays a key role in the translation machinery activation.5 Recently, studies indicated that the MEK/ ERK pathway could also play a major role in p70S6K activation. Therefore, impacts of LY294002, rapamycin, and U0126 on p70S6K and 4E-BP1 phosphorylations are important keys to understand the mechanism involved in hepatocyte proliferation. This study evidenced first that p70S6K is up-phosphorylated after EGF stimulation. We showed that phosphop70 basal level is under a PI3K/mTOR mechanism, whereas U0126 blocked the p70S6K phosphorylation induced by the growth factor (Fig. 6A). This result argues that p70S6K phosphorylation induced by EGF is MEK dependent, indicating that interferences between the MEK/ERK pathway and p70S6K could occur in hepatocytes. As control, we confirmed that LY294002 and rapamycin had no effect on ERK1/2 phosphorylation, whereas these phosphorylations were completely abrogated by U0126. Second, we looked at the role of PI3K pathway activation/inhibition on 4E-BP1 phosphorylation (Fig. 6B, C, and D). Surprisingly, in basal conditions, in the absence of EGF, 4E-BP1 appeared as a protein of low electrophoretic mobilities (/␥) corresponding to phosphorylated forms.24,31 Moreover, EGF stimulation did not influence 4E-BP1 mobility status. However, the 2 inhibitors, LY294002 and rapamycin, dephosphorylated the protein as visualized by a faster electrophoretic mobility (␣), showing that 4E-BP1 phosphorylation is under a PI3K-FRAP/mTOR mechanism that is independent of the presence of EGF (Fig. 6B and C). The MEK/ERK pathway did not interfere with this mechanism because the MEK inhibitor U0126 had no effect on 4E-BP1 electrophoretic mobility.
HEPATOLOGY, Vol. 36, No. 5, 2002
COUTANT ET AL.
1085
tion evidenced by the decrease of the mitochondrial transmembrane potential (⌬m), measured by flow cytometry, using the potential-sensitive dye DiOC6. In basal experiments, DEVD-AMC cleavage increased between 72 and 96 hours of culture and the growth factor protected hepatocytes from apoptosis during this period of stimulation (Fig. 7A). This protection by EGF was PI3K and FRAP/mTOR independent because inhibition by LY294002 as well as rapamycin did not induce DEVD-AMC cleavage. On the opposite, MEK/ERK inhibition by U0126 induced apoptosis as in basal conditions. Last, we confirmed that the MEK/ERK pathway is a key cascade in hepatocyte survival by showing that the mitochondrial potential is altered in U0126-treated cells (Fig. 7B). After 24 hours of exposure, the number of cells exhibiting a high ⌬m decreased, whereas that of cells with low ⌬m, corresponding to dead cells, increased by 21% ⫾ 3% in U0126-treated hepatocytes, showing that mitochondrial permeability transition occurred in this condition. In contrast, in LY294002- and rapamycintreated hepatocytes, the number of high-⌬m cells remained elevated, as in the presence of EGF alone.
Discussion
Fig. 6. Regulation of p70S6K and 4E-BP1 phosphorylation state. (A-C) Hepatocytes (48 hours old) in basal medium containing insulin were stimulated or not by EGF in the presence or absence of LY294002 (15 mol/L), rapamycin (1 nmol/L), or U0126 (50 mol/L) and analyzed at the indicated time after stimulation. (D) Freshly isolated hepatocytes were cultured in the absence or presence of insulin or EGF and analyzed 1 hour after stimulation.
Because insulin, used as a comitogen in the basal culture medium for hepatocytes, could contribute to the 4EBP1 activation, experiments were performed in the absence of the hormone (Fig. 6D). Under this condition, the ␣ form disappeared and only higher apparent molecular forms were detected in the presence of insulin (, ␥). In all cases, contrary to p70S6K, EGF did not influence the phosphorylation rate of the protein. Survival Is PI3K Independent but MEK Dependent in Mid-Late G1 Phase. Cell death was investigated by 2 distinct approaches: (1) the executioner caspases 3/7 activity measured through the proteolytic DEVD-AMC degradation, and (2) mitochondrial permeability transi-
In hepatocytes, growth factors have been shown to activate different signaling cascades, including MEK/ERK, PI3K, c-Jun-N-terminal kinase, and P38 pathways.32-35 This complex network participates in liver homeostasis by the regulation of hepatocyte proliferation and survival. Our study underlines the role of the 2 MEK/ERK and PI3K pathways in the regulation of hepatocyte proliferation and apoptosis. Hepatocytes in primary culture progress in G1 independently of growth factor stimulation up to the restriction point located in mid-late G1 phase, where they remain arrested in the absence of mitogenic signal. After mitogenic stimulation, hepatocytes progress in late G1 phase and undergo DNA synthesis. We and others previously showed that after growth factor stimulation for progression up to S phase, there was marked induction of cyclin D1 in late G1 phase.14,15,27,36 From previous data, the question arose as to whether PI3K cascade regulates DNA synthesis and expression of cyclin D1 or not. Although dominant-negative PI3K did not inhibit DNA synthesis in hepatocytes, other data suggested that the cascade inhibition could abolish or only delay entry to S phase.17,19,37 Our study clearly shows that the PI3K inhibitor, LY294002, and the FRAP/mTOR inhibitor, rapamycin, do not delay entry to S phase and expression of cyclin D1 but totally abolish DNA replication and the cyclin induction in growth factor–stimulated hepatocytes. Uncoupling of mRNA and protein levels has been
1086
COUTANT ET AL.
HEPATOLOGY, November 2002
Fig. 7. EGF protects hepatocytes from apoptosis via the MEK/ERK pathway. (A) Hepatocytes (48 hours old) were maintained in basal conditions (basal) or stimulated by EGF in the absence or presence of U0126 (50 mol/L), LY294002 (15 mol/L), or rapamycin (1 nmol/L) and analyzed for (A) DEVDAMC caspase activities at the indicated times and (B) mitochondrial transmembrane potential (⌬m) using the DiOC6 potential-sensitive dye 24 hours after stimulation. Experiments were performed on hepatocytes attached to the plate.
pointed out in hepatocytes, suggesting that cyclin D1 is subject to control on a translation level and/or protein stability in rat liver.30 In this context, we showed that LY294002 and rapamycin completely prevented cyclin D1 at the mRNA and protein levels. These results show that PI3K and FRAP/mTOR as well as the MEK/ERK cascade15 are linked to expression of cyclin D1, highlighting the pivotal role of this protein. We previously showed that hepatocytes treated with the MEK inhibitor, PD98059, behaved as if they remained blocked at the restriction point.15 We confirmed these results with U0126; in reversion experiments, hepatocytes were able to undergo DNA synthesis with a corresponding 24-hour delay like cells blocked at the restriction point. Whereas LY294002 completely blocked EGF-stimulated DNA synthesis, evidence was provided that hepatocytes progressed in late G1 phase in the presence of this inhibitor. Indeed, in reversion experiments, the time interval between removal of LY294002 and DNA synthesis was only 6 to 12 hours, suggesting that cells were not blocked at the restriction point. Moreover, cyclin D1 protein could be rapidly detected 3 hours after removal of LY294002, contrasting with cells treated with U0126. All these data indicate that hepatocytes could progress in late G1 phase independently of PI3K activation, whereas its activation controls the replication process. Inhibition of FRAP/mTOR was sufficient to inhibit DNA synthesis and expression of cyclin D1 without af-
fecting AKT phosphorylation. Moreover, AKT dominant-negative mutant did not interfere with DNA synthesis induced by EGF. Other data also suggest that inhibition of hepatocyte proliferation by transforming growth factor  did not involve AKT but might be partly mediated by the inhibition of ERK2 and p70S6K.18 Whereas previous studies have shown that Raf/MEK signaling is neither necessary nor sufficient for p70S6K activation by mitogen, recent data show that the expression of either c-Raf or MEK is sufficient for S6K1/2 activation in an ERK-dependent manner.38-40 Our results argue that p70S6K phosphorylation induced by EGF is under a U0126-sensitive regulation. On the opposite hand, in this report, we highlight that PI3K cascade activation at the 4E-BP1 level is independent of EGF stimulation but controlled by an insulin-regulation mechanism. From the literature, 4E-BP1 activation seems to be complex and the effect of insulin could involve multiple phosphorylation events, which have been described as rapamycin sensitive and insensitive.20,31,41 We show that a rapamycin-sensitive pathway is involved in 4E-BP1 phosphorylation in response to insulin. This regulation could account for the permissive role of insulin on hepatocyte proliferation. This finding, which highlights the key role of FRAP/ mTOR in hepatocyte replication, was also established for vascular endothelial cells and lymphocytes.29 However, Jiang et al.20 have recently reported that the effect of rapamycin on 4E-BP1 function in vivo can be significantly different from its effect in cultured cells. This discrepancy
HEPATOLOGY, Vol. 36, No. 5, 2002
could be in part due to the environmental conditions and via the release of humoral factors from other tissues in vivo, as reported by these investigators. Moreover, 4EBP1 underwent partial dephosphorylation after 4 days of treatment with rapamycin in vivo. We established that the antiapoptotic function of EGF in primary cultures of rat hepatocytes is dependent on the MEK/ERK pathway in mid-late G1 phase, whereas the inhibition of the PI3K cascade has no effect on hepatocyte survival. Moreover, our data highlight that hepatocyte progression in late G1 generates survival signals that counteract cell death induced in hepatocytes blocked at the restriction point in basal conditions or in the presence of MEK inhibitor. Cells blocked at the restriction point did not survive and rapidly progressed in apoptosis. Interestingly, the location of cells in G1, the state of differentiation, and the presence of specific apoptotic inducers in the medium could be of prime importance to transduce specific signals. In 20-day-old fetal hepatocytes, PI3K mediated the survival effect of EGF on transforming growth factor –induced death, whereas the MEK inhibitors did not block the survival effect of growth factor.42 In adult hepatocytes, EGF inhibition of transforming growth factor –induced apoptosis is dependent on the PI3K and ERK activities, whereas suppression of apoptosis induced by tumor necrosis factor ␣ does not require PI3K but is dependent on ERK and P38 mitogen-activated protein kinase activation.43 The survival and/or mitogenic orderly events induced by the growth factor seem to be dependent on a cellular clock in adult hepatocytes. In conclusion, the activation of different pathways might contribute to the complex regulation of hepatocyte DNA replication and survival. The data presented here emphasize the importance of PI3K-FRAP/mTOR in DNA replication, whereas MEK activation is fundamental for replication and survival. The 2 cascade activations are not redundant and are necessary for a full response to growth factor. Our results help to understand the intracellular signaling activation in G1 phase, which is of prime importance for the knowledge of hepatocyte growth and survival controls that regulate homeostasis during the regeneration process following liver injury and diseases. Acknowledgment: The authors thank Drs. S. Pyronnet and N. Sonenberg for giving us the anti– 4E-BP1 antibody and J. C. Andrieux for critical reading of the manuscript.
References 1. Datta SR, Brunet A, Greenberg ME. Cellular survival: a play in three Akts. Genes Dev 1999;13:2905-2927.
COUTANT ET AL.
1087
2. Chou MM, Blenis J. The 70 kDa S6 kinase: regulation of a kinase with multiple roles in mitogenic signalling. Curr Opin Cell Biol 1995;7:806814. 3. Chung J, Grammer TC, Lemon KP, Kazlauskas A, Blenis J. PDGF- and insulin-dependent pp70S6k activation mediated by phosphatidylinositol3-OH kinase. Nature 1994;370:71-75. 4. Conus NM, Hemmings BA, Pearson RB. Differential regulation by calcium reveals distinct signaling requirements for the activation of Akt and p70S6k. J Biol Chem 1998;273:4776-4782. 5. Gingras AC, Raught B, Sonenberg N. Regulation of translation initiation by FRAP/mTOR. Genes Dev 2001;15:807-826. 6. Berra E, Diaz-Meco MT, Moscat J. The activation of p38 and apoptosis by the inhibition of Erk is antagonized by the phosphoinositide 3-kinase/Akt pathway [published erratum appears in J Biol Chem 1998 Jun 26;273(26): 16630]. J Biol Chem 1998;273:10792-10797. 7. Rommel C, Clarke BA, Zimmermann S, Nunez L, Rossman R, Reid K, Moelling K, et al. Differentiation stage-specific inhibition of the RafMEK-ERK pathway by Akt. Science 1999;286:1738-1741. 8. Brunet A, Pouyssegur J. Mammalian MAP kinase modules: how to transduce specific signals. Essays Biochem 1997;32:1-16. 9. Kauffmann-Zeh A, Rodriguez-Viciana P, Ulrich E, Gilbert C, Coffer P, Downward J, Evan G. Suppression of c-Myc-induced apoptosis by Ras signalling through PI(3)K and PKB. Nature 1997;385:544-548. 10. Fausto N, Laird AD, Webber EM. Liver regeneration. 2. Role of growth factors and cytokines in hepatic regeneration. FASEB J 1995;9:15271536. 11. Fausto N. Liver regeneration. J Hepatol 2000;32:19-31. 12. Michalopoulos GK, DeFrances MC. Liver regeneration. Science 1997; 276:60-66. 13. Etienne PL, Baffet G, Desvergne B, Boisnard-Rissel M, Glaise D, GuguenGuillouzo C. Transient expression of c-fos and constant expression of c-myc in freshly isolated and cultured normal adult rat hepatocytes. Oncogene Res 1988;3:255-262. 14. Loyer P, Cariou S, Glaise D, Bilodeau M, Baffet G, Guguen-Guillouzo C. Growth factor dependence of progression through G1 and S phases of adult rat hepatocytes in vitro. Evidence of a mitogen restriction point in mid-late G1. J Biol Chem 1996;271:11484-11492. 15. Talarmin H, Rescan C, Cariou S, Glaise D, Zanninelli G, Bilodeau M, Loyer P, et al. The mitogen-activated protein kinase kinase/extracellular signal-regulated kinase cascade activation is a key signalling pathway involved in the regulation of G(1) phase progression in proliferating hepatocytes. Mol Cell Biol 1999;19:6003-6011. 16. Rescan C, Coutant A, Talarmin H, Theret N, Glaise D, Guguen-Guillouzo C, Baffet G. Mechanism in the sequential control of cell morphology and s phase entry by epidermal growth factor involves distinct mek/erk activations. Mol Biol Cell 2001;12:725-738. 17. Auer KL, Contessa J, Brenz-Verca S, Pirola L, Rusconi S, Cooper G, Abo A, et al. The Ras/Rac1/Cdc42/SEK/JNK/c-Jun cascade is a key pathway by which agonists stimulate DNA synthesis in primary cultures of rat hepatocytes. Mol Biol Cell 1998;9:561-573. 18. Dixon M, Agius L, Yeaman SJ, Day CP. Inhibition of rat hepatocyte proliferation by transforming growth factor beta and glucagon is associated with inhibition of ERK2 and p70 S6 kinase. HEPATOLOGY 1999;29:14181424. 19. Liu Y, Gorospe M, Kokkonen GC, Boluyt MO, Younes A, Mock YD, Wang X, et al. Impairments in both p70 S6 kinase and extracellular signalregulated kinase signaling pathways contribute to the decline in proliferative capacity of aged hepatocytes. Exp Cell Res 1998;240:40-48. 20. Jiang YP, Ballou LM, Lin RZ. Rapamycin-insensitive regulation of 4e-BP1 in regenerating rat liver. J Biol Chem 2001;276:10943-10951. 21. Guguen C, Guillouzo A, Boisnard M, Le Cam A, Bourel M. [Ultrastructural study of monolayer hepatocytes in adult rat cultures in the presence of hydrocortisone hemisuccinate]. Biol Gastroenterol (Paris) 1975;8: 223-231. 22. Floch V, Legros N, Loisel S, Guillaume C, Guilbot J, Benvegnu T, Ferrieres V, et al. New biocompatible cationic amphiphiles derivative from
1088
23.
24.
25. 26. 27.
28. 29.
30.
31.
32.
33.
34.
COUTANT ET AL.
glycine betaine: a novel family of efficient nonviral gene transfer agents. Biochem Biophys Res Commun 1998;251:360-365. Gilot D MM, Benvegnu T, Ferrieres V, Loreal O, Guguen-Guillouzo C, Plusquellec D, Loyer P. Cationic lipids derived from glycine betaine promote efficient and non-toxic gene transfection in cultured hepatocytes. J. Gene Med 2002;4:1-13. Gingras AC, Svitkin Y, Belsham GJ, Pause A, Sonenberg N. Activation of the translational suppressor 4E-BP1 following infection with encephalomyocarditis virus and poliovirus. Proc Natl Acad Sci U S A 1996;93:55785583. Stennicke HR, Salvesen GS. Biochemical characteristics of caspases-3, -6, -7, and -8. J Biol Chem 1997;272:25719-25723. Feldmann G. Liver apoptosis. J Hepatol 1997;26:1-11. Nelsen CJ, Rickheim DG, Timchenko NA, Stanley MW, Albrecht JH. Transient expression of cyclin D1 is sufficient to promote hepatocyte replication and liver growth in vivo. Cancer Res 2001;61:8564-8568. Toker A, Cantley LC. Signalling through the lipid products of phosphoinositide-3-OH kinase. Nature 1997;387:673-676. Vinals F, Chambard JC, Pouyssegur J. p70 S6 kinase-mediated protein synthesis is a critical step for vascular endothelial cell proliferation. J Biol Chem 1999;274:26776-26782. Albrecht JH, Hu MY, Cerra FB. Distinct patterns of cyclin D1 regulation in models of liver regeneration and human liver. Biochem Biophys Res Commun 1995;209:648-655. Khaleghpour K, Pyronnet S, Gingras AC, Sonenberg N. Translational homeostasis: eukaryotic translation initiation factor 4E control of 4Ebinding protein 1 and p70 S6 kinase activities. Mol Cell Biol 1999;19: 4302-4310. Francavilla A, Carr BI, Starzl TE, Azzarone A, Carrieri G, Zeng QH. Effects of rapamycin on cultured hepatocyte proliferation and gene expression. HEPATOLOGY 1992;15:871-877. Westwick JK, Fleckenstein J, Yin M, Yang SQ, Bradham CA, Brenner DA, Diehl AM. Differential regulation of hepatocyte DNA synthesis by cAMP in vitro in vivo. Am J Physiol 1996;271:G780-G790. Jarvis WD, Auer KL, Spector M, Kunos G, Grant S, Hylemon P, Mikkelsen R, et al. Positive and negative regulation of JNK1 by protein
HEPATOLOGY, November 2002
35.
36. 37.
38.
39.
40.
41.
42.
43.
kinase C and p42 (MAP kinase) in adult rat hepatocytes. FEBS Lett 1997; 412:9-14. Spector MS, Auer KL, Jarvis WD, Ishac EJ, Gao B, Kunos G, Dent P. Differential regulation of the mitogen-activated protein and stress-activated protein kinase cascades by adrenergic agonists in quiescent and regenerating adult rat hepatocytes. Mol Cell Biol 1997;17:3556-3565. Albrecht JH, Hansen LK. Cyclin D1 promotes mitogen-independent cell cycle progression in hepatocytes. Cell Growth Differ 1999;10:397-404. Band CJ, Mounier C, Posner BI. Epidermal growth factor and insulininduced deoxyribonucleic acid synthesis in primary rat hepatocytes is phosphatidylinositol 3-kinase dependent and dissociated from protooncogene induction. Endocrinology 1999;140:5626-5634. Iijima Y, Laser M, Shiraishi H, Willey CD, Sundaravadivel B, Xu L, McDermott PJ, et al. c-Raf/MEK/ERK pathway controls protein kinase C-mediated p70s6k activation in adult cardiac muscle cells. J Biol Chem 2002;277:23065-23075. Martin KA, Schalm SS, Romanelli A, Keon KL, Blenis J. Ribosomal S6 kinase 2 inhibition by a potent C-terminal repressor domain is relieved by mitogen-activated protein-extracellular signal-regulated kinase kinase-regulated phosphorylation. J Biol Chem 2001;276:7892-7898. Wang L, Gout I, Proud CG. Cross-talk between the ERK and p70 S6 kinase (S6K) signaling pathways. MEK-dependent activation of S6K2 in cardiomyocytes. J Biol Chem 2001;276:32670-32677. Diggle TA, Subkhankulova T, Lilley KS, Shikotra N, Willis AE, Redpath NT. Phosphorylation of elongation factor-2 kinase on serine 499 by cAMP-dependent protein kinase induces Ca2⫹/calmodulin-independent activity. Biochem J 2001;353:621-626. Fabregat I, Herrera B, Fernandez M, Alvarez AM, Sanchez A, Roncero C, Ventura JJ, et al. Epidermal growth factor impairs the cytochrome C/Caspase-3 apoptotic pathway induced by transforming growth factor beta in rat fetal hepatocytes via a phosphoinositide 3-kinase-dependent pathway. HEPATOLOGY 2000;32:528-535. Roberts RA, James NH, Cosulich SC. The role of protein kinase B and mitogen-activated protein kinase in epidermal growth factor and tumor necrosis factor alpha-mediated rat hepatocyte survival and apoptosis. HEPATOLOGY 2000;31:420-427.